BULK IEEE MATLAB PROJECTS 2015-16

Sno.

Topic

Abstract

Year

1.

MATLAB2015_01

Machine Learning-Based Coding Unit Depth
Decisions for Flexible Complexity Allocation
in High Efficiency Video Coding

In this paper, we propose a machine learning-based
fast coding unit (CU) depth decision method for High Efficiency Video Coding (HEVC), which optimizes the complexity allocation at CU level with given rate-distortion (RD) cost constraints. First, we analyze quad-tree CU depth decision process in HEVC
and model it as a three-level of hierarchical binary decision problem. Second, a flexible CU depth decision structure is presented, which allows the performances of each CU depth decision be smoothly transferred between the coding complexity and RD performance. Then, a three-output joint classifier consists of multiple binary classifiers with different parameters is designed to control the risk of false prediction. Finally, a sophisticated RD-complexity model is derived to determine the optimal
parameters for the joint classifier, which is capable of minimizing the complexity in each CU depth at given RD degradation constraints. Comparative experiments over various sequences show that the proposed CU depth decision algorithm can reduce the computational complexity from 28.82% to 70.93%, and 51.45% on average when compared with the original HEVC test model.

2015

2.

MATLAB2015_02

Distinguishing Local and Global Edits for Their
Simultaneous Propagation in a Uniform Framework

In propagating edits for image editing, some edits are intended to affect limited local regions, while others act
globally over the entire image. However, the ambiguity problem in propagating edits is not adequately addressed in existing methods. Thus, tedious user input requirements remain since the user must densely or repeatedly input control samples to suppress ambiguity. In this paper, we address this challenge to propagate edits suitably by marking edits for local or global
propagation and determining their reasonable propagation scopes automatically. Thus, our approach avoids propagation conflicts, effectively resolving the ambiguity problem. With the reduction of ambiguity, our method allows fewer and less-precise control samples than existing methods. Furthermore, we provide a uniform
framework to propagate local and global edits simultaneously, helping the user to quickly obtain the intended results with reduced labor. With our unified framework, the potentially ambiguous interaction between local and global edits (evident
in existing methods that propagate these two edit types in series) is resolved. We experimentally demonstrate the effectiveness of our method compared with existing methods.

2015

3.

MATLAB2015_03

Face Recognition Across Non-Uniform Motion
Blur, Illumination, and Pose

Existing methods for performing face recognition
in the presence of blur are based on the convolution model and cannot handle non-uniform blurring situations that frequently arise from tilts and rotations in hand-held cameras. In this paper, we propose a methodology for face recognition in the presence of space-varying motion blur comprising of arbitrarily-shaped kernels. We model the blurred face as a convex combination of
geometrically transformed instances of the focused gallery face, and show that the set of all images obtained by non-uniformly blurring a given image forms a convex set. We first propose a non uniform blur-robust algorithm by making use of the assumption of a sparse camera trajectory in the camera motion space to build an energy function with l1-norm constraint on the camera
motion. The framework is then extended to handle illumination variations by exploiting the fact that the set of all images obtained from a face image by non-uniform blurring and changing the illumination forms a bi-convex set. Finally, we propose an elegant extension to also account for variations in pose.

2015

4.

MATLAB2015_04

Swarm Intelligence for Detecting Interesting Events
in Crowded Environments

This work focuses on detecting and localizing
anomalous events in videos of crowded scenes, i.e. divergences from a dominant pattern. Both motion and appearance information are considered, so as to robustly distinguish different kinds of anomalies, for a wide range of scenarios. A newly introduced concept based on swarm theory, Histograms of Oriented Swarms (HOS), is applied to capture the dynamics of crowded environments. HOS, together with the well known Histograms of Oriented Gradients (HOG), are combined to build a descriptor that effectively characterizes each scene. These appearance and motion features are only extracted within
spatiotemporal volumes of moving pixels to ensure robustness to local noise, increase accuracy in the detection of local, non dominant anomalies, and achieve a lower computational cost. Experiments on benchmark datasets containing various situations with human crowds, as well as on traffic data, led to results that
surpassed the current state of the art, confirming the method’s efficacy and generality. Finally, the experiments show that our approach achieves significantly higher accuracy, especially for pixel-level event detection compared to State of the Art (SoA)
methods, at a low computational cost.

2015

5.

MATLAB2015_05

Content-Based Image Retrieval Using Features
Extracted From Halftoning-Based Block
Truncation Coding

This paper presents a technique for Content-Based
Image Retrieval (CBIR) by exploiting the advantage of low complexity Ordered-Dither Block Truncation Coding (ODBTC) for the generation of image content descriptor. In encoding step, ODBTC compresses an image block into corresponding quantizers and bitmap image. Two image features are proposed to index an image, namely Color Co-occurrence Feature (CCF) and Bit Pattern Features (BPF), which are generated directly from
ODBTC encoded data streams without performing the decoding process. The CCF and BPF of an image are simply derived from the two ODBTC quantizers and bitmap, respectively, by involving the visual codebook. Experimental results show that the proposed method is superior to the Block Truncation Coding (BTC) image
retrieval systems and the other former methods, and thus prove that the ODBTC scheme is not only suited for image compression since of its simplicity, but also offers a simple and effective descriptor to index images in CBIR

2015

6.

MATLAB2015_06

Approximation and Compression with Sparse
Orthonormal Transforms

We propose a new transform design method that
targets the generation of compression-optimized transforms for next-generation multimedia applications. The fundamental idea behind transform compression is to exploit regularity within signals such that redundancy is minimized subject to a fidelity cost. Multimedia signals, in particular images and video, are well known to contain a diverse set of localized structures, leading to many different types of regularity and to non stationary signal
statistics. The proposed method designs sparse orthonormal transforms (SOT) that automatically exploit regularity over different signal structures and provides an adaptation method that determines the best representation over localized regions. Unlike earlier work that is motivated by linear approximation constructs and model-based designs that are limited to specific types of
signal regularity, our work uses general nonlinear approximation ideas and a data-driven setup to significantly broaden its reach. We show that our SOT designs provide a safe and principled
extension of the Karhunen-Loeve transform (KLT) by reducing to the KLT on Gaussian processes and by automatically exploiting non-Gaussian statistics to significantly improve over the KLT on more general processes. We provide an algebraic optimization
framework that generates optimized designs for any desired transform structure (multi-resolution, block, lapped, etc.) with significantly better n-term approximation performance. For each structure, we propose a new prototype codec and test over a
database of images. Simulation results show consistent increase in compression and approximation performance compared with conventional methods.

2015

7.

MATLAB2015_07

High-Resolution Face Verification Using
Pore-Scale Facial Features

Face recognition methods, which usually represent
face images using holistic or local facial features, rely heavily on alignment. Their performances also suffer a severe degradation under variations in expressions or poses, especially when there is one gallery per subject only. With the easy access to high resolution (HR) face images nowadays, some HR face databases
have recently been developed. However, few studies have tackled the use of HR information for face recognition or verification. In this paper, we propose a pose-invariant face-verification method, which is robust to alignment errors, using the HR information based on pore-scale facial features. A new key point descriptor, namely, pore-Principal Component Analysis (PCA)- Scale Invariant Feature Transform (PPCASIFT)—adapted from
PCA-SIFT—is devised for the extraction of a compact set of distinctive pore-scale facial features. Having matched the porescale features of two-face regions, an effective robust-fitting scheme is proposed for the face-verification task. Experiments show that, with one frontal-view gallery only per subject, our proposed method outperforms a number of standard verification
methods, and can achieve excellent accuracy even the faces are under large variations in expression and pose.

2015

8.

MATLAB2015_08

DERF: Distinctive Efficient Robust Features From
the Biological Modeling of the P Ganglion Cells

Studies in neuroscience and biological vision have
shown that the human retina has strong computational power, and its information representation supports vision tasks on both ventral and dorsal pathways. In this paper, a new local image descriptor, termed distinctive efficient robust features (DERF), is derived by modeling the response and distribution properties of the parvocellular-projecting ganglion cells in the primate
retina. DERF features exponential scale distribution, exponential grid structure, and circularly symmetric function difference of Gaussian (DoG) used as a convolution kernel, all of which are consistent with the characteristics of the ganglion cell array found in neurophysiology, anatomy, and biophysics. In addition,
a new explanation for local descriptor design is presented from the perspective of wavelet tight frames. DoG is naturally a wavelet, and the structure of the grid points array in our descriptor is closely related to the spatial sampling of wavelets. The DoG wavelet itself forms a frame, and when we modulate the parameters of our descriptor to make the frame tighter, the performance of the DERF descriptor improves accordingly. This
is verified by designing a tight frame DoG, which leads to
much better performance. Extensive experiments conducted in the image matching task on the multiview stereo correspondence data set demonstrate that DERF outperforms state of the art methods for both hand-crafted and learned descriptors, while remaining robust and being much faster to compute.

2015

9.

MATLAB2015_09

Blind Inpainting using ℓ0 and Total Variation
Regularization

In this paper, we address the problem of image
reconstruction with missing pixels or corrupted with impulse noise, when the locations of the corrupted pixels are not known. A logarithmic transformation is applied to convert the multiplication between the image and binary mask into an additive problem. The image and mask terms are then estimated iteratively with total variation regularization applied on the image, and ℓ0 regularization on the mask term which imposes sparseness on the support set of the missing pixels. The resulting
alternating minimization scheme simultaneously estimates the image and mask, in the same iterative process. The logarithmic transformation also allows the method to be extended to the Rayleigh multiplicative and Poisson observation models. The method can also be extended to impulse noise removal by relaxing
the regularizer from the ℓ0 norm to the ℓ1 norm. Experimental results show that the proposed method can deal with a larger fraction of missing pixels than two phase methods which first estimate the mask and then reconstruct the image.

2015

10.

MATLAB2015_10

A Source-Channel Coding Approach to Digital
Image Protection and Self-Recovery

Watermarking algorithms have been widely applied
to the field of image forensics recently. One of these very forensic applications is the protection of images against tampering. For this purpose, we need to design a watermarking algorithm fulfilling two purposes in case of image tampering: 1) detecting the tampered area of the received image and 2) recovering the lost information in the tampered zones. State-of-the-art techniques
accomplish these tasks using watermarks consisting of check bits and reference bits. Check bits are used for tampering detection, whereas reference bits carry information about the whole image. The problem of recovering the lost reference bits still stands. This paper is aimed at showing that having the tampering location known, image tampering can be modeled
and dealt with as an erasure error. Therefore, an appropriate design of channel code can protect the reference bits against tampering. In the present proposed method, the total watermark bit-budget is dedicated to three groups: 1) source encoder output bits; 2) channel code parity bits; and 3) check bits. In watermark embedding phase, the original image is source coded and the output bit stream is protected using appropriate
channel encoder. For image recovery, erasure locations detected by check bits help channel erasure decoder to retrieve the original source encoded image. Experimental results show that our proposed scheme significantly outperforms recent techniques in terms of image quality for both watermarked and recovered
image. The watermarked image quality gain is achieved through spending less bit-budget on watermark, while image recovery quality is considerably improved as a consequence of consistent performance of designed source and channel codes.

