Jiexin Pu

A novel classification method for palmprint recognition based on reconstruction error and normalized distance

In this paper, we propose a fusion classification method based on reconstruction error and normalized distance for palmprint recognition. This method first obtains an approximate representation of the test sample by solving a linear system in which the test sample is assumed to be a linear combination of all the original training samples. Then it replaces the test sample by its approximate representation and decomposes the approximate representation as a weighted sum of all the training samples. The proposed method calculates the reconstruction error of the approximate representation from the weighted sum of the training samples from each class. The method also computes the normalized distance between the test sample and each class. Finally, the method integrates the reconstruction error and normalized distance between the test sample and a class to form the matching score and assigns the test sample into the class that has the smallest matching score. Experimental results on the palmprint databases demonstrate the effectiveness of our method.

Face Recognition Via Weighted Two Phase Test Sample Sparse Representation

Sparse representation (SR) for signals over an overcomplete dictionary fascinates a lot of researchers in the past decade. The two-phase test sample sparse representation method (TPTSSR) achieved an excellent performance in face recognition. However, TPTSSR exploits the global information and tends to lose local information. In this paper, the weighted two phase test sample sparse representation method (WTPTSSR) is proposed. WTPTSSR utilizes both data locality and linearity and it can be regarded as extensions of TPTSSR. Experiments on the face databases demonstrate that WTPTSSR is more effective than TPTSSR.

Quaternion Based Maximum Margin Criterion Method for Color Face Recognition

Color is one of the basic features of images, which can provide very useful information and play an important role in face recognition. By using the quaternion matrix representation, the R, G, B information of each pixel is not destroyed and it can be taken as a organic body. Therefore, this paper proposes a quaternion based maximum margin criterion (QMMC) algorithm. Firstly, the quaternion number is used to denote the pixel of the color image, and a quaternion vector is taken to represent the color image. Secondly, the maximum margin criterion algorithm is used to project the quaternion vector in the high-dimension space into a low-dimension space. Finally, the nearest neighbor classification are taken for classification recognition. Numerous experiments show that the proposed QMMC can achieve better recognition performance.

Salient object detection in complex scenes via D-S evidence theory based region classification

In complex scenes, multiple objects are often concealed in cluttered backgrounds. Their saliency is difficult to be detected by using conventional methods, mainly because single color contrast can not shoulder the mission of saliency measure; other image features should be involved in saliency detection to obtain more accurate results. Using Dempster-Shafer (D-S) evidence theory based region classification, a novel method is presented in this paper. In the proposed framework, depth feature information extracted from a coarse map is employed to generate initial feature evidences which indicate the probabilities of regions belonging to foreground or background. Based on the D-S evidence theory, both uncertainty and imprecision are modeled, and the conflicts between different feature evidences are properly resolved. Moreover, the method can automatically determine the mass functions of the two-stage evidence fusion for region classification. According to the classification result and region relevance, a more precise saliency map can then be generated by manifold ranking. To further improve the detection results, a guided filter is utilized to optimize the saliency map. Both qualitative and quantitative evaluations on three publicly challenging benchmark datasets demonstrate that the proposed method outperforms the contrast state-of-the-art methods, especially for detection in complex scenes.

Multi-scale counting and difference representation for texture classification

Multi-scale analysis has been widely used for constructing texture descriptors by modeling the coefficients in transformed domains. However, the resulting descriptors are not robust to the rotated textures when performing texture classification. To alleviate this problem, we in this paper propose a multi-scale counting and difference representation (CDR) of image textures for texture classification. Particularly, we first extract a single-scale CDR feature consisting of the local counting vector (LCV) and the differential excitation vector (DEV). The LCV is established to capture different types of textural structures using the discrete local counting projection, while the DEV is used to describe the difference information of textures in accordance with the differential excitation projection. Finally, the multi-scale CDR feature of a texture image is constructed by combining CDRs at different scales. Experimental results on Brodatz, VisTex, and Outex databases demonstrate that our proposed multi-scale CDR-based texture classification method outperforms five representative texture classification methods.

