Time filter

Source Type

Liu Z.,Chongqing University | Chai Y.,Chongqing University | Yin H.,Chongqing University | Yin H.,Key Laboratory of Dependable Service Computing in Cyber Physical Society | And 2 more authors.
Information Fusion | Year: 2017

Multi-focus image fusion is an effective technique to integrate the relevant information from a set of images with the same scene, into a comprehensive image. The fused image would be more informative than any of the source images. In this paper, a novel fusion scheme based on image cartoon-texture decomposition is proposed. Multi-focus source images are decomposed into cartoon content and texture content by an improved iterative re-weighted decomposition algorithm. It can achieve rapid convergence and naturally approximates the morphological structure components. The proper fusion rules are constructed to fuse the cartoon content and the texture content, respectively. Finally, the fused cartoon and texture components are combined to obtain the all-in-focus image. This fusion processing can preserve morphological structure information from source images and performs few artifacts or additional noise. Our experimental results have clearly shown that the proposed algorithm outperforms many state-of-the-art methods, in terms of visual and quantitative evaluations. © 2016 Elsevier B.V.


Liu L.,Key Laboratory of Dependable Service Computing in Cyber Physical Society | Liu L.,Chongqing University | Peng Y.,National University of Singapore | Wang S.,Lanzhou University of Technology | And 2 more authors.
Information Sciences | Year: 2016

Sensor-based human activity recognition has become an important research field within pervasive and ubiquitous computing. Techniques for recognizing atomic activities such as gestures or actions are mature for now, but complex activity recognition still remains a challenging issue. In this paper, we address the problem of complex activity recognition using time series extracted from multiple sensors. We first build a dictionary of time series patterns, called shapelets, to represent atomic activities, then present three shapelet-based models to recognize sequential, concurrent, and generic complex activities. We use the datasets collected from three different labs to evaluate our shapelet-based approach and the results show that our approach can handle complex activity recognition effectively. Our experimental results also show that the shapelet-based approach outperforms other competing approaches in terms of recognition accuracy and system usage. © 2016 Elsevier Inc. All rights reserved.


Ge Y.,Chongqing University | Ge Y.,Key Laboratory of Dependable Service Computing in Cyber Physical Society | Yang D.,Chongqing University | Lu J.,Advanced Digital science Center | And 3 more authors.
Journal of Visual Communication and Image Representation | Year: 2013

Active appearance model (AAM) has been successfully applied to register many types of deformable objects in images. However, the high dimension of intensity used in AAM usually leads to an expensive storage and computational cost. Moreover, intensity values cannot provide enough information for image alignment. In this paper, we propose a new AAM method based on Gabor texture feature representation. Our contributions are two-fold. On one hand, based on the assumption that Gabor magnitude and Gabor phase follow a lognormal distribution and a general Gaussian distribution respectively, three simplified texture representations are proposed. One the other hand, we apply the proposed texture representations in AAM, which is the first time to extract statistical features from both Gabor magnitude and Gabor phase as the texture representation in AAM. Tests on public and our databases show that the proposed Gabor representations lead to more accurate and robust matching between model and images. © 2013 Elsevier B.V. All rights reserved.


Liong V.E.,Advanced Digital science Center | Lu J.,Advanced Digital science Center | Ge Y.,Key Laboratory of Dependable Service Computing in Cyber Physical Society | Ge Y.,Chongqing University
Pattern Recognition Letters | Year: 2015

In this paper, we propose a regularized local metric learning (RLML) method for person re-identification. Unlike existing metric learning based person re-identification methods which learn a single distance metric to measure the similarity of each pair of human body images, our method combines global and local metrics to represent the within-class and between-class variances. By doing so, we utilize the local distribution of the training data to avoid the overfitting problem. In addition, to address the lacking of training samples in most person re-identification systems, our method also regulates the covariance matrices in a parametric manner, so that discriminative information can be better exploited. Experimental results on four widely used datasets demonstrate the advantage of our proposed RLML over both existing metric learning and state-of-the-art person re-identification methods. © 2015 Elsevier B.V.