2015

11.

MATLAB2015_11

Structured Sparse Priors for Image Classification

Model-based compressive sensing (CS) exploits the
structure inherent in sparse signals for the design of better
signal recovery algorithms. This information about structure is often captured in the form of a prior on the sparse coefficients, with the Laplacian being the most common such choice (leading to l1-norm minimization). Recent work has exploited the discriminative capability of sparse representations for image classification by employing class-specific dictionaries in the CS framework. Our contribution is a logical extension of these ideas into structured sparsity for classification. We introduce the notion of discriminative class-specific priors in conjunction with class specific dictionaries, specifically the spike-and-slab prior widely applied in Bayesian sparse regression. Significantly, the proposed framework takes the burden off the demand for abundant training image samples necessary for the success of sparsity-based classification schemes. We demonstrate this practical benefit of our approach in important applications,
such as face recognition and object categorization.

2015

12.

MATLAB2015_12

Video Tracking Using Learned Hierarchical Features

In this paper, we propose an approach to learn
hierarchical features for visual object tracking. First, we offline learn features robust to diverse motion patterns from auxiliary video sequences. The hierarchical features are learned via a twolayer convolutional neural network. Embedding the temporal slowness constraint in the stacked architecture makes the learned
features robust to complicated motion transformations, which is important for visual object tracking. Then, given a target video sequence, we propose a domain adaptation module to online adapt the pre-learned features according to the specific target object. The adaptation is conducted in both layers of the deep feature learning module so as to include appearance information of the specific target object. As a result, the learned hierarchical features can be robust to both complicated motion transformations and appearance changes of target objects. We
integrate our feature learning algorithm into three tracking
methods. Experimental results demonstrate that significant improvement can be achieved by using our learned hierarchical features, especially on video sequences with complicated motion transformations.

2015

13.

MATLAB2015_13

A Global/Local Affinity Graph for Image
Segmentation

Construction of a reliable graph capturing perceptual grouping cues of an image is fundamental for
graph-cut based image segmentation methods. In this paper, we propose a novel sparse global/local affinity
graph over superpixels of an input image to capture both short and long range grouping cues, thereby
enabling perceptual grouping laws, e.g., proximity, similarity, continuity, to enter in action through a
suitable graph cut algorithm. Moreover, we also evaluate three major visual features, namely color, texture
and shape,for their effectiveness in perceptual segmentation and propose a simple graph fusion scheme
to implement some recent findings from psychophysics which suggest combining these visual features
with different emphases for perceptual grouping. Specifically, an input image is first oversegmented into
superpixels at different scales. We postulate a gravitation law based on empirical observations and divide
superpixels adaptively into small, medium and large sized sets. Global grouping is achieved using medium
sized superpixels through a sparse representation of superpixels’ features by solving a `0-minimization
problem, thereby enabling continuity or propagation of local smoothness over long range connections.
Small and large sized superpixels are then used to achieve local smoothness through an adjacent graph
in a given feature space, thus implementing perceptual laws, e.g., similarity and proximity. Finally, a
bipartite graph is also introduced to enable propagation of grouping cues between superpixels of different
scales. Extensive experiments are carried out on the Berkeley Segmentation Database in comparison with
several state of the art graph constructions.

2015

14.

MATLAB2015_14

A Database for Evaluating No-Reference
Image Quality Assessment Algorithms

This paper presents a new database, CID2013,
to address the issue of using no-reference (NR) image quality assessment algorithms on images with multiple distortions. Current NR algorithms struggle to handle images with many concurrent distortion types, such as real photographic images captured by different digital cameras. The database consists of six image sets; on average, 30 subjects have evaluated 12–14 devices depicting eight different scenes for a total of 79 different cameras, 480 images, and 188 subjects (67% female).
The subjective evaluation method was a hybrid absolute category rating-pair comparison developed for the study and presented in this paper. This method utilizes a slideshow of all images within a scene to allow the test images to work as references to each other. In addition to mean opinion score value, the images are also rated using sharpness, graininess, lightness, and color saturation scales. The CID2013 database contains images used
in the experiments with the full subjective data plus extensive background information from the subjects. The database is madefreely available for the research community.

2015

15.

MATLAB2015_15

An Efficient MRF Embedded Level Set Method for
Image Segmentation

This paper presents a fast and robust level set
method for image segmentation. To enhance the robustness against noise, we embed a Markov random field (MRF) energy function to the conventional level set energy function. This MRF energy function builds the correlation of a pixel with its neighbors and encourages them to fall into the same region. To obtain
a fast implementation of the MRF embedded level set model, we explore algebraic multigrid (AMG) and sparse field method (SFM) to increase the time step and decrease the computation domain, respectively. Both AMG and SFM can be conducted in a parallel fashion, which facilitates the processing of our method for big image databases. By comparing the proposed fast and
robust level set method with the standard level set method and its popular variants on noisy synthetic images, synthetic aperture radar (SAR) images, medical images and natural images, we comprehensively demonstrate the new method is robust against various kinds of noises. Especially, the new level set method can segment an image of size 500 by 500 within three seconds on
MATLAB R2010b installed in a computer with 3.30GHz CPU and 4GB memory.

2015

16.

MATLAB2015_16

Weighted Guided Image Filtering

It is known that local filtering-based edgepreserving smoothing techniques suffer from halo artifacts.
In this paper, a weighted guided image filter (WGIF) is introduced by incorporating an edge-aware weighting into an existing guided image filter (GIF) to address the problem. The WGIF inherits advantages of both global and local smoothing filters in the sense that: 1) the complexity of the WGIF is O(N) for an image with N pixels, which is same as the GIF and 2) the WGIF
can avoid halo artifacts like the existing global smoothing filters. The WGIF is applied for single image detail enhancement, single image haze removal, and fusion of differently exposed images. Experimental results show that the resultant algorithms produce images with better visual quality and at the same time halo artifacts can be reduced/avoided from appearing in the final images with negligible increment on running times.

2015

17.

MATLAB2015_17

Distinctive Efficient Robust Features From
the Biological Modeling of the P Ganglion Cells

Studies in neuroscience and biological vision have
shown that the human retina has strong computational power, and its information representation supports vision tasks on both ventral and dorsal pathways. In this paper, a new local image descriptor, termed distinctive efficient robust features (DERF), is derived by modeling the response and distribution properties of the parvocellular-projecting ganglion cells in the primate retina. DERF features exponential scale distribution, exponential grid structure, and circularly symmetric function difference of
Gaussian (DoG) used as a convolution kernel, all of which are consistent with the characteristics of the ganglion cell array found in neurophysiology, anatomy, and biophysics. In addition, a new explanation for local descriptor design is presented from the perspective of wavelet tight frames. DoG is naturally a
wavelet, and the structure of the grid points array in our
descriptor is closely related to the spatial sampling of wavelets. The DoG wavelet itself forms a frame, and when we modulate the parameters of our descriptor to make the frame tighter, the performance of the DERF descriptor improves accordingly. This is verified by designing a tight frame DoG, which leads to much better performance. Extensive experiments conducted in the image matching task on the multiview stereo correspondence data set demonstrate that DERF outperforms state of the art methods for both hand-crafted and learned descriptors, while remaining robust and being much faster to compute.

2015

18.

MATLAB2015_18

Multi-task Pose-Invariant Face Recognition

Face images captured in unconstrained environments usually contain significant pose variation, which dramatically degrades the performance of algorithms designed to recognize frontal faces. This paper proposes a novel face identification framework capable of handling the full range of pose variations within ±90° of yaw. The proposed framework first transforms the original pose-invariant face recognition problem into a partial frontal face recognition problem. A robust patch-based face representation scheme is then developed to represent the synthesized partial frontal faces. For each patch,
a transformation dictionary is learnt under the proposed multitask learning scheme. The transformation dictionary transforms the features of different poses into a discriminative subspace. Finally, face matching is performed at patch level rather than at the holistic level. Extensive and systematic experimentation on FERET, CMU-PIE, and Multi-PIE databases shows that the proposed method consistently outperforms single-task-based baselines as well as state-of-the-art methods for the pose problem. We further extend the proposed algorithm for the unconstrained face verification problem and achieve top-level performance on the challenging LFW data set.

2015

19.

MATLAB2015_19

A Feature-Enriched Completely Blind Image
Quality Evaluator

Existing blind image quality assessment (BIQA)
methods are mostly opinion-aware. They learn regression models from training images with associated human subjective scores to predict the perceptual quality of test images. Such opinion-aware methods, however, require a large amount of training samples with associated human subjective scores and of a variety of distortion types. The BIQA models learned by opinion-aware methods often have weak generalization capability, hereby limiting their usability in practice. By comparison, opinion-unaware methods do not need human subjective scores for training, and thus have greater potential for good generalization capability. Unfortunately, thus far no opinion-unaware BIQA method has shown consistently better quality prediction accuracy than the opinion-aware methods. Here, we aim to develop an opinion unaware BIQA method that can compete with, and perhaps outperform, the existing opinion-aware methods. By integrating the features of natural image statistics derived from multiple cues, we learn a multivariate Gaussian model of image patches from a collection of pristine natural images. Using the learned multivariate Gaussian model, a Bhattacharyya-like distance is used to measure the quality of each image patch, and then an overall quality score is obtained by average pooling. The proposed BIQA method does not need any distorted sample images nor subjective quality scores for training, yet extensive experiments demonstrate its superior quality-prediction performance to the state-of-the-art opinion-aware BIQA methods.

2015

20.

MATLAB2015_20

Spatiotemporal Saliency Detection for Video
Sequences Based on Random Walk With Restart

A novel saliency detection algorithm for video
sequences based on the random walk with restart (RWR) is proposed in this paper. We adopt RWR to detect spatially and temporally salient regions. More specifically, we first find a temporal saliency distribution using the features of motion distinctiveness, temporal consistency, and abrupt change. Among them, the motion distinctiveness is derived by comparing the motion profiles of image patches. Then, we employ the temporal
saliency distribution as a restarting distribution of the random walker. In addition, we design the transition probability matrix for the walker using the spatial features of intensity, color, and compactness. Finally, we estimate the spatiotemporal saliency distribution by finding the steady-state distribution of the walker.
The proposed algorithm detects foreground salient objects
faithfully, while suppressing cluttered backgrounds effectively, by incorporating the spatial transition matrix and the temporal restarting distribution systematically. Experimental results on various video sequences demonstrate that the proposed algorithm outperforms conventional saliency detection algorithms qualitatively and quantitatively.

2015

21.

