Chuanbo Chen

Huazhong University of Science and Technology, Wu-han-shih, Hubei, China

Are you Chuanbo Chen?

Claim your profile

Publications (47)19.12 Total impact

  • Jun Shang · Chuanbo Chen · Xiaobing Pei · Hu Liang · He Tang · Mudar Sarem
    [Show abstract] [Hide abstract]
    ABSTRACT: Designing efficient and effective keypoint descriptors for an image plays a vital role in many computer vision tasks. The traditional binary descriptors such as local binary pattern and its variants directly perform a binarization operation on the intensity differences of the local affine covariant regions, thus their performance usually drops a lot because of the limited distinctiveness. In this paper, we propose a novel image keypoint descriptor, namely local derivative quantized binary pattern for object recognition. To incorporate the spatial information, we first divide the local affine covariant region into several subregions according to the intensity orders. For each sub region, we quantize the intensity differences between the central pixels and their neighbors in an adaptive way, and then we order the differences and use a hash function to map the differences into binary codes. The binary codes are histogramed to form the feature of each subregion. Furthermore, we utilize multi-scale support regions and pool the histograms together to represent the features of the image. Our approach does not need prior codebook training and hence it is more flexible and easy to be implemented. Moreover, our descriptor can preserve more local brightness and edge information than the traditional binary descriptors. Also, our descriptor is robust to rotation, illumination variations and other geometric transformations. Finally we conduct extensive experiments on three challenging datasets (i.e., 53 Objects, ZuBuD, and Kentucky) for object recognition and the experimental results show that our descriptor outperforms the existing state-of-the-art descriptors.
    No preview · Article · Nov 2015 · The Visual Computer
  • [Show abstract] [Hide abstract]
    ABSTRACT: Multi-frame super-resolution image reconstruction aims to restore a high-resolution image by fusing a set of low-resolution images. The low-resolution images are usually subject to some degradation, such as warping, blurring, down-sampling, or noising, which causes substantial information loss in the low-resolution images, especially in the texture regions. The missing information is not well estimated using existing traditional methods. In this paper, having analyzed the observation model describing the degradation process starting with a high-resolution image and moving to the low-resolution images, we propose a more reasonable observation model that integrates the missing information into the super-resolution reconstruction. Our approach is fully formulated in a Bayesian framework using the KullbackÔÇôLeibler divergence. In this way, the missing information is estimated simultaneously with the high-resolution image, motion parameters, and hyper-parameters. Our proposed estimation of the missing information improves the quality of the reconstructed image. Experimental results presented in this paper show improved performance compared with that of existing traditional methods.
    No preview · Article · Jul 2015 · Circuits Systems and Signal Processing
  • Jun Shang · Chuanbo Chen · Xiaobing Pei · Hu Liang · He Tang · Mudar Sarem
    [Show abstract] [Hide abstract]
    ABSTRACT: Binary image descriptors have received a lot of attention in recent years, since they provide numerous advantages, such as low memory footprint and efficient matching strategy. However, they utilize intermediate representations and are generally less discriminative than floating-point descriptors. We propose an image region descriptor, namely local derivative ordinal binary pattern, for object recognition and image categorization. In order to preserve more local contrast and edge information, we quantize the intensity differences between the central pixels and their neighbors of the detected local affine covariant regions in an adaptive way. These differences are then sorted and mapped into binary codes and histogrammed with a weight of the sum of the absolute value of the differences. Furthermore, the gray level of the central pixel is quantized to further improve the discriminative ability. Finally, we combine them to form a joint histogram to represent the features of the image. We observe that our descriptor preserves more local brightness and edge information than traditional binary descriptors. Also, our descriptor is robust to rotation, illumination variations, and other geometric transformations. We conduct extensive experiments on the standard ETHZ and Kentucky datasets for object recognition and PASCAL for image classification. The experimental results show that our descriptor outperforms existing state-of-the-art methods.
