You are on page 1of 15

Elysium Technologies Private Limited

ISO 9001:2008 A leading Research and Development Division

Madurai | Chennai | Kollam | Ramnad | Tuticorin | Singapore

Abstract Digital Image Processing 2010 - 2011

01 A Detection Model for Driver’s Unsafe States Based on Real-Time Face-Vision

The auto industry has developed fast in the last 100 years. Although it brings us convenience, more and more traffic
accidents are happening every day. There are many factors which can cause a car accident. Based on the record of a
large number of accidents, fatigue is one of the most important factors. Additionally, driver’s distraction and
conversations with passengers during driving can lead to serious results. In this paper, a real-time vision-based model
is proposed for monitoring driver’s unsafe states, including fatigue state, distraction state and talking state, etc. By
analyzing driver’s real-time face vision, a method for detecting driver’s fatigue, distraction and talking states is given.
Also, the model is its extreme high speed and very simple equipment. It can run at about 20 frames per second in
video with 640*480 resolutions on a normal computer platform.

02 A Perceptually Relevant Approach to Ringing Region Detection

An efficient approach toward a no-reference ringing metric intrinsically exists of two steps: first detecting regions in an
image where ringing might occur, and second quantifying the ringing annoyance in these regions. This paper presents
a novel approach toward the first step: the automatic detection of regions visually impaired by ringing artifacts in
compressed images. It is a no-reference approach, taking into account the specific physical structure of ringing
artifacts combined with properties of the human visual system (HVS). To maintain low complexity for real-time
applications, the proposed approach adopts a perceptually relevant edge detector to capture regions in the image
susceptible to ringing, and a simple yet efficient model of visual masking to determine ringing visibility. The approach
is validated with the results of a psychovisual experiment, and its performance is compared to existing alternatives in
literature for ringing region detection. Experimental results show that our method is promising in terms of both
reliability and computational efficiency

03 A Pedestrian Classification Method Based on Transfer Learning

Pedestrian detection is a challenging research task of computer vision, which can be seen as a classification problem
in the sliding window framework. Many supervised learning based methods require a large number of labeled data for
training. However, training and testing data are not independent identically distributed in most cases, due to the
complex background, and it is expensive to re-collect and label the data. This paper proposes a semi-supervised
method for pedestrian classification, which is based on transfer learning and sparse coding and just requires a small
quantity of labeled data. Firstly, we use sparse coding to learn a slightly higher-level, more succinct feature
representation from the unlabeled data that randomly downloaded from the Internet. Then we apply this representation
to the target classification problem by transfer learning. The quantitative experiment results demonstrate that this
method can improve the performance of pedestrian classification and just needs only a few labeled data

#230, Church Road, Anna Nagar, Madurai 625 020, Tamil Nadu, India
(: +91 452-4390702, 4392702, 4390651
Elysium Technologies Private Limited
ISO 9001:2008 A leading Research and Development Division
Madurai | Chennai | Kollam | Ramnad | Tuticorin | Singapore

04 A Novel Image Classification Method Based on Manifold Learning and Gaussian Mixture Model

Image classification is one of the important parts of digital image processing. We propose a novel feature space-based
image classification method by combining manifold learning and mixture model. In this paper, the process of image
classification can be viewed as two parts: a coarse-grained classification and a fine-grained classification. In the
coarse-grained classification, we apply the ISOMAP (Isometric Mapping) algorithm to do a dimensional reduction
based on manifold learning. Thus, solving the classification problem is transformed from a high dimensional data
space to a low-dimensional feature space. And then, during the fine-grained classification, we present an improved EM
algorithm of finite Gaussian mixture model to do clustering. Experimental results have demonstrated that the proposed
method performs well in both accuracy and time. Additionally, our algorithm is robust to some extent

A New Image Fusion Algorithm Based on Wavelet Transform and the Second Generation Curve let
05 Transform

This paper analyzes the characteristics of the Second Generation Curve let Transform and put forward an image fusion
algorithm based on Wavelet Transform and the Second Generation Curve let Transform. We looked at the selection
principles about low and high frequency coefficients according to different frequency domain after Wavelet and the
Second Generation Curve let Transform. In choosing the low-frequency coefficients, the concept of local area variance
was chosen to measuring criteria. In choosing the high frequency coefficients, the window property and local
characteristics of pixels were analyzed. Finally, the proposed algorithm in this article was applied to experiments of
multi focus image fusion and complementary image fusion. According to simulation results, the proposed algorithms
hold useful information from source multiple images quite well.