MATLAB2015_21

Sorted Consecutive Local Binary Pattern
for Texture Classification

In this paper, we propose a sorted consecutive local
binary pattern (scLBP) for texture classification. Conventional methods encode only patterns whose spatial transitions are not more than two, whereas scLBP encodes patterns regardless of their spatial transition. Conventional methods do not encode
patterns on account of rotation-invariant encoding; on the other hand, patterns with more than two spatial transitions have discriminative power. The proposed scLBP encodes all patterns with any number of spatial transitions while maintaining their rotation-invariant nature by sorting the consecutive patterns. In addition, we introduce dictionary learning of scLBP based on kd-tree which separates data with a space partitioning strategy. Since the elements of sorted consecutive patterns lie in different space, it can be generated to a discriminative code with kd-tree. Finally, we present a framework in which scLBPs
and the kd-tree can be combined and utilized. The results
of experimental evaluation on five texture data sets—Outex, CUReT, UIUC, UMD, and KTH-TIPS2-a—indicate that our proposed framework achieves the best classification rate on the CUReT, UMD, and KTH-TIPS2-a data sets compared with conventional methods. The results additionally indicate that only a marginal difference exists between the best classification rate
of conventional methods and that of the proposed framework on the UIUC and Outex data sets.

2015

22.

MATLAB2015_22

Robust 2D Principal Component Analysis:
A Structured Sparsity Regularized Approach

Principal component analysis (PCA) is widely
used to extract features and reduce dimensionality in various computer vision and image/video processing tasks. Conventional approaches either lack robustness to outliers and corrupted data or are designed for one-dimensional signals. To address this problem, we propose a robust PCA model for two-dimensional
images incorporating structured sparse priors, referred to
as structured sparse 2D-PCA. This robust model considers the prior of structured and grouped pixel values in two dimensions. As the proposed formulation is jointly nonconvex and nonsmooth, which is difficult to tackle by joint optimization, we develop a two-stage alternating minimization approach to solve the problem. This approach iteratively learns the projection matrices by bidirectional decomposition and utilizes the proximal method to obtain the structured sparse outliers.
By considering the structured sparsity prior, the proposed
model becomes less sensitive to noisy data and outliers in two dimensions. Moreover, the computational cost  indicates that the robust two-dimensional model is capable of processing quarter common intermediate format video in real time, as well as
handling large-size images and videos, which is often intractable with other robust PCA approaches that involve image-to-vector conversion. Experimental results on robust face reconstruction, video background subtraction data set, and real-world videos
show the effectiveness of the proposed model compared with conventional 2D-PCA and other robust PCA algorithms.

2015

23.

MATLAB2015_23

Accurate Vessel Segmentation With
Constrained B-Snake

We describean active contour framework with
accurate shape and size constraints on the vessel cross-sectional planes to produce the vessel segmentation. It starts with a multiscale vessel axis tracing in a 3D computed tomography (CT) data, followed by vessel boundary delineation on the cross-sectional planes derived from the extracted axis. The vessel boundary surface is deformed under constrained movements on the cross sections and is voxelized to produce the final vascular segmentation. The novelty of this paper lies
in the accurate contour point detection of thin vessels based on the CT scanning model, in the efficient implementation of missing contour points in the problematic regions and in the active contour model with accurate shape and size constraints. The main advantage of our framework is that it avoids disconnected and incomplete segmentation of the vessels in the problematic regions that contain touching vessels (vessels in close
proximity to each other), diseased portions (pathologic structure attached to a vessel), and thin vessels. It is  particularly suitable for accurate segmentation of thin and low contrast vessels. Our method is evaluated and demonstrated on CT data sets from our partner site, and its results are compared with three related
methods. Our method is also tested on two publicly available databases and its results are compared with the recently published method. The applicability of the proposed method to some challenging clinical problems, the segmentation of the vessels in the problematic regions, is demonstrated with good results on both quantitative and qualitative experimentations; our segmentation algorithm can delineate vessel boundaries that have level of variability similar to those obtained manually.

2015

24.

MATLAB2015_24

PatchMatch With Potts Model for Object
Segmentation and Stereo Matching

This paper presents a unified variational formulation for joint object segmentation and stereo matching, which takes both accuracy and efficiency into account. In our approach, depth-map consists of compact objects, each object is represented through three different aspects: 1) the perimeter in image space; 2) the slanted object depth plane; and 3) the planar bias, which is to add an additional level of detail on top of each object plane in order to model depth variations within an object. Compared with traditional high quality solving methods in low level, we use a convex formulation of the multilabel Potts Model
with PatchMatch stereo techniques to generate depth-map at each image in object level and show that accurate multiple view reconstruction can be achieved with our formulation by means of induced homography without discretization or staircasing artifacts. Our model is formulated as an energy minimization that is optimized via a fast primal-dual algorithm, which can handle several hundred object depth segments efficiently. Performance evaluations in the Middlebury benchmark data sets
show that our method outperforms the traditional integer-valued disparity strategy as well as the original PatchMatch algorithm and its variants in subpixel accurate disparity estimation. The proposed algorithm is also evaluated and shown to produce consistently good results for various real-world data sets (KITTI benchmark data sets and multiview benchmark
data sets).

2015

25.

MATLAB2015_25

Robust Representation and Recognition of
Facial Emotions Using Extreme Sparse
Learning

Recognition of natural emotions from human faces is an interesting topic with a wide range of potential applications like human-computer interaction, automated tutoring systems, image and video retrieval, smart environments, and driver warning systems. Traditionally, facial emotion recognition systems have been evaluated on laboratory controlled data, which is not representative of the environment faced in real-world applications. To robustly recognize facial emotions in real-world natural situations, this paper proposes an approach called Extreme Sparse Learning (ESL), which has the ability to jointly learn a dictionary (set of basis) and a non-linear classification model. The proposed approach combines the discriminative power of Extreme Learning Machine (ELM) with the reconstruction property of sparse representation to enable accurate classification when presented with noisy signals and imperfect data
recorded in natural settings. Additionally, this work presents a new local spatio-temporal descriptor that is distinctive and pose-invariant. The proposed framework is able to achieve state-of-the-art recognition accuracy on both acted and spontaneous facial emotion databases.

2015

26.

MATLAB2015_26

Adaptive Image Denoising by Targeted Databases

We propose a data-dependent denoising procedure
to restore noisy images. Different from existing denoising
algorithms which search for patches from either the noisy image or a generic database, the new algorithm finds patches from a database that contains relevant patches. We formulate the denoising problem as an optimal filter design problem and make two contributions. First, we  determine the basis function ofthe denoising filter by solving a group sparsity minimization problem. The optimization formulation generalizes existing denoising algorithms and offers systematic analysis of the performance. Improvement methods are proposed to enhance the patch search process. Second, we determine the spectral coefficients of the denoising filter by considering a localized Bayesian prior. The
localized prior leverages the similarity of the targeted database, alleviates the intensive Bayesian computation, and links the new method to the classical linear minimum mean squared error estimation. We demonstrate applications of the proposed method in a variety of scenarios, including text images, multiview images,
and face images. Experimental results show the superiority of the new algorithm over existing methods.

2015

27.

MATLAB2015_27

Progressive Halftone Watermarking Using
Multi-layer Table Lookup Strategy

In this work, a halftoning-based multi-layer watermarking of low computational complexity is proposed. An additional data hiding technique is also employed to embed multiple watermarks into the watermark to be embedded to improve the security and embedding capacity. At the encoder, the Efficient Direct Binary Search (EDBS) method is employed to generate 256 reference tables to ensure the output is in halftone format. Subsequently, watermarks are embedded by a set of optimized compressed tables with various textural angles for table lookup. At the decoder, the Least-MeanSquare (LMS) metric is considered to increases the differences among those generated phenotypes.

2015

28.

MATLAB2015_28

Learning Multiple Linear Mappings for Efficient
Single Image Super-Resolution

Example learning-based superresolution (SR)
algorithms show promise for restoring a high-resolution (HR) image from a single low-resolution (LR) input. The most popular approaches, however, are either time- or space-intensive, which limits their practical applications in many resource-limited settings. In this paper, we propose a novel computationally efficient single image SR method that learns multiple linear
mappings (MLM) to directly transform LR feature subspaces into HR subspaces. In particular, we first partition the large nonlinear feature space of LR images into a cluster of linear subspaces. Multiple LR subdictionaries are then learned, followed by inferring the corresponding HR subdictionaries based on the
assumption that the LR–HR features share the same representation coefficients. We establish MLM from the input LR features to the desired HR outputs in order to achieve fast yet stable SR recovery. Furthermore, in order to suppress displeasing artifacts generated by the MLM-based method, we apply a fast nonlocal means algorithm to construct a simple yet effective similaritybased regularization term for SR enhancement. Experimental results indicate that our approach is both quantitatively and qualitatively superior to other application-oriented SR methods, while maintaining relatively low time and space complexity.

2015

29.

MATLAB2015_29

Cross-Domain Person Re-Identification Using
Domain Adaptation Ranking SVMs

This paper addresses a new person re-identification
problem without label information of persons under non overlapping target cameras. Given the matched (positive) and unmatched (negative) image pairs from source domain cameras, as well as unmatched (negative) and unlabeled image pairs from target domain cameras, we propose an Adaptive Ranking Support Vector Machines (AdaRSVM) method for re-identification under target domain cameras without person labels. To overcome
the problems introduced due to the absence of matched (positive) image pairs in the target domain, we relax the discriminative constraint to a necessary condition only relying on the positive mean in the target domain. To estimate the target positive mean, we make use of all the available data from source and target domains as well as constraints in person re-identification. Inspired by adaptive learning methods, a new discriminative
model with high confidence in target positive mean and low confidence in target negative image pairs is developed by refining the distance model learnt from the source domain. Experimental results show that the proposed AdaRSVM outperforms existing supervised or unsupervised, learning or non-learning reidentification methods without using label information in target
cameras. Moreover, our method achieves better re-identification performance than existing domain adaptation methods derived under equal conditional probability assumption.

2015

30.

MATLAB2015_30

Structure-Sensitive Saliency Detection
via Multilevel Rank Analysis in
Intrinsic Feature Space

This paper advocates a novel multiscale,
structure-sensitive saliency detection method, which can
distinguish multilevel, reliable saliency from various natural pictures in a robust and versatile way. One key challenge for saliency detection is to guarantee the entire salient object being characterized differently from nonsalient background. To tackle this, our strategy is to design a structure-aware descriptor based on the intrinsic biharmonic distance metric. One benefit of introducing this descriptor is its ability to simultaneously integrate local and global structure information, which is extremely valuable for separating the salient object from nonsalient background in a multiscale sense. Upon devising such powerful shape descriptor, the remaining challenge is
to capture the saliency to make sure that salient subparts
actually stand out among all possible candidates. Toward this goal, we conduct multilevel low-rank and sparse analysis in the intrinsic feature space spanned by the shape descriptors defined on over-segmented super-pixels. Since the low-rank property emphasizes much more on stronger similarities among super-pixels, we naturally obtain a scale space along the rank
dimension in this way. Multiscale saliency can be obtained by simply computing differences among the low-rank components across the rank scale. We conduct extensive experiments on some public benchmarks, and make comprehensive, quantitative evaluation between our method and existing state-of-the-art techniques. All the results demonstrate the superiority of our method in accuracy, reliability, robustness, and versatility.