    No preview · Article · May 2015 · Journal of Electronic Imaging
  • Xiaobing Pei · Tao Wu · Chuanbo Chen
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, a novel projective nonnegative matrix factorization (PNMF) method for enhancing the clustering performance is presented, called automated graph regularized projective nonnegative matrix factorization (AGPNMF). The idea of AGPNMF is to extend the original PNMF by incorporating the automated graph regularized constraint into the PNMF decomposition. The key advantage of this approach is that AGPNMF simultaneously finds graph weights matrix and dimensionality reduction of data. AGPNMF seeks to extract the data representation space that preserves the local geometry structure. This character makes AGPNMF more intuitive and more powerful than the original method for clustering tasks. The kernel trick is used to extend AGPNMF model related to the input space by some nonlinear map. The proposed method has been applied to the problem of document clustering using the well-known Reuters-21578, TDT2, and SECTOR data sets. Our experimental evaluations show that the proposed method enhances the performance of PNMF for document clustering.
    No preview · Article · Oct 2014 · Cybernetics, IEEE Transactions on
  • Xiaobing Pei · Zehua Lyu · Changqing Chen · Chuanbo Chen
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, a novel label propagation (LP) method is presented, called the manifold adaptive label propagation (MALP) method, which is to extend original LP by integrating sparse representation constraint into regularization framework of LP method. Similar to most LP, first of all, MALP also finds graph edges from given data and gives weights to the graph edges. Our goal is to find graph weights matrix adaptively. The key advantage of our approach is that MALP simultaneously finds graph weights matrix and predicts the label of unlabeled data. This paper also derives efficient algorithm to solve the proposed problem. Extensions of our MALP in kernel space and robust version are presented. The proposed method has been applied to the problem of semi-supervised face clustering using the well-known ORL, Yale, extended YaleB, and PIE datasets. Our experimental evaluations show the effectiveness of our method.
    No preview · Article · Sep 2014 · Cybernetics, IEEE Transactions on
  • Jun Shang · Chuanbo Chen · Hu Liang · He Tang · Mudar Sarem

    No preview · Article · Sep 2014 · Journal of Computers
  • Chuanbo Chen · He Tang · Zehua Lyu · Hu Liang · Jun Shang · Mudar Serem
    [Show abstract] [Hide abstract]
    ABSTRACT: Based on the fact that human attention is more likely to be attracted by different objects or statistical outliers of a scene, a bottom-up saliency detection model is proposed. Our model regards the saliency patterns of an image as the outliers in a dataset. For an input image, first, each image element is described as a feature vector. The whole image is considered as a dataset and an image element is classified as a saliency pattern if its corresponding feature vector is an outlier among the dataset. Then, a binary label map can be built to indicate the salient and the nonsalient elements in the image. According to the Boolean map theory, we compute multiple binary maps as a set of Boolean maps which indicate the outliers in multilevels. Finally, we linearly fused them into the final saliency map. This saliency model is used to predict the human eye fixation, and has been tested on the most widely used three benchmark datasets and compared with eight state-of-the-art saliency models. In our experiments, we adopt the shuffled the area under curve metric to evaluate the accuracy of our model. The experimental results show that our model outperforms the state-of-the-art models on all three datasets. (C) 2014 SPIE and IS&T
    No preview · Article · Sep 2014 · Journal of Electronic Imaging
  • Chuanbo Chen · Hu Liang · Shengrong Zhao · Zehua Lyu · Mudar Sarem
    [Show abstract] [Hide abstract]
    ABSTRACT: A high-resolution image is obtained by fusing the information derived from blurred, sub-pixel shifted, and noisy low-resolution observations. In this paper, a novel regularization model based on an Anisotropic Fractional Order Adaptive (AFOA) norm is proposed and then we apply the AFOA model into the Super-Resolution Reconstruction technology. Compared with the existing models, the proposed AFOA model can remove the noise and protect the edges adaptively according to the local features of the images. Meanwhile, the proposed AFOA model can avoid the staircase effect effectively in the smooth region. To obtain the solution to the proposed AFOA model, the Gradient Descent Method is used in this paper. Finally, the experimental results show that the proposed method has much improvement than the existing methods in the respect of the Peak Signal-to-Noise Ratio and the visual quality.