Extended Locality-Constrained Linear Self-Coding for Saliency Detection

In complex scenes, foreground saliency can hardly be detected completely, which may further result in the ambiguous cues of objects for other computer vision tasks. In this letter, an extended locality-constrained linear self-coding (eLLsC) scheme is proposed to assist to solve the saliency detection problem under the complex scenes. The locality of both spatial relation and feature distance is preserved in eLLsC, thus making the transformed code involved in the manifold ranking to prompt the generation of the saliency map with more complete foreground and clearer boundary. Experimental results on three saliency detection benchmarks demonstrate the effectiveness of the proposed hybrid method.

Scene classification-oriented saliency detection via the modularized prescription

Saliency detection technology has been greatly developed and applied in recent years. However, the performance of current methods is not satisfactory in complex scenes. One of the reasons is that the performance improvement is often carried out through utilizing complicated mathematical models and involving multiple features rather than classifying the scene complexity and respectively detecting saliency. To break this unified detection schema for generating better results, we propose a method of scene classification-oriented saliency detection via the modularized prescription in this paper. Different scenes are described by a scene complexity expression model, and they are analyzed and discriminately detected by different pipelines. This process seems like that doctors can tailor the treatment prescriptions when they meet different symptoms. Moreover, two SVM-based classifiers are trained for scene classification and sky region identification, and the proposed sky region discrimination and erase model can be used to efficiently decrease the saliency interference by the high luminance of the background sky regions. Experimental results demonstrate the effectiveness and superiority of the proposed method in both higher precision and better smoothness, especially for detecting in structure complex scenes.

Structural difference histogram representation for texture image classification

Local binary pattern (LBP) is a frequently-used texture descriptor. Lots of LBP-variants have been proposed to improve its performance of representing textures. However, most of them ignore the global and neighbour-difference information of an image texture. In this study, the authors propose a structural difference histogram representation by fusing the segmented structure pattern (SSP), the refined LBP (RLBP) and the neighbour-difference pattern (NDP) for texture classification. Particularly, the segmented structure, which contains the global contour information of an image texture, is first constructed to compute its SSP histogram feature. Simultaneously, the RLBP is defined to represent the local texture information. Furthermore, the NDP is presented to describe differences between neighbours of a centre pixel in the local patch of texture images. Experimental results on Brodatz and Columbia-Utrecht reflectance databases indicate that the proposed method can achieve the satisfactory classification accuracy compared with several representative methods.

Hybrid of extended locality-constrained linear coding and manifold ranking for salient object detection

Abstract Recent years have witnessed great progress of salient object detection methods. However, due to the emerging complex scenes, two problems should be solved urgently: one is on the fast locating of the foreground while preserving the precision, and the other is about reducing the noise near the foreground boundary in saliency maps. In this paper, a hybrid method is proposed to ameliorate the above two issues. At first, to reduce the essential runtime of integrating the prior knowledge, a novel Prior Knowledge Learning based Region Classification (PKL-RC) method is proposed for classifying image regions and preliminarily locating foreground; furthermore, to generate more accurate saliency, a Locality-constrained Linear self-Coding based Region Clustering (LLsC-RC) model is proposed to improve the adjacency structure of the similarity graph for Manifold Ranking (MR). Experimental results demonstrate the effectiveness and superiority of the proposed method in both higher precision and better smoothness.

Orthogonal sparse linear discriminant analysis

ABSTRACT Linear discriminant analysis (LDA) is a linear feature extraction approach, and it has received much attention. On the basis of LDA, researchers have done a lot of research work on it, and many variant versions of LDA were proposed. However, the inherent problem of LDA cannot be solved very well by the variant methods. The major disadvantages of the classical LDA are as follows. First, it is sensitive to outliers and noises. Second, only the global discriminant structure is preserved, while the local discriminant information is ignored. In this paper, we present a new orthogonal sparse linear discriminant analysis (OSLDA) algorithm. The k nearest neighbour graph is first constructed to preserve the locality discriminant information of sample points. Then, L2,1-norm constraint on the projection matrix is used to act as loss function, which can make the proposed method robust to outliers in data points. Extensive experiments have been performed on several standard public image databases, and the experiment results demonstrate the performance of the proposed OSLDA algorithm.