2015

31.

MATLAB2015_31

Depth Reconstruction From Sparse Samples:
Representation, Algorithm, and Sampling

The rapid development of 3D technology and
computer vision applications has motivated a thrust of
methodologies for depth acquisition and estimation. However, existing hardware and software acquisition methods have limited performance due to poor depth precision, low resolution, and high computational cost. In this paper, we present a computationally efficient method to estimate dense depth maps from sparse measurements. There are three main contributions. First, we provide empirical evidence that depth maps can be encoded much more sparsely than natural images using common
dictionaries, such as wavelets and contourlets. We also show that a combined wavelet–contourlet dictionary achieves better performance than using either dictionary alone. Second, we propose an alternating direction method of multipliers (ADMM) for depth map reconstruction. A multiscale warm start procedure
is proposed to speed up the convergence. Third, we propose a two-stage randomized sampling scheme to optimally choose the sampling locations, thus maximizing the reconstruction performance for a given sampling budget. Experimental results show that the proposed method produces high-quality dense depth estimates, and is robust to noisy measurements. Applications to real data in stereo matching are demonstrated.

2015

32.

MATLAB2015_32

Image Denoising by Exploring External
and Internal Correlations

Single image denoising suffers from limited data
collection within a noisy image. In this paper, we propose a novel image denoising scheme, which explores both internal and external correlations with the help of web images. For each noisy patch, we build internal and external data cubes by finding similar patches from the noisy and web images, respectively. We then propose reducing noise by a two-stage strategy using different filtering approaches. In the first stage, since the noisy patch may lead to inaccurate patch selection, we
propose a graph based optimization method to improve patch matching accuracy in external denoising. The internal denoising is frequency truncation on internal cubes. By combining the internal and external denoising patches, we obtain a preliminary denoising result. In the second stage, we propose reducing noise by filtering of external and internal cubes, respectively, on transform domain. In this stage, the preliminary denoising result
not only enhances the patch matching accuracy but also provides reliable estimates of filtering parameters. The final denoising image is obtained by fusing the external and internal filtering results. Experimental results show that our method constantly outperforms state-of-the-art denoising schemes in both subjective and objective quality measurements, e.g., it achieves >2 dB gain
compared with BM3D at a wide range of noise levels.

2015

33.

MATLAB2015_33

Motion-Compensated Coding and Frame Rate
Up-Conversion: Models and Analysis

Block-based motion estimation (ME) and motion
compensation (MC) techniques are widely used in modern video processing algorithms and compression systems. The great variety of video applications and devices results in diverse compression specifications, such as frame rates and bit rates.
In this paper, we study the effect of frame rate and compression bit rate on block-based ME and MC as commonly utilized in inter-frame coding and frame rate up-conversion (FRUC). This joint examination yields a theoretical foundation for comparing MC procedures in coding and FRUC. First, the video signal is locally modeled as a noisy translational motion of an image.
Then, we theoretically model the motion-compensated prediction of available and absent frames as in coding and FRUC applications, respectively. The theoretic MC-prediction error is studied further and its autocorrelation function is calculated, yielding useful separable-simplifications for the coding application. We argue that a linear relation exists between the variance of the MC-prediction error and temporal distance. While the relevant distance in MC coding is between the predicted and reference frames, MC-FRUC is affected by the distance between the frames available for interpolation. We compare our estimates with experimental results and show that the theory explains qualitatively the empirical behavior. Then, we use the models proposed to analyze a system for improving of video coding at low bit rates, using a spatiotemporal scaling. Although this concept is practically employed in various forms, so far it lacked a theoretical justification. We here harness the proposed MC models and present a comprehensive analysis of the system, to qualitatively predict the experimental
results.

2015

34.

MATLAB2015_34

Fractal Analysis for Reduced Reference
Image Quality Assessment

In this paper, multifractal analysis is adapted to
reduced-reference image quality assessment (RR-IQA). A novel RR-QA approach is proposed, which measures the difference of spatial arrangement between the reference image and the distorted image in terms of spatial regularity measured by fractal dimension. An image is first expressed in Log-Gabor domain. Then, fractal dimensions are computed on each Log-Gabor subband and concatenated as a feature vector. Finally, the
extracted features are pooled as the quality score of the distorted image using 1 distance. Compared with existing approaches, the proposed method measures image quality from the perspective of the spatial distribution of image patterns. The proposed method was evaluated on seven public benchmark data sets. Experimental results have demonstrated the excellent performance of the proposed method in comparison with state-of-the-art approaches.

2015

35.

MATLAB2015_35

Criteria-Based Modulation for Multilevel Inverters

Pulse-width modulation schemes are aimed at adjusting the fundamental component while reducing the harmonic
content of an inverter output voltage or current. This paper addresses the topic of optimal inverter operation in reference to a given objective function. The objective function could embody either a single performance criterion, such as voltage or current total harmonic distortion, or a weighted sum of multiple criteria.
The proposed method ensures primacy of the chosen solution while imposing no restriction over its modulation index. In particular, operating the inverter by the chosen solution would result in performances superior to any other modulation scheme commutating in an equal number of switching angles per fundamental cycle. The proposed method allows for the consideration
of practical inverter constraints and prevents the possibility of impractical switching sequence. A detailed investigation of the method is given, accompanied by two practical cases minimizing, respectively, phase-voltage THD and line-current THD of a three level inverter. Selected simulation and experimental results are
presented to validate the theoretical part.

2015

36.

MATLAB2015_36

A Fully Soft-Switched Single Switch Isolated
DC-DC Converter

This paper proposes a soft-switched single switch
isolated converter. The proposed converter is able to offer low cost and high power density in step up application due to the following features: ZCS turn-on and ZVS turn-off of switch and ZCS turn-off of diodes regardless of voltage and load variation; low rated lossless snubber; reduced transformer volume compared to flyback based converters due to low magnetizing current. Experimental results on a 100kHz, 250W prototype are provided to validate the proposed concept.

2015

37.

MATLAB2015_37

Functional Modeling of Symmetrical Multipulse
Autotransformer Rectifier Units
for Aerospace Applications

This paper aims to develop a functional model of symmetrical multipulse autotransformer rectifier units (ATRUs) for more-electric aircraft (MEA) applications. The ATRU is seen as the most reliable way readily to be applied in the MEA. Interestingly, there is no model of ATRUs suitable for unbalanced or faulty
conditions at the moment. This paper is aimed to fill this gap and develop functional models suitable for both balanced and unbalanced conditions. Using the fact that the dc voltage and current are strongly related to the voltage and current vectors at the ac terminals of ATRUs, a functional model has been developed for the asymmetric ATRUs. The developed functional models are validated through simulation and experiment. The efficiency of the developed model is also demonstrated by comparing with corresponding detailed switching models. The developed functional model shows significant improvement of simulation efficiency, especially under balanced conditions.

2015

38.

MATLAB2015_38

Model Predictive Control Methods to Reduce Common-Mode Voltage
for Three-Phase Voltage Source Inverters

In this paper, we propose model predictive control methods to reduce the common-mode voltage of three-phase voltage source inverters (VSIs). In the reduced common-mode voltage-model predictive control (RCMV-MPC) methods proposed in this paper, only nonzero voltage vectors are utilized to reduce the common-mode voltage as well as to control the load currents. In addition, two nonzero voltage vectors are selected from the cost function at every sampling period, instead of using only one optimal vector during one sampling period. The two selected nonzero vectors are distributed in one sampling period in such a way as to minimize the error between
the measured load current and the reference. Without utilizing the zero vectors, the common-mode voltage controlled by the proposed RCMV-MPC algorithms can be restricted within ±Vdc/6. Furthermore, application of the two nonzero vectors with optimal time sharing
between them can yield satisfactory load current ripple performance without using the zero vectors. Thus, the proposed RCMV-MPC methods can reduce the common-mode voltage as well as control the load currents with fast transient response and satisfactory load current ripple performance compared with the conventional model predictive control method. Simulation and experimental results are included to verify the effectiveness of the proposed RCMV-MPC methods.

2015

39.

MATLAB2015_39

Interleaved Phase-Shift Full-Bridge Converter With
Transformer Winding Series–Parallel Autoregulated
(SPAR) Current Doubler Rectifier

The analysis and design guidelines for a two-phase interleaved phase-shift full-bridge converter with transformer winding series–parallel autoregulated current doubler rectifier are presented in this paper. The secondary windings of two transformers
work in parallel when the equivalent duty cycle is smaller than 0.25 but in series when the duty cycle is larger than 0.25 owing to the series–parallel autoregulated rectifier. With the proposed rectifying structure, the voltage stress of the rectifier is reduced. Also, the interleaving operation reduces the output current ripple. A 1-kW prototype with 200–400-V input and 50-V/20-A output is built up
to verify the theoretical analysis.

2015

40.

MATLAB2015_40

Analysis of Active-Network Converter with Coupled
Inductors

High step-up voltage gain DC/DC converters are widely applied in fuel cell stacks, photovoltaic arrays,
battery sources, and high intensity discharge (HID) lamps power systems. Active-network converters with coupled
inductors (CL-ANC) are derived from switched inductor active-network converters (SL-ANC). The proposed converter contains two coupled inductors which can be integrated into one magnetic core and two power switches. The converter can provide a relatively high voltage conversion ratio with a small duty cycle; the voltage and current stress of power switches are low which is helpful to reduce the losses. This paper shows the key waveforms of the CL-ANC and detailed derivation of the steady-state operation principle. The voltage conversion ratio and the effect of the leakage
inductance on voltage gain are discussed. The voltage stress and current stress on the power devices are illustrated and the comparison between the proposed converter and SL-ANC are given. Finally, the prototype has been established in the lab with 200V and 400V output under different turn ratios. Experimental results are given to verify the correctness of the analysis.

2015

41.

MATLAB2015_41

Modeling and Controller Design of a Semi-Isolated
Multi-Input Converter for Hybrid PV/Wind Power
Charger System

The objective of this paper is to propose the
development of a multi-input dc-dc converter (MIC) family which is composed of isolated and/or non-isolated dc-dc converters. By analyzing five basic isolated dc-dc converters, four isolated pulsating voltage source cells (I-PVSCs) and three isolated pulsating current source cells (I-PCSCs) are generated. Moreover, a semi-isolated multi-input converter (S-MIC) for hybrid PV/wind power charger system which can simplify the power
system, reduce the cost, deliver continuous power and overcome high voltage-transfer-ratio problems is proposed. In this paper, the operational principle of the proposed S-MIC is explained, the small-signal ac model is derived and the controller design is developed. Computer simulations and experimental results are presented to verify the accuracy of the proposed small signal ac model and the performance of the proposed S-MIC.

2015

42.