    No preview · Article · Aug 2014 · The Visual Computer
  • Chuanbo Chen · Hu Liang · Shengrong Zhao · Zehua Lyu · Mudar Sarem
    [Show abstract] [Hide abstract]
    ABSTRACT: Multi-frame Super-Reconstruction (SR) is a technique for reconstructing a High-Resolution (HR) image by fusing a set of low-resolution images of the same scene. One of the most difficult problems in SR is to preserve the edges while removing the noise. Therefore, in this paper we propose a novel model, which combines the total variation model and the model by using a pair of different weighting parameters. Also, a hierarchical Bayesian framework is used and the weighting parameters can be modeled together with the HR image and other parameters. Thus the weighting parameters are updated according to the global features of the HR image in iterations. In this way, the proposed model can not only preserve the edges but also it can remove the noise. Our experimental results show that the proposed method has much improvement over the existing methods.
    No preview · Article · Aug 2014 · Computers & Electrical Engineering
  • Source
    Peng Wu · Kai Xie · Houquan Yu · Chuanbo Chen · Chao Wu
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, we propose a new color image representation method based on double-rectangle Non-symmetry Anti-packing pattern representation Model (NAM), which is used to reduce the number of subpatterns in image representation. The method adopts bit-plane optimization strategies at first to reduce the correlation between color image bit planes, then employ the correlation among pixels in bit planes and proceed with double rectangle NAM segmentation to decrease data storage space. Experimental results show that this method is an effective lossless color image coding method.
    Preview · Article · Jan 2012
  • Chuanbo Chen · Guangwei Wang · Mudar Sarem
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, a non-symmetry and anti-packing image representation model (NAM) has been proposed. NAM is a hierarchical image representation method and it aims to provide faster operations and less storage requirement. By taking a rectangle sub-pattern, for example, we describe the idea of NAM and its encoding algorithm.In addition, an approach for adaptive area histogram equalization for image contrast enhancement based on a NAM image is presented. The contrast enhancement approach is designed to meet the NAM image representation and it can be duplicated with faster operation. The complexity analysis and the experimental results show that the NAM based algorithm for image contrast enhancement is faster and more effective than that based on matrix image.
    No preview · Article · Sep 2011 · Computers & Electrical Engineering
  • Guangwei Wang · Chuanbo Chen
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, we present a non-symmetry and anti-packing object pattern representation model (NAM) for object detection. A set of distinctive sub-patterns (object parts) is constructed from a set of sample images of the object class; object pattern are then represented using sub-patterns, together with spatial relations observed among the sub-patterns. Many feature descriptors can be used to describe these sub-patterns.The NAM model codes the global geometry of object category, and the local feature descriptor of sub-patterns deal with the local variation of object. By using Edge Direction Histogram (EDH) features to describe local sub-pattern contour shape within an image, we found that richer shape information is helpful in improving recognition performance. Based on this representation, several learning classifiers are used to detect instances of the object class in a new image. The experimental results on a variety of categories demonstrate that our approach provides successful detection of the object within the image.
    No preview · Article · Dec 2010
  • Guangwei Wang · Chuanbo Chen
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, we propose a non-symmetry and anti-packing image representation model (NAM). NAM is a hierarchical image representation method and its aim is less data amount and faster operation. By taking a rectangle sub-pattern for example, we describe the idea of NAM. In addition, an approach for adaptive area histogram equalization image contrast enhancement based on a NAM image is presented. The contrast enhancement approach is designed to meet the NAM image representation and can be fast implemented. In this work, the contrast enhancement method combines dynamic range modification and adaptive area histogram equalization to improve the visualization of images. Complexity analysis and experimental results show that the NAM based algorithm for image contrast enhancement is faster and more effective than that based on matrix image.
    No preview · Conference Paper · Nov 2010
  • Guangwei Wang · Chuanbo Chen · Kang Chen
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, we propose a non-symmetry and anti-packing image representation model (NAM). NAM is a hierarchical image representation method and its aim is less data amount and faster operation. By taking a rectangle sub-pattern for example, we describe the idea of NAM. In this work, based on block-wise LS linear predictor method, a gray-coded bit-plane binary image NAM representation method is presented to compress the error image. The compression performance has been further improved. Experimental results show that our compression scheme is clearly better from data compression point of view than other methods.