06 A New Feature Descriptor and Selection Method to Space Image Registration

In order to extract the flat feature stars, which are used to implement the space image registration, a feature descriptor
based on WLD (Weber Local Descriptor) and feature dynamic selection method is proposed in this paper. Firstly, the
feature stars are extracted and described by excitation and orientation using the WLD transform. Then we utilize a
dynamic feature selection algorithm (DFSA), which is used in registration method selection for registering image with
varying scene content, to choose the best match features for the registration. Many space images are used to test the
method proposed in this paper and the results show that this method has higher performance compared to traditional

#230, Church Road, Anna Nagar, Madurai 625 020, Tamil Nadu, India
(: +91 452-4390702, 4392702, 4390651
Elysium Technologies Private Limited
ISO 9001:2008 A leading Research and Development Division
Madurai | Chennai | Kollam | Ramnad | Tuticorin | Singapore

07 An Adaptive Threshold for the Canny Operator of Edge Detection

The thresholds play an important role in the Canny Operator which used in the image edge detection. Many self-
adaptive threshold algorithms have been proposed to improve the performance of Canny Operator. The Otsu method
is one of the most popular improvements. However the Otsu method cannot automatically set the low threshold
according to the different image intensity adaptively. In order to overcome this defect, an adaptive threshold algorithm
for the Canny Operator was proposed which calculated the low threshold adaptively based on a probability model.
Experiments show that this method produces better edge detection results both on objective and subjective
evaluations than the Otsu method.

Adaptive Non separable Wavelet Transform via Lifting and its Application to Content-Based Image
08 Retrieval

We present in this paper a novel way to adapt a multidimensional wavelet filter bank, based on the non separable
lifting scheme framework, to any specific problem. It allows the design of filter banks with a desired number of degrees
of freedom, while controlling the number of vanishing moments of the primal wavelet ( ?? moments) and of the dual
wavelet ( moments). The prediction and update filters, in the lifting scheme based filter banks, are defined as Neville
filters of order ?? and , respectively. However, in order to introduce some degrees of freedom in the design, these
filters are not defined as the simplest Neville filters. The proposed method is convenient: the same algorithm is used
whatever the dimensionality of the signal, and whatever the lattice used. The method is applied to content-based image
retrieval (CBIR): an image signature is derived from this new adaptive non separable wavelet transform. The method is
evaluated on four image databases and compared to a similar CBIR system, based on an adaptive separable wavelet
transform. The mean precision at five of the non separable wavelet based system is notably higher on three out of the
four databases, and comparable on the other one. The proposed method also compares favorably with the dual-tree
complex wavelet transform, an over complete non separable wavelet transform.

09 Accurate and Efficient Method for Smoothly Space-Variant Gaussian Blurring

This paper presents a computationally efficient algorithm for smoothly space-variant Gaussian blurring of images. The
proposed algorithm uses a specialized filter bank with optimal filters computed through principal component analysis.
This filter bank approximates perfect space-variant Gaussian blurring to arbitrarily high accuracy and at greatly
reduced computational cost compared to the brute force approach of employing a separate low-pass filter at each
image location. This is particularly important for spatially variant image processing such as foveated coding.
Experimental results show that the proposed algorithm provides typically 10 to 15 dB better approximation of perfect
Gaussian blurring than the blended Gaussian Pyramid blurring approach when using a bank of just eight filters.

#230, Church Road, Anna Nagar, Madurai 625 020, Tamil Nadu, India
(: +91 452-4390702, 4392702, 4390651
Elysium Technologies Private Limited
ISO 9001:2008 A leading Research and Development Division
Madurai | Chennai | Kollam | Ramnad | Tuticorin | Singapore

A Salt & Pepper Noise Fast Filtering Algorithm for Grayscale Images Based on Neighborhood Correlation
10 Detection

A salt & pepper noise fast filtering algorithm for grayscale images based on neighborhood correlation detection is
presented. By utilizing a 4 × 4 pixel template, the algorithm can discriminate and filter various patterns of salt & pepper
noise spots or blocks within 2 × 2 pixel size range. In contrast with many kinds of median filtering algorithm, which
may cause image blurring, it has much higher edge preserving ability. Furthermore, this algorithm is able to
synchronously reflect image quality via amount, location and density statistics of salt & pepper noise spots and make
good sense to guide parameter selection for imaging systems.

11 A Robust Fuzzy Local Information C-Means Clustering Algorithm

This paper presents a variation of fuzzy c-means (FCM) algorithm that provides image clustering. The proposed
algorithm incorporates the local spatial information and gray level information in a novel fuzzy way. The new algorithm
is called fuzzy local information C-Means (FLICM). FLICM can overcome the disadvantages of the known fuzzy c-means
algorithms and at the same time enhances the clustering performance. The major characteristic of FLICM is the use of
a fuzzy local (both spatial and gray level) similarity measure, aiming to guarantee noise insensitiveness and image
detail preservation. Furthermore, the proposed algorithm is fully free of the empirically adjusted parameters ( , , , etc.)
incorporated into all other fuzzy c-means algorithms proposed in the literature. Experiments performed on synthetic
and real-world images show that FLICM algorithm is effective and efficient, providing robustness to noisy images

12 A Review of Segmentation Method for MR Image

Magnetic resonance (MR) is a typical medical imaging technique. It can provide high resolution 3D image with
anatomical and function information through analyzing MRI sequence, which facilitates and improves diagnosis and
patient treatment. The first important step in image analysis is image segmentation. In this paper, numerous methods
that have been developed for segmentation in MRI are reviewed. We study these segmentation strategies and perform
a qualitative discussion according to 3 categories, i.e. traditional image processing method, statistical-based
segmentation method and partition technique with bias field estimation.