MATLAB2015_42

A Four-Switch Three-Phase SEPIC-Based Inverter

The four-switch three-phase (FSTP) inverter has been proposed as an innovative inverter design to
reduce the cost, complexity, size, and switching losses of
the DC-AC conversion system. Traditional FSTP inverter
usually operates at half the DC input voltage, hence, the
output line voltage cannot exceed this value. This paper
proposes a novel design for the FSTP inverter based on the topology of the single-ended primary-inductance converter (SEPIC). The proposed topology provides pure sinusoidal output voltages with no need for output filter. Compared to traditional FSTP inverter, the proposed FSTP SEPIC inverter improves the voltage utilization factor of the input DC supply, where the proposed topology provides higher output line voltage which can be extended up to the full value of the DC input voltage. The integral sliding-mode control is used with the proposed topology to optimize its dynamics and to ensure robustness of the system during different operating conditions. Derivation of the equations describing the parameters design, components ratings, and the operation of the proposed SEPIC inverter is presented in this paper. Simulation model and experimental setup are used to validate the proposed concept. Simulations and experimental results show the effectiveness of the proposed inverter.

2015

43.

MATLAB2015_43

High-Efficiency Isolated Single-Input Multiple-Output
Bidirectional Converter

This study presents a high-efficiency isolated single-input multiple-output bidirectional (HISMB) converter for a power storage system. According to the power management, the proposed HISMB converter can operate at a step-up state (energy release) and a step-down state (energy storage). At the step-up state, it can boost the voltage of a low-voltage input power source to a high-voltage-side dc bus and middle-voltage terminals. When the high-voltage-side dc bus has excess energy, one can reversely transmit the energy. The high-voltage dc bus can take as the main power, and middle-voltage output terminals can supply powers for individual middle-voltage dc loads or to charge auxiliary power sources (e.g., battery modules). In this study, a coupled-inductor-based HISMB converter accomplishes the bidirectional power control with the properties of voltage clamping and soft switching, and the corresponding device specifications are adequately designed. As a result, the energy of the leakage inductor of the coupled inductor can be recycled and released to the high-voltage-side dc bus and auxiliary power sources, and the voltage stresses on power switches can be greatly reduced. Moreover, the switching losses can be significantly decreased because of all power switches with zero-voltage-switching (ZVS) features. Therefore, the objectives of high-efficiency power conversion, electric isolation, bidirectional energy transmission, and various output voltage with different levels can be obtained. The effectiveness of the proposed HISMB converter is verified by experimental results of a kW-level prototype in practical applications.

2015

44.

MATLAB2015_44

Modularized Control Strategy and Performance
Analysis of DFIG System under Unbalanced and
Harmonic Grid Voltage

The paper presents a modularized control
strategy of doubly fed induction generator (DFIG) system, including the grid-side converter (GSC) and rotor-side converter (RSC), under unbalanced and harmonic grid voltage. The sequence decomposition process and complicated control reference calculation can be avoided in the proposed control strategy. From the perspective of power grid friendly-operation, the control targets of DFIG system in this paper are chosen as: 1) smooth active and reactive power injected into the power grid; 2) balanced and sinusoidal current injected into the power grid. The RSC and GSC can work as two independent modules and the communication between RSC and GSC can be removed. Furthermore, the 3rd harmonic current component, DC link voltage fluctuation and electromagnetic torque pulsation under the different
control targets are theoretically analyzed. Finally, the
availability of the proposed modularized control strategy
of DFIG system under unbalanced and distorted grid
voltage is verified by experiment results.

2015

45.

MATLAB2015_45

Resonant Switched-Capacitor Voltage
Regulator with Ideal Transient Response

A new, small and efficient voltage regulator,
realized using a resonant switched capacitor converter
technology, is introduced. Voltage regulation is
implemented by means of simple digital pulse density
modulation. It displays an ideal transient response with a
zero-order nature to all disturbance types. The newly
developed topology acts as a gyrator with a wide range of
voltage conversion ratios (below as well as above unity)
with constant efficiency characteristics for the entire
operation range. The operation of the voltage regulator is
verified on a 20W experimental prototype, demonstrating
ideal transient recovery without over/under-shoots in
response to load and line transients. Simple design
guidelines for the voltage regulation system are provided
and verified by experiments.

2015

46.

MATLAB2015_46

On the Performance of Multiobjective Evolutionary
Algorithms in Automatic Parameter Extraction of
Power Diodes

In this paper, a general, robust, and automatic
parameter extraction of nonlinear compact models is presented. The parameter extraction is based on multiobjective optimization using evolutionary algorithms which allow fitting of several highly
nonlinear and highly conflicting characteristics simultaneously. Two multiobjective evolutionary algorithms which have been proved to be robust for a wide range of multiobjective problems [1]–[3], the Nondominated Sorting Genetic Algorithm II
and the Multiobjective Covariance Matrix Adaptation Evolution Strategy, are used in the parameter extraction of a novel power diode compact model based on the lumped charge technique. The performance of the algorithms is assessed using a systematic statistical approach. Good agreement between the simulated and
measured characteristics of the power diode shows the accuracy of the used compact model and the efficiency and effectiveness of the proposed multiobjective optimization scheme.

2015

47.

MATLAB2015_47

Development of a Wind Interior Permanent-Magnet
Synchronous Generator Based Microgrid and Its
Operation Control

This paper presents the development of a wind
interior permanent-magnet synchronous generator (IPMSG) based DC micro-grid and its operation control. First, the derated characteristics of PMSG systems with various AC/DC converters and operation controls are comparatively analyzed. Then the IPMSG followed by three-phase Vienna switch mode rectifier (SMR) is developed to establish the common DC bus of DC micro-grid. Good developed power and voltage regulation characteristics are achieved via the proposed commutation tuning, robust current and voltage controls.
Second, a single-phase three-wire (1P3W) inverter is
constructed to serve as the test load. Good AC 220V/110V output voltage waveforms under unknown and nonlinear loads are preserved by the developed robust waveform tracking control scheme. Third, a battery energy storage system (BESS) is established, and the fast energy storage support response is obtained via the proposed droop control approach with adaptive predictive current control method. In addition, a chopped dump load is equipped to enhance the energy balance control flexibility.

2015

48.

MATLAB2015_48

A Novel Drive Method for High-Speed
Brushless DC Motor Operating in a Wide
Range

In this paper, a novel drive method, which is different from the traditional motor drive techniques, for high-speed brushless DC (BLDC) motor is proposed and verified by a series of experiments. It is well known that the BLDC motor can be driven by either Pulse-Width Modulation (PWM) techniques with a constant DC-link voltage or Pulse-Amplitude Modulation (PAM) techniques with an adjustable DC-link voltage. However, to our best knowledge, there is rare study providing a proper drive method for high-speed BLDC motor with a large power over a wide speed range. Therefore, the detailed theoretical analysis comparison of the PWM control and the PAM control for high-speed BLDC motor is first given. Then a conclusion that the PAM control is superior to the PWM control at high speed is obtained
because of decreasing the commutation delay and high frequency harmonic wave. Meanwhile, a new high-speed BLDC motor drive method based on the hybrid approach combining PWM and PAM is proposed. At last, the feasibility and effectiveness of the performance analysis comparison and the new drive method are verified by several experiments.

2015

49.

MATLAB2015_49

The Dynamic Control of Reactive Power for the
Brushless Doubly Fed Induction Machine with
Indirect Stator-quantities Control Scheme

Compared to the doubly fed induction
machine (DFIM), the brushless doubly fed induction
machine (BDFIM) has higher reliability by virtue of the
absence of a brush gear. Recent research on structure
optimization design and control strategy of BDFIM has
made remarkable progress. BDFIM indirect
stator-quantities control (ISC) is a new control strategy,
which, in comparison to vector control strategy, requires
fewer parameters and does not need rotating coordinate
transformation. This paper further develops the dynamic
control of reactive power for the BDFIM with ISC scheme. Detailed theoretical analysis is done to show the controller structure of the reactive power. The experimental results of the prototype show the feasibility of the proposed scheme. As a result, the proposed ISC controllers have been able to control not only speed and torque, but also the reactive power.

2015

50.

MATLAB2015_50

An LCL-LC Filter for Grid-Connected
Converter: Topology, Parameter and Analysis

In order to further cut down the cost of filter for
grid-connected pulse width modulation (PWM) converter under the more and more stringent grid code, a new kind of high order filter, named LCL-LC filter, is presented in this paper. The resonant frequency characteristics of the filter are analyzed and a parameter design method on the base of the characteristics is also proposed in the paper. The proposed parameter design method can easily make full use of the existing research results about the traditional LCL filter parameter design. And then a
parameter robustness analysis method based on
four-dimensional graphics is proposed to analyze parameter robustness of the presented filter. Compared with the traditional one, the proposed analysis method can analyze the filter performance under variations of several parameters at a time without any iteration. The comparative analysis and discussion considering the LCL filter, the trap filter, and the LCL-LC filter,
are presented and verified through the experiments on a 5kW grid-connected converter prototype. Experiment results demonstrate the accuracy of theoretical analysis and prove the presented filter has a better performance than two others.

2015

51.

MATLAB2015_51

3D microtransformers for DC-DC on-chip
power conversion

We address the miniaturization of power converters by introducing novel, 3D micro transformers with magnetic core for low-MHz frequency applications. The core is fabricated by lamination and micro structuring of Metglas® 2714A magnetic alloy. The solenoids of the micro transformers are wound around the core using a ball-wedge wire bonder. The wire bonding process is fast, allowing the fabrication of solenoids with up to 40 turns in 10 s. The fabricated devices yield the high inductance per unit volume of 2.95 µH/mm3 and energy per unit volume of 133 nJ/mm3 at the frequency of 1 MHz. The power efficiency of 64-76% are measured for different turns ratio with coupling factors as high as 98%.

2015

52.

MATLAB2015_52

Indirect Matrix Converter-Based Topology
and Modulation Schemes for Enhancing
Input Reactive Power Capability

A new topology based on indirect matrix converter
(IMC) is proposed to enhance the input reactive power capability. This topology consists of a conventional IMC and an auxiliary switching network (ASN), which is connected to the dc-link of the IMC in parallel. With the aid of ASN, an implicit current source converter-based static synchronous compensator can be embedded
into an IMC, which lays a foundation for the input reactive power control. Based on the proposed topology, two modulation schemes are presented, and the formations of the output voltage and input reactive current are decoupled in both of them. To minimize
power loss and improve input current quality, a double closed-loop control algorithm is introduced, in which the current through the dc inductor in ASN is controlled to be minimum. Different from the conventional IMC, the input reactive power of the topology is independent of its load condition without considering the practical constraints. The effectiveness of the proposed topology and modulation scheme is confirmed by experimental results.

2015

53.