    No preview · Conference Paper · Sep 2010
  • Pengyi Gao · Chuanbo Chen · Kui Zhang · Yingsong Hu · Dan Li
    [Show abstract] [Hide abstract]
    ABSTRACT: In this work, a hybrid neural network model (HNNM) is proposed, which combines the advantages of genetic algorithm, multi-agents and reinforcement learning. In order to generate networks with few connections and high classification performance, HNNM could dynamically prune or add hidden neurons at different stages of the training process. Experimental results have shown to be better than those obtained by the most commonly used optimization techniques.
    No preview · Conference Paper · Jun 2010
  • Pengyi Gao · Chuanbo Chen · Sheng Qin · Yingsong Hu
    [Show abstract] [Hide abstract]
    ABSTRACT: Although many global optimization search algorithms may be used to train feedforward neural networks, these algorithms have some weaknesses such as dependence of initial solution. This paper proposes a novel hybrid global optimization method for classification problem, called GTA, which combines the advantages of Genetic algorithm and Tabu search. The training process in proposed method is divided into two phase. First, a promising initial solution is searched by GA algorithm, and next the best solution is selected by tabu search. In this work, the optimization method and test are discussed. Results obtained by testing Diabetes Data Set have shown that the approach performs better than other optimization algorithm.
    No preview · Conference Paper · Mar 2010
  • Pengyi Gao · Chuanbo Chen · Sheng Qin
    [Show abstract] [Hide abstract]
    ABSTRACT: The selection for the number of hidden nodes for a neural network is of critical importance. This paper proposes a novel algorithm to determine the number of hidden nodes of a neural network and optimize it. In the method, the number of hidden nodes H is first computed by empirical formulas, and the range of H is determined according to computed result. Then, the "three points search" is applied to search the best number of hidden nodes within the range. Finally, a GTA(Genetic algorithm and Tabu search Algorithm Approach) is developed to train the weights of neural network constructed with the best H. Test results obtained by using Iris data set has shown to be efficient, and better than those by the most commonly used optimization techniques.
    No preview · Article · Jan 2010
  • Xiaobing Pei · ShaoHong Fang · ChuanBo Chen
    [Show abstract] [Hide abstract]
    ABSTRACT: Non-negative matrix factorization (NMF) is useful in finding basis information of non-negative data. It is a new dimension reduction method. In this paper, a Group Locality Preserving Orthogonal Nonnegative Matrix Factorization (GLPONMF) is investigated. The idea is to extend the NMF method in order to extract basis vectors for each sample class and at the same time enforce the locality preserving orthogonal properties. Finally, the experimental evaluation has been given.
    No preview · Conference Paper · Jan 2010
  • Source
    Xueli Wu · Chuanbo Chen · Mudar Sarem · Wei Huang
    [Show abstract] [Hide abstract]
    ABSTRACT: The nonsymmetry and antipacking pattern representation model (NAM), inspired by the concept of the packing problem, uses a set of subpatterns to represent an original pattern. The NAM is a promising method for image representation because of its ability to focus on the interesting subsets of an image. In this paper, we develop a new method for gray-scale image representation based on NAM, called NAM-structured plane decomposition (NAMPD), in which each subpattern is associated with a rectangular region in the image. The luminance function of pixels in this region is approximated by an oblique plane model. Then, we propose a new and fast edge detection algorithm based on NAMPD. The theoretical analyses and experimental results presented in this paper show that the edge detection algorithm using NAMPD performs faster than the classical ones because it permits the execution of operations on subpatterns instead of pixels.
    Full-text · Article · Nov 2009 · Geo-spatial Information Science
  • Guangwei Wang · Chuanbo Chen · Mudar Sarem · Shaohong Fang
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, we propose a new Non-symmetry and Anti-packing Model (NAM) representation method for gray image. By describing the NAM and Gray-Coded Bit-Plane Decomposition (GC-BPD), a novel NAM-based representation algorithm for gray image is presented. The theoretical analyses and experimental results show that the NAM-based representation method can reduce the data storage much more effectively than the linear quadtree-based representation method and is a better method to represent the gray image. The method is a new research area and it is valuable for the theoretical research and potential practical values such as decreasing the storage room, quickening the process procedure, matching pattern, and so on.
    No preview · Article · Sep 2009