13 Fusing Local Patterns of Gabor Magnitude and Phase for Face Recognition

Gabor features have been known to be effective for face recognition. However, only a few approaches utilize phase
feature and they usually perform worse than those using magnitude feature. To investigate the potential of Gabor
phase and its fusion with magnitude for face recognition, in this paper, we first propose local Gabor XOR patterns
(LGXP), which encodes the Gabor phase by using the local XOR pattern (LXP) operator. Then, we introduce block-

#230, Church Road, Anna Nagar, Madurai 625 020, Tamil Nadu, India
(: +91 452-4390702, 4392702, 4390651
Elysium Technologies Private Limited
ISO 9001:2008 A leading Research and Development Division
Madurai | Chennai | Kollam | Ramnad | Tuticorin | Singapore

based Fisher’s linear discriminate (BFLD) to reduce the dimensionality of the proposed descriptor and at the same
time enhance its discriminative power. Finally, by using BFLD, we fuse local patterns of Gabor magnitude and phase
for face recognition. We evaluate our approach on FERET and FRGC 2.0 databases. In particular, we perform
comparative experimental studies of different local Gabor patterns. We also make a detailed comparison of their
combinations with BFLD, as well as the fusion of different descriptors by using BFLD. Extensive experimental results
verify the effectiveness of our LGXP descriptor and also show that our fusion approach outperforms most of the state-
of-the-art approaches

14 Feature Extraction and Soft Segmentation of Texture Images

In this paper, we propose an efficient method for texture image segmentation. First we extract four feature channels
smoothed with the total variation (TV) flow. Then we propose a soft segmentation model based on the Chan- Vese
model by adding a weight in the arc length term and using a soft membership function instead of level set function to
represent the region. We derive a fast algorithm using the Additive Operator Scheme (AOS) and Chambolle's fast dual
projection method. Experimental results on texture and Synthetic Aperture Radar (SAR) images show the effectiveness
of our algorithm.

15 Feature Extraction of Hyper spectral Images Based on Preserving Neighborhood Discriminate Embedding

A novel manifold learning feature extraction approach—preserving neighborhood discriminate embedding (PNDE) of
hyper spectral image is proposed in this paper. The local geometrical and discriminate structure of the data manifold
can be accurately characterized by within-class neighboring graph and between-class neighboring graph. Unlike
manifold learning, such as LLE, Iso map and LE, which cannot deal with new test samples and images larger than
70×70, the method here can process full scene hyper spectral images. Experiments results on hyper spectral datasets
and real-word datasets show that the proposed method can efficiently reduce the dimensionality while maintaining
high classification accuracy. In addition, only a small amount of training samples are needed.

16 Fast Image Segmentation Based on Two dimensional Minimum Tsallis-Cross Entropy

Image segmentation based on 2-D (two dimensional) histogram is an effective method because the structure
information is taken into account in image. However, it always is on the assumption that partial region of 2-D
histogram equals to zero, while utilizing Shannon entropy as optimization function. As a result, the efficiency of image
segmentation is degraded seriously. In this paper, we proposed a fast thresholding segmentation based on two-
dimensional minimum Tsallis-cross entropy and PSO, which utilizes minimum Tsallis-cross entropy as optimization
function which is non-extensive and calculates optimal threshold in improved gray level-gradient histogram which
cancels previous hypothesis that partial region value equals to zero in histogram. At the same time, the improved 2-D
histogram is clustered before searching optimal threshold value to shorten the time. Experiment results show that the
proposed algorithm achieves a better segmentation quality and computation efficiency.

#230, Church Road, Anna Nagar, Madurai 625 020, Tamil Nadu, India
(: +91 452-4390702, 4392702, 4390651
Elysium Technologies Private Limited
ISO 9001:2008 A leading Research and Development Division
Madurai | Chennai | Kollam | Ramnad | Tuticorin | Singapore

17 Face Recognition Using a Color Tensor Framework

: Facial images are affected by multiple factors including facial geometries, expressions, viewpoints and illuminations.
We apply multi-linear algebra to separate these factors and extract the people factor used for face recognition.
Compared to standard PCA and its variants, the method allows for bigger changes in viewpoints andilluminations. Our
method is based on the method, adding color information, using (2D)2-PCA for dimensionality reduction and centering
face recognition in a color sensor framework. Good results are obtained by doing experiments on Weizman and
Facepix database.

Edge-based Text Localization and Character Segmentation Algorithms for Automatic Slab Information
18 Recognition

Content-based image indexing is to label image based on its content such as color, texture, shape, face, text, and etc.
Because the text can be easily localized and recognized compared to other image contents, many researchers have
studied about the general text localization in images actively. Generally, it is known that localizing scene text is more
difficult than localizing caption text, and still it is not easy. However, although our target texts are scene texts, it is
clear that they are not general, that is, the properties of target texts are fixed. In this paper, we propose an edge-based
text localization and segmentation algorithms for automatic slab information recognition system