MATLAB2015_53

Closed Loop Discontinuous Modulation Technique
for Capacitor Voltage Ripples and Switching Losses
Reduction in Modular Multilevel Converters

In this paper, a new discontinuous modulation
technique is presented for the operation of the modular multilevel converter (MMC). The modulation technique is based on adding a zero-sequence to the original modulation signals so that the MMC arms are clamped to the upper or lower terminals of the dc-link bus. The clamping intervals are controlled according to
the absolute value of the output current to minimize the switching losses of the MMC. A significant reduction in the capacitor voltage ripples is achieved, especially when operating with low modulation indices. Furthermore, a circulating current control strategy suitable for this modulation technique is also proposed. Simulation and experimental results under various operating points are reported along with evaluation and comparison results
against a conventional carrier-based pulse-width modulation method.

2015

54.

MATLAB2015_54

Decentralized Inverse-Droop Control for
Input-Series-Output-Parallel DC-DC Converters

Input-series-output-parallel (ISOP) DC-DC converters are suited for high input-voltage and low output-voltage applications. This letter presents a decentralized inverse-droop control for this configuration. Each module is self-contained and no central controller is needed, thus improving the system modularity, reliability and flexibility. With the proposed inverse-droop control,
the output voltage reference rises as the load becomes
heavy. Even though the input voltages is not used in the
inverse-droop loop, the power sharing including input
voltage sharing (IVS) and output current sharing (OCS)
can still be well achieved. Besides, the output voltage
regulation characteristic is not affected by the variation
of input voltage. The operation principle is introduced,
and stability of the strategy is also revealed based on
small signal modeling. Finally, the experiment is
conducted to verify the effectiveness of the control
strategy.

2015

55.

MATLAB2015_55

Detailed Analysis of DC-Link Virtual Impedance
based Suppression Method for Harmonics Interaction
in High-Power PWM Current-Source Motor Drives

For high-power PWM current-source motor drive
systems, due to the low converter switching frequency and the relative small dc choke for reduced cost/weight, the converters’ switching harmonics may interact through dc link and produce inter harmonics in the entire system. Such harmonics interaction phenomenon may give rise to the system resonance at certain motor speeds, which degrades the grid-side power quality and generates excessive torque ripples on the motor side. The resonance caused by the harmonics interaction in high-power PWM
current-source motor drives is investigated in previous work. In addition, to actively suppress such resonance, the basic idea of a dc-link virtual impedance based suppression method has also been proposed. This paper extends the previous work to thoroughly analyze the mechanism and realization of resonance suppression by the dc-link virtual impedance based method. The indepth analysis shows that the dc-link virtual impedance based
method successfully enables the active inter harmonics compensation capability of high-power PWM current-source drives, which is not addressed in previous researches. Moreover, simulations and experiments demonstrate that, by following the selection of coefficient in the suppression method discussed in this paper, the
dc-link virtual impedance based method can effectively enhance the attenuation effect of dc link in high-power PWM current source drive systems so as to suppress the resonance due to the harmonics interaction under all resonance conditions.

2015

56.

MATLAB2015_56

An Online Frequency-Domain Junction Temperature Estimation
Method for IGBT Modules

This letter proposes a new frequency-domain
thermal model for online junction temperature estimation of insulated-gate bipolar transistor (IGBT) modules. The proposed model characterizes the thermal behavior of an IGBT module by a linear time-invariant (LTI) system, whose frequency response is obtained by applying the fast Fourier transform (FFT) to the time derivative of the transient thermal impedance from junction to a
reference position of the IGBT module. The junction temperature of the IGBT is then estimated using the frequency responses of the LTI system and the heat sources of the IGBT module. Simulation results show that the proposed method is computationally efficient
for an accurate online junction temperature estimation of IGBT modules in both steady-state and transient loading conditions.

2015

57.

MATLAB2015_57

Characterization of a Silicon IGBT and Silicon
Carbide MOSFET Cross Switch Hybrid

A parallel arrangement of a Silicon (Si) IGBT and a
Silicon Carbide (SiC) MOSFET is experimentally demonstrated. The concept referred to as the Cross Switch “XS” hybrid aims to reach optimum power device performance by providing low static and dynamic losses while improving the overall electrical and thermal properties due to the combination of both the bipolar Si
IGBT and unipolar SiC MOSFET characteristics. For the purpose of demonstrating the XS hybrid, the parallel configuration was implemented experimentally in a single package for devices rated at 1200V. Test results were obtained to validate this approach with respect to the static and dynamic performance when compared to
a full Si IGBT and a full SiC MOSFET reference devices having the same power ratings as for the XS hybrid samples.

2015

58.

MATLAB2015_58

LCL Filter Design and Inductor Current Ripple Analysis for 3-
level NPC Grid Interface Converter

The harmonic filter for a 3-level neutral point
clamped (NPC) grid interface converter is designed in this paper with good filtering performance and small component size. LCL topology is selected because of the attenuation and size tradeoff. The design of the inverter side inductor L1 is emphasized due to its cost. A detailed inductor current ripple analysis is given based on the space vector modulation (SVM). The analysis derives the inductor volt-second and the maximum current
ripple equation in line cycle. It also reveals the switching cycle current ripple distribution over a line cycle, with the
consideration of power factor. The total system loss is calculated with different ripple current. Inductor L1 is determined by the loss and size tradeoff. Also the capacitor and grid side inductor L2 is designed based on attenuation requirement. Different damping circuits for LCL filter are compared and investigated in detail. The filter design is verified by both simulation and a 200kVA 3-level NPC converter hardware.

2015

59.

MATLAB2015_59

Virtual RC Damping of LCL-Filtered Voltage
Source Converters with Extended Selective
Harmonic Compensation

Active damping and harmonic compensation are
two common challenges faced by LCL-filtered voltage source converters. To manage them holistically, this paper begins by proposing a virtual RC damper in parallel with the passive filter capacitor. The virtual damper is actively inserted by feeding back the passive capacitor current through a high-pass filter, which indirectly, furnishes two superior features. They are the
mitigation of phase lag experienced by a conventional damper and the avoidance of instability caused by the negative resistance inserted unintentionally. Moreover, with the virtual RC damper, the frequency region, within which the harmonic compensation is effective, can be extended beyond the gain crossover frequency. This is of interest to some high-performance applications, but has
presently not been achieved by existing schemes. Performance of the proposed scheme has been tested in the laboratory with results obtained for demonstrating stability and harmonic compensation.

2015

60.

MATLAB2015_60

Versatile Control of Unidirectional AC-DC
Boost Converters for Power Quality Mitigation

This paper introduces a versatile control scheme for
unidirectional ac-dc boost converters for the purpose of
mitigating grid power quality. Since most power factor correction circuits available in the commercial market utilize unidirectional ac-dc boost converter topologies, this is an almost no-cost solution for compensating harmonic current and reactive power in residential applications. Harmonic current and reactive power
compensation methods in the unidirectional ac-dc boost converter are investigated. The additional focus of this paper is to quantify the input current distortions by the unidirectional ac-dc boost converter used for supplying not only active power to the load but also reactive power. Due to input current distortions, the amount of reactive power injected from an individual converter to the grid
should be restricted. Experimental results are presented to
validate the effectiveness of the proposed control method.

2015

61.

MATLAB2015_61

Aalborg Inverter — A new type of “Buck in
Buck, Boost in Boost” Grid-tied Inverter

This paper presents a new family of high
efficiency DC/AC grid-tied inverter with a wide
variation of input DC voltage. It is a “Boost in Boost,
Buck in Buck” inverter, meaning that only one power
stage works at high frequency in order to achieve
minimum switching loss. The minimum voltage drop of
the filtering inductor in the power loop is achieved to
reduce the conduction power loss in both “Boost” and
“Buck” mode. The principle of operation is
demonstrated through the analysis on the equivalent
circuits of a “half-bridge” single-phase inverter. The
theoretical analysis shows that when input DC voltage
is larger than the magnitude of the AC voltage, it is a
Voltage Source Inverter (VSI), and on the contrary it is
Current Source Inverter (CSI) in the other mode. A
220 V/50 Hz/ 2000 W prototype has been constructed.
Simulations and experiments show it has a good control
and system performance.

2015

62.

MATLAB2015_62

Grid-connected Forward Micro-inverter with Primary-Parallel Secondary-Series
Transformer

This paper presents a primary-parallel secondaryseries multicore forward micro-inverter for photovoltaic ACmodule application. The presented micro-inverter operates with a constant off-time boundary mode control, providing MPPT capability and unity power factor. The proposed multi transformer solution allows using low-profile unitary turns ratio transformers. Therefore, the transformers are better coupled and the overall performance of the micro-inverter is improved. Due to the multiphase solution the number of devices increases but, the current stress and losses per device
are reduced contributing to an easier thermal management. Furthermore, the decoupling capacitor is split among the phases, contributing to a low-profile solution without electrolytic capacitors suitable to be mounted in the frame of a PV module. The proposed solution is compared to the classical parallel interleaved approach, showing better efficiency in a wide power range and improving the weighted efficiency.

2015

63.

MATLAB2015_63

A Single-Stage PhotoVoltaic System for a DualInverter fed Open-End Winding Induction Motor
Drive for Pumping Applications

This paper presents an integrated solution for
PhotoVoltaic (PV) fed water-pump drive system, which uses an Open-End Winding Induction Motor (OEWIM). The dualinverter fed OEWIM drive achieves the functionality of a threelevel inverter and requires low value DC bus voltage. This helps in an optimal arrangement of PV modules, which could avoid large strings and helps in improving the PV performance with
wide band-width of operating voltage. It also reduces the voltage rating of the DC-link capacitors and switching devices used in the system. The proposed control strategy achieves an integration of both Maximum Power Point Tracking (MPPT) and V/f control for the efficient utilization of the PV panels and the motor. The
proposed control scheme requires the sensing of PV voltage and current only. Thus, the system requires less number of sensors. All the analytical, simulation and experimental results of this work under different environmental conditions are presented in this paper.

2015

SN

PROJECT CODE

PROJECT TOPIC

YEAR

1

MAT1425

Topic: Diagnosis of diabetic retinopathy by employing image processing technique to detect exudates in retinal images

Abstract: Diabetic retinopathy (DR) is a micro vascular complication of long-term diabetes and it is the major cause of visual impairment because of changes in blood vessels of the retina. Major vision loss because of DR is highly preventable with regular screening and timely intervention at the earlier stages. The presence of exudates is one of the primitive signs of DR and the detection of these exudates is the first step in automated screening for DR. Hence, exudates detection becomes a significant diagnostic task, in which digital retinal imaging plays a vital role. In this study, the authors propose an algorithm to detect the presence of exudates automatically and this helps the ophthalmologists in the diagnosis and follow-up of DR. Exudates are normally detected by their high grey-level variations and they have used an artificial neural network to perform this task by applying colour, size, shape and texture as the features. The performance of the authors algorithm has been prospectively tested by using DIARETDB1 database and evaluated by comparing the results with the ground-truth images annotated by expert ophthalmologists. They have obtained illustrative results of mean sensitivity 96.3%, mean specificity 99.8%, using lesion-based evaluation criterion and achieved a classification accuracy of 99.7%.

