Chengzhi Deng

Curvelet Domain Watermark Detection Using Alpha-Stable Models

This paper address issues that arise in copyright protection systems of digital images, which employ blind watermark verification structures in the curvelet domain. First, we observe that statistical distribution with heavy algebraic tails, such as the alpha-stable family, are in many cases more accurate modeling tools for the curvelet coefficients than families with exponential tails such as generalized Gaussian. Motivated by our modeling results, we then design a new processor for blind watermark detection using the Cauchy member of the alpha-stable family. We analyze the performance of the new detector in terms of the associated probabilities of detection and false alarm and we compare it to the performance of the generalized Gaussian detector and the traditional correlation-based detector by performance experiments. The experiments prove that Cauchy detector is superior to the others.

Approximate Sparsity and Nonlocal Total Variation Based Compressive MR Image Reconstruction

Recent developments in compressive sensing (CS) show that it is possible to accurately reconstruct the magnetic resonance (MR) image from undersampled -space data by solving nonsmooth convex optimization problems, which therefore significantly reduce the scanning time. In this paper, we propose a new MR image reconstruction method based on a compound regularization model associated with the nonlocal total variation (NLTV) and the wavelet approximate sparsity. Nonlocal total variation can restore periodic textures and local geometric information better than total variation. The wavelet approximate sparsity achieves more accurate sparse reconstruction than fixed wavelet and norm. Furthermore, a variable splitting and augmented Lagrangian algorithm is presented to solve the proposed minimization problem. Experimental results on MR image reconstruction demonstrate that the proposed method outperforms many existing MR image reconstruction methods both in quantitative and in visual quality assessment.

Convex hull for visual tracking with EMD

Developing an effective target appearance model is a challenging task due to the influence of factors such as partial occlusion, illumination variations, fast motion, etc. Existing appearance models usually utilize the tracking results from previous frames as target templates upon which the target appearance model is built by linear combinations of the templates. With such kind of representation, visual tracking is not robust when drastic appearance variations occur. We propose a simple but effective tracking algorithm with a novel appearance model in a particle filter framework. A target candidate is represented by the convex combination of a set of target templates. Additionally, the distance between a target candidate and the templates is measured using the EMD. Experimental results on challenging video sequences against state-of-the-art algorithms demonstrate the robustness and effectiveness of the proposed tracking algorithm.

Approximate Sparse Regularized Hyperspectral Unmixing

Sparse regression based unmixing has been recently proposed to estimate the abundance of materials present in hyperspectral image pixel. In this paper, a novel sparse unmixing optimization model based on approximate sparsity, namely, approximate sparse unmixing (ASU), is firstly proposed to perform the unmixing task for hyperspectral remote sensing imagery. And then, a variable splitting and augmented Lagrangian algorithm is introduced to tackle the optimization problem. In ASU, approximate sparsity is used as a regularizer for sparse unmixing, which is sparser than regularizer and much easier to be solved than regularizer. Three simulated and one real hyperspectral images were used to evaluate the performance of the proposed algorithm in comparison to regularizer. Experimental results demonstrate that the proposed algorithm is more effective and accurate for hyperspectral unmixing than state-of-the-art regularizer.

Robust Visual Tracking with Discrimination Dictionary Learning

It is a challenging issue to deal with kinds of appearance variations in visual tracking. Existing tracking algorithms build appearance models upon target templates. Those models are not robust to significant appearance variations due to factors such as illumination variations, partial occlusions, and scale variation. In this paper, we propose a robust tracking algorithm with a learnt dictionary to represent target candidates. With the learnt dictionary, a target candidate is represented with a linear combination of dictionary atoms. The discriminative information in learning samples is exploited. In the meantime, the learning processing of dictionaries can learn appearance variations. Based on the learnt dictionary, we can get a more stable representation for target candidates. Additionally, the observation likelihood is evaluated based on both the reconstruct error and dictionary coefficients with constraint. Comprehensive experiments demonstrate the superiority of the proposed tracking algorithm to some state-of-the-art tracking algorithms.

Robust Visual Tracking Based on Convex Hull with EMD-L1

Factors such as drastic illumination variations, partial occlusion, rotation make robust visual tracking a difficult problem. Some tracking algorithms represent a target appearances based on obtained tracking results from previous frames with a linear combination of target templates. This kind of target representation is not robust to drastic appearance variations. In this paper, we propose a simple and effective tracking algorithm with a novel appearance model. A target candidate is represented by convex combinations of target templates. Measuring the similarity between a target candidate and the target templates is a key problem for a robust likelihood evaluation. The distance between a target candidate and the templates is measured using the earth mover’s distance with L1 ground distance. Comprehensive experiments demonstrate the robustness and effectiveness of the proposed tracking algorithm against state-of-the-art tracking algorithms.

Hyperspectral unmixing based on weighted abundance vector regularization

Sparse unmixing is based on the assumption that each mixed pixel in the hyperspectral image can be expressed in the form of linear combination of a number of pure spectral signatures that are known in advance. Despite the success of sparse unmixing based on the L0 or L1 regularizer, the limitation of these approach focuses on analyzing the hyperspectral data without incorporating spatial structure information of hyperspectral data. In this paper, considering the correlation between the abundance coefficients of neighboring pixels we proposed a weighted abundance vector sparse unmixing model for the hyperspectral unmixing, named Wav-SU model. Based on this model, we use L1 norm and L1/2 norm as the regularizer and adapt the variable splitting and augmented Lagrangian algorithm to solve them. Our experimental results with both simulated and real hyperspectral data sets demonstrate that the proposed Wav-SU method is an effective and simple spectral unmixing algorithm for hyperspectral unmixing.

Hyperspectral unmixing based on iteratively re-weighted smoothed l0 regularisation

Hyperspectral unmixing HU plays a fundamental role in a wide range of hyperspectral applications. Sparse-based approach has recently received much attention in hyperspectral unmixing area. Sparse unmixing is based on the assumption that each mixed pixel in the hyperspectral image can be expressed in the form of linear combination of a number of pure spectral signatures that are known in advance. Despite the success of sparse unmixing based on the L

Sparse Affine Hull for Visual Tracking

It is a challenging task to develop a robust appearance model due to various factors such as partial occlusion, fast motion, background clutters and illumination variations. In this paper, we propose a novel target representation for visual tracking. Namely, a target candidate is represented by sparse affine combinations of dictionary templates in a particle filter framework. Affine combinations based target appearances can cover unknown appearances. In order to adapt the dynamic scenes across a video sequence, the dictionary templates are updated in the tracking process. Experimental results on several challenging video sequences against some state-of-the-art tracking algorithms demonstrate that the proposed algorithm is robust to illumination variations, background clutters, etc.

Redundant Ridgelet Transform and its Application to Image Processing

One of the problems encountered in image transmission is the cut-off or error of a bits chain at the moment of transmission. To protect the transmitted signal, it is necessary to couple quantized transform coding with a redundant transform. Ridgelet transform is a new directional resolution transform and it is more suitable for describing the signals with line or super-plane singularities. Finite ridgelet transform is a discrete orthonormal version of ridgelet transform proposed by Minh N.Do and Martin Vetterli. In this paper, we propose a new conception of redundant ridgelet transform, which we implement it mainly by controlling the ratio of the redundancy at the step of radon transform in the ridgelet transform. We first briefly introduce the concept of ridgelet transform. Then, we illustrate finite ridgelet transform and the new method. Finally, the main features of the redundant ridgelet transform and its applications for image coding and image analysis are discussed.