19 Contour let-Based Image Watermarking Using Optimum Detector in a Noisy Environment

In this paper, an improved multiplicative image watermarking system is presented. Since human visual system is less
sensitive to the image edges, watermarking is applied in the contour let domain, which represents image edges
sparsely. In the presented scheme, watermark data is embedded in directional sub band with the highest energy. By
modeling the contour let coefficients with General Gaussian Distribution (GGD), the distribution of watermarked noisy
coefficients is analytically calculated. The tradeoff between the transparency and robustness of the watermark data is
solved in a novel fashion. At the receiver, based on the Maximum Likelihood (ML) decision rule, an optimal detector by
the aid of channel side information is proposed. In the next step, a blind extension of the suggested algorithm is
presented using the patchwork idea. Experimental results confirm the superiority of the proposed method against
common attacks, such as Additive White Gaussian Noise (AWGN), JPEG compression, and rotation attacks, in
comparison with the recently proposed techniques

20 Color Image Enhancement with a Human Visual System Based Adaptive Filter

In this paper, considering the adaptive characteristics of human visual system, a new color image enhancement
algorithm based on human visual system adaptive filter is proposed. The new algorithm is divided into three major
parts: obtain luminance image and background image, adaptive adjustment and color restoration. Unlike traditional
color image enhancement algorithms, the adaptive filter in the algorithm takes color information into consideration.
The algorithm finds the importance of color information in color image enhancement and utilizes color space
conversion to obtain a much better visibility. Experimental results show that the algorithm proposed has better
effectiveness in reducing halo and color distortion

#230, Church Road, Anna Nagar, Madurai 625 020, Tamil Nadu, India
(: +91 452-4390702, 4392702, 4390651
Elysium Technologies Private Limited
ISO 9001:2008 A leading Research and Development Division
Madurai | Chennai | Kollam | Ramnad | Tuticorin | Singapore

21 Blind and Semi-Blind Deblurring of Natural Images

A method for blind image deblurring is presented. The method only makes weak assumptions about the blurring filter and is
able to undo a wide variety of blurring degradations. To overcome the ill-posedness of the blind image deblurring problem,
the method includes a learning technique which initially focuses on the main edges of the image and gradually takes details
into account. A new image prior, which includes a new edge detector, is used. The method is able to handle unconstrained
blurs, but also allows the use of constraints or of prior information on the blurring filter, as well as the use of filters defined in
a parametric manner. Furthermore, it works in both single-frame and multiframe scenarios. The use of constrained blur
models appropriate to the problem at hand, and/or of multiframe scenarios, generally improves the deblurring results. Tests
performed on monochrome and color images, with various synthetic and real-life degradations, without and with noise, in
single-frame and multiframe scenarios, showed good results, both in subjective terms and in terms of the increase of signal
to noise ratio (ISNR) measure. In comparisons with other state of the art methods, our method yields better results, and
shows to be applicable to a much wider range of blurs.

22 Automatic Detection of Vocal Folds from High-Speed Imaging

An automatic image processing method is proposed to extract vocal fold vibrations from high-speed digital imaging.
Extracted regular vocal fold vibrations contribute to periodic voices, and irregular vocal fold vibrations contribute to
periodic voices. The effects of image resolutions are investigated. This study combines the features of both image and
voice signal processing and provides a valuable automatic image processing technique for the measurement and
quantification of voice productions from high-speed imaging.

23 An Improved DFT-SOFDM Scheme Based on Inter-Carrier Interference Cancellation

In this paper, a novel and efficient improved spectrum allocation scheme of DFT-SOFDM in a frequency selective
channel is presented. The proposed scheme exploits data conjugate mapping on the basis of inter-carrier interference
(ICI) cancellation, which has low complexity but high performance. The illustrative examples over Least Square (LS)
and Linear Minimum Mean-Squared Error (LMMSE) equalization demonstrate that the proposed scheme can achieve
better BER and Peak to Average Power Ratio (PAPR) performance compared to the conventional distributed spectrum
allocation scheme in DFT-SOFDM system

#230, Church Road, Anna Nagar, Madurai 625 020, Tamil Nadu, India
(: +91 452-4390702, 4392702, 4390651
Elysium Technologies Private Limited
ISO 9001:2008 A leading Research and Development Division
Madurai | Chennai | Kollam | Ramnad | Tuticorin | Singapore

24 An Image Segmentation Method Based on Type-2 Fuzzy Gaussian Mixture Models

This paper proposes a new image segmentation method based on Type-2 fuzzy Gaussian Mixture Models (T2 FGMMs).
First, the core-region and the open-region of image are extracted according to spatial information of pixels. Then, the
GMMs parameters are estimated by EM algorithm. The interval in which T2 FGMMs parameters vary is constrained by
the GMMs parameters of the core-region and the open-region of image. Finally, Bayesian decision is used to realize
image segmentation. In the end, the method is compared with image segmentation using Otsu’s method, FCM and
GMM. Experiments demonstrate the effectiveness of this method