2014

2

MAT1424

Topic: Data Hiding in Encrypted H.264/AVC Video Streams by Codeword Substitution

Abstract: Digital video sometimes needs to be stored and processed in an encrypted format to maintain security and privacy. For the purpose of content notation and/or tampering detection, it is necessary to perform data hiding in these encrypted videos. In this way, data hiding in encrypted domain without decryption preserves the confidentiality of the content In addition, it is more efficient without decryption followed by data hiding and re-encryption. In this paper, a novel scheme of data hiding directly in the encrypted version of H.264/AVC video stream is proposed, which includes the following three parts, i.e., H.264/AVC video encryption, data embedding, and data extraction. By analyzing the property of H.264/AVC codec, the code words of intra prediction modes, the code words of motion vector differences, and the code words of residual coefficients are encrypted with stream ciphers. Then, a data hider may embed additional data in the encrypted domain by using codeword substitution technique, without knowing the original video content. In order to adapt to different application scenarios, data extraction can be done either in the encrypted domain or in the decrypted domain. Furthermore, video file size is strictly preserved even after encryption and data embedding. Experimental results have demonstrated the feasibility and efficiency of the proposed scheme.

2014

3

MAT1423

Topic: Edge Detection Method for Image Processing based on Generalized Type-2 Fuzzy Logic

Abstract: This  paper  presents  an  edge  detection  method based  on  the  morphological  gradient  technique  and  generalized type-2  fuzzy  logic.  The  theory  of  alpha  planes  is  used  to implement generalized type-2 fuzzy logic  for edge detection. For the  defuzzification  process,  the  heights  and  approximation methods  are  used.  Simulation  results  with  a  type-1  fuzzy inference  system  (T1FIS),  an  interval  type-2  fuzzy  inference system  (IT2FIS)  and  with  a  generalized  type-2  fuzzy  inference system (GT2FIS) for edge detection are presented. The proposed generalized  type-2  fuzzy  edge  detection  method  was  tested  with benchmark  images  and  synthetic  images.  We  used  the  merit  of Pratt  measure  to  illustrate  the  advantages  of  using  generalized type-2 fuzzy logic.

2014

4

MAT1422

Topic: Deblurred images post-processing by Poisson warping

Abstract: In this work we develop a post-processing algorithm which enhances the results of the existing image deblurring methods. It performs additional edge sharpening using grid warping. The idea of the proposed algorithm is to transform the neighborhood of the edge so that the neighboring pixels move closer to the edge, and then resample the image from the warped grid to the original uniform grid. The proposed technique preserves image textures while making the edges sharper. The effectiveness of the method is shown for basic deblurring methods on LIVE database images with added blur and noise.

2014

5

MAT1421

Topic: Image Contrast Enhancement Using Color and Depth Histograms

Abstract: In this letter, we propose a new global contrast enhancement algorithm using the histograms of color and depth images. On the basis of the histogram-modification framework, the color and depth image histograms arefirst partitioned into subintervals using the Gaussian mixture model. The positions partitioning the color histogram are then adjusted such that spatially neighboring pixels with the similar intensity and depth values can be grouped into the same sub-interval. By estimating the mapping curve of the contrast enhancement for each sub-interval, the global image contrast can be improved without over-enhancing the local image contrast. Experimental results demonstrate the effectiveness

of the proposed algorithm.

2014

6

MAT1420

Topic: Object Tracking Based on Active Contour Modeling

Abstract: Object Tracking based on Active Contour Modeling is an image processing based technology that uses snapshots of the object under consideration to track it via robot in the real world. The objective has been to implement a unique methodology that employs the pursuing and adapting of contour to the current state of image, and hence track the object. The system can be implemented in drone planes wherein this algorithm can be used to guide the movement of the gun based on the movements of the object, or, in robot games with a slightly more advanced robot. Initially Image Processing is performed to reduce operation complexity and achieve swift real-time performance. A set of contour-based modeling algorithms is then implemented to ‘actively’ track the subject. Also, relative

transformation calculations are made to lock the target via robot, continuously. MATLAB is used to simulate and implement the system and it is tested on field with a ball placed on it and a robot tracking the ball. The experiments prove that the system successfully detects and tracks the object efficiently in the real world for all horizontal and vertical transitions.

2014

7

MAT1419

Topic: Vision Based Data Extraction of Vehicles in Traffic

Abstract: With the rise in traffic related crimes the need for an efficient automated surveillance system has become of utmost importance. This paper proposes a system to monitor video from traffic cameras and process it in real time for storing essential information of the vehicles in traffic. Histogram of Oriented Gradients (HOG) of extracted frames is used as features for classification (vehicle frame and non vehicle frame). The classifier is designed based on Support Vector Machine (SVM) . The subtracted image acquired from a dynamically updated background image is used to extract the vehicle image for recognition using trained Artificial Neural Network(ANN). The system is designed to store details like vehicle make, model, color and time of passing the camera in a database (Microsoft Access (MS Access)). Finally the stored details are made available through a Graphical User Interface(GUI) designed using Visual Basic(VB) that will provide an user with the options of selecting a time window to look for the vehicles that have passed within that interval or to enter a car model to check if it has passed that point at any time. The system is modeled in MATLAB and tested in a real time environment in one of the busiest road in Kamrup district of Assam and provides satisfactory performance.

2014

8

MAT1418

Topic: Digital Right Management Control for Joint Ownership of Digital Images using Biometric Features

Abstract: This paper proposes a method to establish joint ownership of digital images by embedding imperceptible digital pattern in the image. This digital pattern is generated from biometric features of more than one subject in a strategic matter so that the identification of individual subject can be done and the multiple ownership of the digital images can be established. This digital pattern was embedded and extracted from the image and the experiments were also carried out when the image was subjected to signal processing attacks. Coefficients of mid frequency band discrete cosine transform was used for

embedding as these coefficients do not adversely affect the perceptual transparency and is also significantly robust to normal signal processing attacks. Experimental results indicate that the insertion of this digital pattern does not change the perceptual properties of the image and the pattern survives

signal processing attacks which can be extracted for unique identification.

2014

9

MAT1417

Topic: Intelligent Water Metering System: An Image Processing Approach (MATLAB simulations)

Abstract: The scarcity and misuse of fresh water pose a serious and growing threat to sustainable development. The population growth, severe droughts and uneven distribution of water resources are the reasons for water scarcity, and this scarcity will only continue to grow more severe. The technical

sophistication of meters for measuring water flows has increased noticeably in recent decades in order to improve management of water. This paper proposes simple image processing approach for an intelligent metering system. The proposed system uses simple image processing algorithms and DSP processor, capable of executing MIPS; which makes whole system respond faster. As meter image is being captured from set distance, meter mask generation reduces the need of algorithms for detection and segmentation of meter reading. The proposed system improves the efficiency of drinking water management and reduces power consumption as image sensor is activated as per predefined billing cycle.

2014

10

MAT1416

Topic: Fingerprint Recognition Using Gabor Filter

Abstract: Fingerprint recognition is the most popular methods used for identification with higher degree of success. The fingerprint has unique characteristics called minutiae, which are points where a curve track finishes, intersect or branches off. In this work a method for Fingerprint recognition is considered using a combination of Fast Fourier Transform (FFT) and Gabor Filters for enhancing the image. The proposed method involves combination of Gabor filter and Frequency domain filtering for enhancing the fingerprint. With eight different orientations of Gabor filter, features of the fingerprint

extracting are combined. In Frequency domain filtering, the fingerprint image is subdivided into 32*32 small frames. Features are extracted from these frames in frequency domain. Final enhanced fingerprint is obtained with the results of Gabor filter and frequency domain filtering. Binarization and Thinning follows next where the enhanced fingerprint is converted into binary and the ridges are thinned to one pixel width. This helps in extracting the Minutiae parts (ridge bifurcation and ridge endings). The overall recognition rate for the proposed method is 95% which is much better than histogram method where the recognition rate is 64%. This project is implemented in MATLAB.

2014

11

MAT1415

Topic: ARIMA Model based Breast Cancer Detection and Classification through Image Processing

Abstract: Computer Aided Diagnosis (CAD) has changed the way of medical diagnostics. As similar to other walk of diagnostics field, CAD is having high potential in breast cancer prognosis because of its highest accuracy. CAD may play a very important role in developing countries i.e. EIT-MEM (Electrical Impedance Tomography –Multi-frequency Electrical Impedance Mammography) device being used for breast cancer defection. MEM-EIT produces tomography based mammograms which are

considered most reliable method of early detection of breast cancer. Cancer diagnostic expert all over the world find this noninvasive technique very accurate as it is one dimensional representation of images in terms of temperature however the accuracy is limited and investigator fail to take into account the spatial co-ordination between the pixels which is crucial in cancerous tumour detection and their classification (cancerous or normal) in EIT (Electrical Impedance Tomography) - based mammogram images. In this study, we are trying to focus an algorithms based CAD (Computer Aided Diagnosis) model for tumour detection and classification. We model it by ARIMA model (autoregressive integrated moving average (ARIMA) model) and parameter estimation will be performed using leassquare method. Our system classifies the tumour into three categories- (i) healthy tissue (ii) benign tissue (iii) cancerous tissue along with above three segments the performance analysis

between 2D image and 1D image will be done for better accuracy and sensitivity detection.

2014

12

MAT1414

Topic: Human Hand Image Analysis Extracting Finger Coordinates and Axial Vectors

Abstract: This paper presents a finger cut-off algorithm for accurate calculation of fingertip coordinates based on hand contours. It provides not only information on exact fingertip position but also orientation and lengths of all fingers in the image. Algorithm can be used for development of user interfaces based on human gesture analysis, such as Touch Table, multimodal gesture based user interface developed by the author. Advantages of proposed algorithm over fingertip detection algorithm originally used in Touch Table are described.

2014

13

MAT1413

Topic: Automatic Brain Tumor Detection and Segmentation in MR Images

Abstract: The MRI or CT scan images are primary follow up diagnostic tools when a neurologic exam indicates a possibility of a primary or metastatic brain tumor existence. The tumor tissue mainly appears in brighter colors than the rest of the regions in the brain. Based on this observation, an automated algorithm for brain tumor detection and medical doctors’ assistance in facilitated and accelerated diagnosis procedure has been developed and initially tested on images obtained from the patients with diagnosed tumors and healthy subjects.

2014

14

MAT1412

Topic: RGB ratios based skin detection

Abstract: Many  different  applications  like  face/people detection,  image  content  interpretation,  de-identification  for privacy  protection  in  multimedia  content,  etc.  requires  skin detection as a pre-processing step. There is no a perfect solution for  skin  detection,  since  this  process  is  a  compromise  on speed, simplicity  and  precision  (detection  quality).  There  are  many different  techniques  for  skin  detection  modeling  ranging  from simple  models  based  on  one  or  several  thresholds  to advanced models  based  on  neural  network,  Bayesian classifier,  maximum entropy,  k-means  clustering,  etc.  This  paper  proposes  a  simple model, based on ratios of red, green and blue components of the RGB color model. It describes how to make a compromise in a skin  detection  modeling  by  using  three  levels  of  rules.  Data analysis  that  supports  conclusions  is  performed  on  the  dataset from Universidad de Chile (UChile, dbskin2  –complete set) that contains 103 images and their annotations.