25 Individual Facial Expression Transferring using Active Appearance Model and FAP in MPEG-4

This paper presents a facial expression transferring system. The inputs of the system are an expression picture and a
neutral picture of a single frontal face, which are captured by an ordinary video camera. And the output is a 3D
individual face with the same expression. We design and implement the system using FDPs and FAPs in MPEG-4 and
the Active Appearance Models (AAMs) comprehensively. Firstly, Adaboost method is used to detect the face. Then 68
feature points of the face are located with the algorithm based on active appearance models (AAM). With these feature
points and according to the MPEG-4 standard, 2D facial expression feature data are registered to FAPs in MPEG-4 by
image normalization and feature alignment. After that, the FAPs can be used to drive the MPEG-4 compatible models
and conduct 3D face with the corresponding expression

26 Improving Fuzzy C-Means Clustering Based on Local Membership Variation

The fuzzy c-means clustering algorithm has been successfully applied to a wide variety of problems. However, the
image may be corrupted by noise, which leads to inaccuracy with segmentation. In the paper, a local fuzzy clustering
regularization model is introduced in the objective function of the standard fuzzy c-means (FCM) algorithm. It can allow
the membership of a pixel to be influenced by the memberships of its immediate neighborhood. Such schemes are
useful for partition data sets affected by noise. Experimental results on both synthetic images and real image are given
to demonstrate the effectiveness of the proposed algorithm

27 Images De-noising with Mapshrink Estimate and Dual-threshold in Curve let Domain

Map shrink algorithm is widely used in the image processing of wavelet Domain, and Curve let transform is a new
multi-scale transform theory with multi-direction. By applying Map shrink to Curve let Domain and analyzing single
threshold de-noising method existing problem, this paper proposes a de-noising method with dual threshold,
combining global threshold and local threshold, which not only retains the quality of global threshold in keeping the
general gray scale of an image but also preserves the advantage in edge detail maintenance of local self-adaptive
method. Based on the experimental results, this method is much more effective than de-noising method of global
threshold or self-adaptive threshold in Curve let Domain

#230, Church Road, Anna Nagar, Madurai 625 020, Tamil Nadu, India
(: +91 452-4390702, 4392702, 4390651
Elysium Technologies Private Limited
ISO 9001:2008 A leading Research and Development Division
Madurai | Chennai | Kollam | Ramnad | Tuticorin | Singapore

28 Image Reconstruction Using Particle Filters and Multiple Hypotheses Testing

In this paper, we introduce a reconstruction framework that explicitly accounts for image geometry when defining the
spatial interaction between pixels in the filtering process. To this end, image structure is captured using local co-
occurrence statistics and is incorporated to the enhancement algorithm in a sequential fashion using the particle
filtering technique. In this context, the reconstruction process is modeled using a dynamical system with multiple
states and its evolution is guided by the prior density describing the image structure. Towards optimal exploration of
the image geometry, an evaluation process of the state of the system is performed at each iteration. The resulting
framework explores optimally spatial dependencies between image content towards variable bandwidth image
reconstruction. Promising results using additive noise models demonstrate the potentials of such an explicit modeling
of the geometry

29 Image Denoising using Contour let and Two-Dimensional Principle Component Analysis

This paper proposes a novel image Denoising algorithm using the Contour let transform and the two dimensional
Principle Component Analysis (2DPCA). The noise image can be decomposed by the Contour let into directional sub
bands. The 2DPCA is then carried out to estimate the threshold for the image blocks in high frequency sub bands. The
soft thresholding shrinkage can hence be employed on the Contour let coefficients without estimating the noise
variance. The Denoising algorithm is validated by numerical experiments on two images. Numerical results show that
the proposed method can obtain higher PSNR than former methods.

30 Hyper spectral Image Classification Using Wavelet Packet Analysis and Gray Prediction Model

The main focus of hyper spectral image classification is the ability to extract information from a pixel’s hyper spectral
curve. In this paper, we propose a new classification method based on wavelet packet analysis and gray prediction
model of hyper spectral reflectance curves. The wavelet packet analysis is used for feature extraction, while the gray
prediction model is applied for dimensionality reduction. The efficiency of the proposed method will be estimated by
the multivariate statistical analysis (i.e. Mahalanobis distance and quantile). Experimental results indicate that our
algorithm has a relatively high efficiency, and classification accuracy of 99.3%.

31 High- Accuracy Sub-Pixel Motion Estimation From Noisy Images in Fourier Domain

In this paper, we propose a new method for estimating sub pixel motion via exploiting the principle of phase
correlation in the Fourier domain. The method is based on linear weighting of the height of the main peak on the one
hand and the difference between its two neighboring side peaks on the other. Using both synthetic and real data we
show that the proposed method outperforms many established approaches and achieves improved accuracy even in
the presence of noisy samples

#230, Church Road, Anna Nagar, Madurai 625 020, Tamil Nadu, India
(: +91 452-4390702, 4392702, 4390651
Elysium Technologies Private Limited
ISO 9001:2008 A leading Research and Development Division
Madurai | Chennai | Kollam | Ramnad | Tuticorin | Singapore

32 Watermarking of Free-view Video

With the advances in image based rendering (IBR) in recent years, generation of a realistic arbitrary view of a scene
from a number of original views has become cheaper and faster. One of the main applications of this progress has
emerged as free-view TV(FTV), where TV-viewers select freely the viewing position and angle via IBR on the
transmitted multi view video. Noting that the TV-viewer might record a personal video for this arbitrarily selected view
and misuse this content, it is apparent that copyright and copy protection problems also exist and should be solved
for FTV. In this paper, we focus on this newly emerged problem by proposing a watermarking method for free-view
video. The watermark is embedded into every frame of multiple views by exploiting the spatial masking properties of
the human visual system. Assuming that the position and rotation of the virtual camera is known, the proposed
method extracts the watermark successfully from an arbitrarily generated virtual image. In order to extend the method
for the case of an unknown virtual camera position and rotation, the transformations on the watermark pattern due to
image based rendering operations are analyzed. Based upon this analysis, camera position and homography
estimation methods are proposed for the virtual camera. The encouraging simulation results promise not only a novel
method, but also a new direction for watermarking research.