2014

15

MAT1411

Topic: Embedding of Sound Clips as a Watermark in StilI Images using Discrete Wavelet Transform

Abstract: Embedding uj’sndkr im+ys  in Iarger images ming the  oppmach of  watermarking is  being efecfively iised ,for image scvutinv.  Wirh che advent of digital image processing; secure addition of wutwmah in digitized  images  ming varivirs  techtiiqzies  has evolved, The me of wavelet transform for the said pz~ipose  has pw ved wry usefit/. This puper presenis a preliminary research carried out to embed audio clips in  still images. The technique  uses  audio puperties aiidfirral disrortiun tfrreshold in the furget image us parameters-for decision moking,fiw various aspects of the  iinplemenfed scheme. Some of these decisions ure selection oJ  either grav scale or color images,  decomposition  level  for  the  wavelet tmnsfbrni,  chanvlel selection,  sound sample  and synrhwis of [he sound sample into minsamples. The research i.y  being exfended ,fbr embedding of  audio samples in image sequences for video  transmissions jbr .secwe artdio commzrnication applicalions

2014

16

MAT1410

Topic: Automatic brain  tumor  detection  and  segmentation  for MRI  using  covariance  and  geodesic  distance

Abstract: In  this  paper,  we present  a  new  approach  that  allows the  detection  and  segmentation  of  brain  tumors  automatically. The  approach  is  based  on  covariance  and  geodesic  distance.  The

detection  of  central  coordinates  of  abnormal  tissues  is  based  on the  covariance  method.  These  coordinates  are  used  to  segment the  brain  tumor  area  using  geodesic  distance  for  Tl  and  T2

weighted  magnetic  resonance  images  (MRI).  The  ultimate objective  is  to  retrieve  the  attributes  of  the  tumor  observed  on the  image  to  use  them  in  the  step  of  segmentation  and classification.  The  present  methods  are  tested  on  images  of  Tl and  T2  weighted  MR  and  have  shown  a  better  performance  in the  analysis  of  biomedical images.

2014

17

MAT1409

Topic: ANALYSIS OF RETINAL BLOOD VESSELS USING IMAGE PROCESSING TECHNIQUES

Abstract: Assessment of blood vessels in human eye allows earlier detection of eye diseases such as glaucoma and diabetic retinopathy. Digital image processing techniques play a vital role in retinal blood vessel detection , Several image processing methods and filters are in practise to detect and extract the attributes of retinal blood vessels such as length ,width, pattern and angles. Automated Digital image processing techniques and methods has to undergo more of improvisation to achieve precise accuracy to study the condition of Retinal Vessels especially in cases of Glaucoma and retinopathy; we have explained various Templates based matched filters, Thresholding Methods, Segmentation methods, and functional approaches to isolate the blood vessels.

2014

18

MAT1408

Topic: Automatic Optic Disc Detection in Digital Fundus Images Using Image Processing

Abstract: Optic disc (OD) is an important part of the eye. OD detection  is  an  important  step  in  developing  systems  for automated  diagnosis  of  various  serious  ophthalmic  diseases like  Diabetic  retinopathy,  Glaucoma,  hypertension  etc.  The variation of intensity within the optic disc and intensity close to the  optic  disc  boundary  are  the  major  hurdle  in  automated optic  disc  detection.  General  edge  detection  algorithms  are frequently  unsuccessful  to  segment  the  optic  disc  because  of this. Complexity increases due to the presence of blood vessels. This  paper  presents  simple  method  for  OD  segmentation  by using  techniques  like  principal  component  analysis  (PCA), mathematical  morphology  and  Watershed  Transform.  PCA used  for  good  presentation  of  input  image  and  mathematical morphology  is  used  to  remove  blood  vessels  from  image. Watershed Transform is used for boundary segmentation.

2014

19

MAT1407

Topic: A  Comparative  Analysis  of  Edge  and  Color  Based Segmentation  for  Orange  Fruit  Recognition

Abstract: In  this  paper,  we  presented  two  segmentation methods.  Edge  based  and  color  based  detection  methods  were used  to  segment  images  of  orange  fruits  obtained  under  natural lighting  conditions.  Twenty digitized  images  of  orange  fruits  were randomly  selected  from  the  Internet  in  order  to  find  an  orange  in each  image  and  to  determine  its  location.  We  compared  the results  of  both  segmentation  results  and  the  color  based segmentation  outperforms  the  edge  based  segmentation  in  all aspects.  The  MATLAB  image  processing  toolbox  is  used  for  the computation  and  comparison  results  are  shown  in  the  segmented image  results.

2014

20

MAT1406

Topic: Detection  of Leukemia  in  Microscopic  Images  Using Image  Processing

Abstract: Leukemia occurs  when lot  of  abnormal  white blood  cells produced  by  the  bone  marrow.  Hematologist  makes  use  of microscopic  study  of  human  blood,  which  leads  to  need  of

methods,  including  microscopic  color  imaging,  segmentation, classification  and  clustering  that  can  allow  identification  of patients  suffering  from  Leukemia.  The  microscopic  images  will be  inspected  visually  by  hematologists  and  the  process  is  time consuming  and  tiring.  The  automatic  image  processing  system  is urgently  needed  and  can  overcome  related  constraints  in  visual inspection.The  proposed  system  will be  on  microscopic images  to detect  Leukemia.  The  early  and  fast  identification  of  Leukemia greatly  aids  in  providing  the  appropriate  treatment.  Initial

segmentation  is  done  using  Statistical  parameters  such  as  mean, standard  deviation  which  segregates  white blood  cells  from  other blood  components  i.e.  erythrocytes  and  platelets.  Geometrical features  such  as  area,  perimeter  of  the  white  blood  cell  nucleusis investigated  for  diagnostic  prediction  of  Leukemia. The  proposed method  is  successfully  applied  to  a  large  number  of  images, showing  promising  results  for  varying  image  quality. Different image  processing  algorithms  such  as  Image  Enhancement, Thresholding,  Mathematical  morphology  and  Labelling  are implemented  using LabVIEW and  MATLAB.

2014

21

MAT1405

Topic: Lung Cancer Diagnosis Using CT-Scan Images Based on Cellular Learning Automata

Abstract: Lung cancer has killed many people in recent years. Early diagnosis of lung cancer can help doctors to treat patients and keep them alive. The most common way to detect lung cancer is using the Computed Tomography (CT) image. The systems that are created by the integration of computers and medical science are called Computer Aided Diagnosis (CAD). A CAD system that is adopted for the diagnosis lung cancer, uses lung CT images as input and based on an algorithm helps doctors to perform an image analysis. With the help of CAD, doctors can make the final decision. This paper is a study concerning automatic detection of lung cancer by using cellular learning automata. Images include some unwanted data and some feature that are important for processing; pre-processing improves images by removing distortion and enhance the important features. This system used lung CT scan so we applied some pre-processing method such as Gabor filter and region growing to improve CT

images. After pre-processing step according features the lung cancer nodule was extracted. The obtained image through previous steps was entered to cellular learning automata lattice for training and making them possess the ability to detect lung cancer. The obtained results show, the proposed approach can reduce the error rate.

2014

22

MAT1404

Topic: Image Processing Based Vehicle Detection and Tracking Method

Abstract: Vehicle detection and tracking plays an effective and significant role in the area of traffic surveillance system where efficient traffic management and safety is the main concern. In this paper, we discuss and address the issue of detecting vehicle / traffic data from video frames. Although various researches have been done in this area and many methods have been implemented, still this area has room for improvements. With a view to do improvements, it is proposed to develop an unique algorithm for vehicle data recognition and tracking using Gaussian mixture model and blob detection methods. First, we differentiate the foreground from background in frames by learning the background. Here, foreground detector detects the object and a binary computation is done to define rectangular regions around every detected object. To detect the moving object correctly and to remove the noise some morphological operations have been applied. Then the final counting is done by tracking the detected objects and their regions. The results are encouraging and we got more than 91% of average accuracy in detection and tracking using the Gaussian Mixture Model and Blob Detection methods.

2014

23

MAT1403

Topic: Image Encryption Based On Diffusion Process And Multiple Chaotic Maps

Abstract: In the modern world, security is a prime important issue and encryption is one of the preeminent way to ensure security. There are many image encryption schemes. Each one of them has its own strength and weakness. This project presents a novel algorithm for the image encryption and decryption scheme. The project provides a secured image encryption technique using multiple chaotic based circular mapping. In this, first, a pair of sub keys is given by using chaotic logistic maps. Second, the image is encrypted using logistic map sub key and its transformation leads to diffusion process. Third, sub keys are generated by four different chaotic maps. Based on the initial conditions, each map may produce various random numbers from various orbits of the maps. Among those random numbers,

a particular number are selected as a key for the encryption algorithm. Based on the key, a binary sequence is generated to manage the encryption algorithm. The input image of 2-D is transformed into a 1- D array by using raster scanning. It is then divided into various sub blocks. Then the position permutation is applied to each binary matrix based on multiple chaotic maps. Finally the receiver uses the same sub keys to decrypt the encrypted images. Also using the same encryption and decryption algorithm video is encrypted and decrypted. Finally shown that video encryption and decryption takes more time. Histogram analysis, correlation analysis are also done and found that there is no statistical similarity between original and encrypted image. Peak Signal to Noise ratio is also calculated and found that the encrypted image is of higher quality.

2014

24

MAT1402

Topic: Real-time Vehicle Color Identification for Surveillance Videos

Abstract: Vehicles are one of the main detection targets of the traffic and security video surveillance system. In this paper, we propose an automatic vehicle color identification method for vehicle classification. The main idea of the proposed scheme is to divide  a  vehicle  into  a  hierarchical  coarse-to-fine  structure  to extract its wheels, windows, main body, and other auto parts. In the proposed method, the main body alone is used by a support vector  machine  (SVM)  for  classification.  Experimental  results show that the proposed scheme is efficient and effective and the proposed  vehicle  color  identification  is  suitable  for  real-time surveillance applications.

2014

25

MAT1401

Topic: Intelligent Water Metering System: An Image Processing Approach (MATLAB simulations)

Abstract : The scarcity and misuse of fresh water pose a serious and growing threat to sustainable development. The population growth, severe droughts and uneven distribution of water resources are the reasons for water scarcity, and this scarcity will only continue to grow more severe. The technical

sophistication of meters for measuring water flows has increased noticeably in recent decades in order to improve management of water. This paper proposes simple image processing approach for an intelligent metering system. The proposed system uses simple image processing algorithms and DSP processor, capable of executing MIPS; which makes whole system respond faster. As meter image is being captured from set distance, meter mask generation reduces the need of algorithms for detection and segmentation of meter reading. The proposed system improves the efficiency of drinking water management and reduces power consumption as image sensor is activated as per predefined billing cycle.

2014