33 Ultrasound Despeckling for Contrast Enhancement

Images produced by ultrasound systems are adversely hampered by a stochastic process known as speckle. A
despeckling method based upon removing outlier is proposed. The method is developed to contrast enhance B-mode
ultrasound images. The contrast enhancement is with respect to decreasing pixel variations in homogeneous regions
while maintaining or improving differences in mean values of distinct regions. A comparison of the proposed
despeckling filter is compared with the other well known despeckling filters. The evaluations of despeckling
performance are based upon improvements to contrast enhancement, structural similarity, and segmentation results
on a Field II simulated image and actual B-mode cardiac ultrasound images captured in vivo.

34 Texture Classification Using Refined Histogram

In this correspondence, we propose a novel, efficient, and effective Refined Histogram (RH) for modeling the wavelet
sub band detail coefficients and present a new image signature based on the RH model for supervised texture
classification. Our RH makes use of a step function with exponentially increasing intervals to model the histogram of
detail coefficients, and the concatenation of the RH model parameters for all wavelet sub bands forms the so-called RH
signature. To justify the usefulness of the RH signature, we discuss and investigate some of its statistical properties.
These properties would clarify the sufficiency of the signature to characterize the wavelet sub band information. In
addition, we shall also present an efficient RH signature extraction algorithm based on the coefficient-counting
technique, which helps to speed up the overall classification system performance. We apply the RH signature to

#230, Church Road, Anna Nagar, Madurai 625 020, Tamil Nadu, India
(: +91 452-4390702, 4392702, 4390651
Elysium Technologies Private Limited
ISO 9001:2008 A leading Research and Development Division
Madurai | Chennai | Kollam | Ramnad | Tuticorin | Singapore

texture classification using the well-known databases. Experimental results show that our proposed RH signature in
conjunction with the use of symmetrized Kullback–Leibler divergence gives a satisfactory classification performance
compared with the current state-of-the-art methods.

35 Texture Classification by Modeling Joint Distributions of Local Patterns with Gaussian Mixtures

Texture classification generally requires the analysis of patterns in local pixel neighborhoods. Statistically, the
underlying processes are comprehensively described by their joint probability density functions (jPDFs). Even for
small neighborhoods, however, stable estimation of jPDFs by joint histograms (jHSTs) is often infeasible, since the
number of entries in the jHST exceeds by far the number of pixels in a typical texture region. Moreover, evaluation of
distance functions between jHSTs is often computationally prohibitive. Practically, the number of entries in a jHST is
therefore reduced by considering only two-pixel patterns, leading to 2D-jHSTs known as co occurrence matrices, or by
quantization of the gray levels in local patterns to only two gray levels, yielding local binary patterns (LBPs). Both
approaches result in a loss of information. We introduce here a framework for supervised texture classification which
reduces or avoids this information loss. Local texture neighborhoods are first filtered by a filter bank. Without further
quantization, the jPDF of the filter responses is then described parametrically by Gaussian mixture models (GMMs). We
show that the parameters of the GMMs can be reliably estimated from small image regions. Moreover, distances
between the thus modelled jPDFs of different texture patterns can be computed efficiently in closed form from their
model parameters.We furthermore extend this texture descriptor to achieve full invariance to rotation. We evaluate the
framework for different filter banks on the Brodatz texture set. We first show that combining the LBP difference filters
with the GMM-based density estimator outperforms the classical LBP approach and its codebook extensions. When
replacing these—rather elementary— difference filters by the wavelet frame transform (WFT), the performance of
exceeds the one obtained more recently by spin image and RIFT descriptors by Lazebnik et al

#230, Church Road, Anna Nagar, Madurai 625 020, Tamil Nadu, India
(: +91 452-4390702, 4392702, 4390651
Elysium Technologies Private Limited
ISO 9001:2008 A leading Research and Development Division
Madurai | Chennai | Kollam | Ramnad | Tuticorin | Singapore

36 Sparse Bayesian Learning of Filters for Efficient Image Expansion

We propose a framework for expanding a given image using an interpolator that is trained in advance with training
data, based on sparse Bayesian estimation for determining the optimal and compact support for efficient image
expansion. Experiments on test data show that learned interpolators are compact yet superior to classical ones

37 Sonar Image Denoising Based on HMT Model in Morphological Wavelet Domain

Sonar images are susceptible to noise pollution that results in low contrast. And sonar image Denoising technology is
the key for subsequent target recognition. In this paper, an image denoising algorithm using wavelet transform was
studied. Firstly, we constructed a morphological mean wavelet for gray image processing. Then the noisy sonar image
was trained by the Hidden Markov Tree model in the morphological wavelet domain. According to the characteristics of
the morphological mean wavelet, we classified multiresolution analysis of the noisy image in different directions, and
removed noise according to the training result with Bayesian estimation. Finally, a desired denoising effect could be
obtained by computing the average of different reconstructed images. Computer experiments show that our denoising
algorithm can remove Gaussian noise of sonar image effectively. Compared with some classical wavelet denoising
methods, image details are retained better.

38 Research on Image Denoising New Method Based on Curvelet Transform

Wavelet transform only has spot singularity but curvelet transform has advantage of multi-scale singularity in each
direction. After analyzing the method of image denoising by curvelet transform based on USFFT, the paper puts
forward a new method WCSC, which uses wrapping and cycle-spinning combined curvelet transform to improve
existing algorithm. According to the simulated experiment, the results of the denoised pictures show that the improved
algorithm has reduced the Pseudo-Gibbs phenomena, can reserve more detail information, texture of images and get
higher seeing impression and PSNR value under the noise standard deviation

39 Region Merging Techniques Using Information Theory Statistical Measures

The purpose of the current work is to propose, under a statistical framework, a family of unsupervised region merging
techniques providing a set of the most relevant region-based explanations of an image at different levels of analysis.
These techniques are characterized by general and nonparametric region models, with neither color nor texture
homogeneity assumptions, and a set of innovative merging criteria, based on information theory statistical measures.
The scale consistency of the partitions is assured through i) a size regularization term into the merging criteria and a
classical merging order, or ii) using a novel scale-based merging order to avoid the region size homogeneity imposed
by the use of a size regularization term. Moreover, a partition significance index is defined to automatically determine
the subset of most representative partitions from the created hierarchy. Most significant automatically extracted
partitions show the ability to represent the semantic content of the image from a human point of view. Finally, a
complete and exhaustive evaluation of the proposed techniques is performed, using not only different databases for

#230, Church Road, Anna Nagar, Madurai 625 020, Tamil Nadu, India
(: +91 452-4390702, 4392702, 4390651
Elysium Technologies Private Limited
ISO 9001:2008 A leading Research and Development Division
Madurai | Chennai | Kollam | Ramnad | Tuticorin | Singapore

the two main addressed problems (object-oriented segmentation of generic images and texture image segmentation),
but also specific evaluation features in each case: under- and over segmentation error, and a large set of region-based,
pixel-based and error consistency indicators, respectively. Results are promising, outperforming in most indicators
both object-oriented and texture state-of-the-art segmentation techniques

40 Real-Time Tracking Algorithm Based on Improved Mean Shift and Kalman Filter

In traditional Mean Shift algorithm, color histogram is usually used as the features vectors, and the dissimilarity
between the referenced targets and the target candidates is expressed by the metric derived from the Bhattacharyya
coefficients. The traditional Mean Shift procedure is used to find the real position of the target by looking for the
regional minimum of the distance function iteratively. While the target’s color is similar to the background, the
algorithm will miss the target. This paper presents a new mean shift algorithm based on spatial edge orientation
histograms, using space distribution and texture information as matching information. Meanwhile a Kalman filter will
be used to predict the target’s position. Experimental results demonstrate that the proposed algorithm can deal with
intricate conditions, such as significant clutter, partial occlusions, and it can track objects efficiently and robustly

41 Perceptual Image Hashing Based on Virtual Watermark Detection

This paper proposes a new robust and secure perceptual image hashing technique based on virtual watermark
detection. The idea is justified by the fact that the watermark detector responds similarly to perceptually close images
using a non embedded watermark. The hash values are extracted in binary form with a perfect control over the
probability distribution of the hash bits. Moreover, a key is used to generate pseudo-random noise whose real values
contribute to the randomness of the feature vector with a significantly increased uncertainty of the adversary,
measured by mutual information, in comparison with linear correlation. Experimentally, the proposed technique has
been shown to outperform related state-of-the art techniques recently proposed in the literature in terms of robustness
with respect to image processing manipulations and geometric attacks.

42 Patch-Based Background Initialization in Heavily Cluttered Video

In this paper, we propose a patch-based technique for robust background initialization that exploits both spatial and
temporal consistency of the static background. The proposed technique is able to cope with heavy clutter, i.e,
foreground objects that stand still for a considerable portion of time. First, the sequence is subdivided in patches that
are clustered along the time-line in order to narrow down the number of background candidates. Then, a tessellation is
grown incrementally by selecting at each step the best continuation of the current background. The method rests on
sound principles in all its stages and only few, intelligible parameters are needed. Experimental results show that the
proposed algorithm is effective and compares favorably with existing techniques.

43 Optimized Block-Based Connected Components Labeling With Decision Trees

#230, Church Road, Anna Nagar, Madurai 625 020, Tamil Nadu, India
(: +91 452-4390702, 4392702, 4390651
Elysium Technologies Private Limited
ISO 9001:2008 A leading Research and Development Division
Madurai | Chennai | Kollam | Ramnad | Tuticorin | Singapore

In this paper, we define a new paradigm for eight-connection labeling, which employees a general approach to improve
neighborhood exploration and minimizes the number of memory accesses. First, we exploit and extend the decision
table formalism introducing OR-decision tables, in which multiple alternative actions are managed. An automatic
procedure to synthesize the optimal decision tree from the decision table is used, providing the most effective
conditions evaluation order. Second, we propose a new scanning technique that moves on a 2 2 pixel grid over the
image, which is optimized by the automatically generated decision tree. An extensive comparison with the state of art
approaches is proposed, both on synthetic and real datasets. The synthetic dataset is composed of different sizes and
densities random images, while the real datasets are an artistic image analysis dataset, a document analysis dataset
for text detection and recognition, and finally a standard resolution dataset for picture segmentation tasks. The
algorithm provides an impressive speedup over the state of the art algorithms.

44 New Learning Based Super-Resolution: Use of DWT and IGMRF Prior

In this paper, we propose a new learning-based approach for super-resolving an image captured at low spatial
resolution. Given the low spatial resolution test image and a database consisting of low and high spatial resolution
images, we obtain super-resolution for the test image. We first obtain an initial high-resolution (HR) estimate by
learning the high-frequency details from the available database. A new discrete wavelet transform (DWT) based
approach is proposed for learning that uses a set of low-resolution (LR) images and their corresponding HR versions.
Since the super-resolution is an ill-posed problem, we obtain the final solution using a regularization framework. The
LR image is modeled as the aliased and noisy version of the corresponding HR image, and the aliasing matrix entries
are estimated using the test image and the initial HR estimate. The prior model for the super-resolved image is chosen
as an Inhomogeneous Gaussian Markov random field (IGMRF) and the model parameters are estimated using the same
initial HR estimate. A maximum a posteriori (MAP) estimation is used to arrive at the cost function which is minimized
using a simple gradient descent approach. We demonstrate the effectiveness of the proposed approach by conducting
the experiments on gray scale as well as on color images. The method is compared with the standard interpolation
technique and also with existing learning-based approaches. The proposed approach can be used in applications such
as wildlife sensor networks, remote surveillance where the memory, the transmission bandwidth, and the camera cost
are the main constraints.

45 Multiple Descriptions Based on Multi rate Coding for JPEG 2000 and H.264/AVC

: Multiple descriptions coding (MDC) makes use of redundant representations of multimedia data to achieve resiliency.
Descriptions should be generated so that the quality obtained when decoding a subset of them only depends on their
number and not on the particular received subset. In this paper, we propose a method based on the principle of
encoding the source at several rates, and properly blending the data encoded at different rates to generate the
descriptions. The aim is to achieve efficient redundancy exploitation, and easy adaptation to different network
scenarios by means of fine tuning of the encoder parameters. We apply this principle to both JPEG 2000 images and
H.264/AVC video data. We consider as the reference scenario the distribution of contents on application-layer overlays
with multiple-tree topology. The experimental results reveal that our method favorably compares with state-of-art MDC

#230, Church Road, Anna Nagar, Madurai 625 020, Tamil Nadu, India
(: +91 452-4390702, 4392702, 4390651
Elysium Technologies Private Limited
ISO 9001:2008 A leading Research and Development Division
Madurai | Chennai | Kollam | Ramnad | Tuticorin | Singapore

46 Multiphase Joint Segmentation-Registration and Object Tracking for Layered Images

In this paper we propose to jointly segment and register objects of interest in layered images. Layered imaging refers
to imageries taken from different perspectives and possibly by different sensors. Registration and segmentation are
therefore the two main tasks which contriute to the bottom level, data alignment, of the multi sensor data fusion
hierarchical structures. Most exploitations of two layered images assumed that scanners are at very high altitudes and
that only one transformation ties the two images. Our data are however taken at mid-range and therefore requires
segmentation to assist us examining different object regions in a divide-and-conquer fashion. Our approach is a
combination of multiphase active contour method with a joint segmentation-registration technique (which we called
MPJSR) carried out in a local moving window prior to a global optimization. To further address layered video
sequences and tracking objects in frames, we propose a simple adaptation of optical flow calculations along the active
contours in a pair of layered image sequences. The experimental results show that the whole integrated algorithm is
able to delineate the objects of interest, align them for a pair of layered frames and keep track of the objects over time

47 Marker-Based Watershed Segmentation Embedded with Edge Information

This paper proposes an edge embedded marker based watershed algorithm for high spatial resolution remote sensing
image segmentation. Two improvement techniques are proposed for the two key steps of maker extraction and pixel
labeling respectively to make it more effective and efficient for high spatial resolution image segmentation. Moreover,
the edge information, detected by the edge detector embedded with confidence, is used to direct the two key steps for
detecting objects with weak boundary and improving the positional accuracy of the objects boundary. Experiments on
Quick bird image show that the proposed method performs well both in retaining the weak boundary and reducing the
undesired over-segmentation

#230, Church Road, Anna Nagar, Madurai 625 020, Tamil Nadu, India
(: +91 452-4390702, 4392702, 4390651