Liu Y.,Chongqing University | Yin H.P.,Chongqing University | Yin H.P.,Key Laboratory of Dependable Service Computing in Cyber Physical Society | Chai Y.,Chongqing University
Lecture Notes in Electrical Engineering | Year: 2016

Kernel K-means is an extended method of K-means, which identifies nonlinearly separable clusters. However it still exits limitations, the one is which repeatedly sets different initial positions to find better local minima, the other is that it can only for linear separable data clustering. In order to overcome this issue, in this paper we propose an improved global kernel k-means. The proposed algorithm adds one cluster at every stage and generates the next centric point at next stage to avoid the unnecessary calculation. Experimental result shows that the proposed algorithm does not depend on initialization which identifies nonlinearly separable cluster, meanwhile, because of the incremental nature and search procedure, the poor local minima is avoided. Moreover, an improvement is put forward to decrease the computational complexity that does not significantly affect the accuracy of classification. © Springer Science+Business Media Singapore 2016.


Liu Z.,Chongqing University | Yin H.,Chongqing University | Yin H.,Key Laboratory of Dependable Service Computing in Cyber Physical Society | Chai Y.,Chongqing University | Yang S.X.,University of Guelph
Expert Systems with Applications | Year: 2014

Fusion of multimodal medical images increases robustness and enhances accuracy in biomedical research and clinical diagnosis. It attracts much attention over the past decade. In this paper, an efficient multimodal medical image fusion approach based on compressive sensing is presented to fuse computed tomography (CT) and magnetic resonance imaging (MRI) images. The significant sparse coefficients of CT and MRI images are acquired via multi-scale discrete wavelet transform. A proposed weighted fusion rule is utilized to fuse the high frequency coefficients of the source medical images; while the pulse coupled neural networks (PCNN) fusion rule is exploited to fuse the low frequency coefficients. Random Gaussian matrix is used to encode and measure. The fused image is reconstructed via Compressive Sampling Matched Pursuit algorithm (CoSaMP). To show the efficiency of the proposed approach, several comparative experiments are conducted. The results reveal that the proposed approach achieves better fused image quality than the existing state-of-the-art methods. Furthermore, the novel fusion approach has the superiority of high stability, good flexibility and low time consumption. © 2014 Elsevier Ltd. All rights reserved.


Chen X.,Chongqing University | Sha E.H.-M.,Key Laboratory of Dependable Service Computing in Cyber Physical Society | Sha E.H.-M.,Chongqing University | Zhuge Q.,Key Laboratory of Dependable Service Computing in Cyber Physical Society | And 3 more authors.
Proceedings - Design Automation Conference | Year: 2015

Domain Wall Memory (DWM) using nanowire with data access port, exhibits extraordinary high density, low power leakage, and low access latency. These properties enable DWM to become an attractive candidate for replacing traditional memories. However, data accesses on DWM may require multIPle shift operations before the port points to requested data, resulting in varying access latencies. Data placement, therefore, has a significant impact on the performance of data accesses on DWM. This paper studies compiler-based optimization techniques for data placement on DWM. To the authors' best knowledge, this is the first work addressing data placement problem on DWM. We present an efficient heuristic, called Grouping-Based Data Placement (GBDP), for the data placement problem of a given data access sequence on DWM. The experimental results show that GBDP has a significant performance improvement; for example, GBDP reduces 82% shift operations on an 8-port DWM compared with non-optimized approach. © 2015 ACM.


Liu Z.,Chongqing University | Yin H.,Chongqing University | Yin H.,Key Laboratory of Dependable Service Computing in Cyber Physical Society | Fang B.,Chongqing University | Chai Y.,Chongqing University
Optics Communications | Year: 2014

An appropriate fusion of infrared and visible images can integrate their complementary information and obtain more reliable and better description of the environmental conditions. Compressed sensing theory, as a low signal sampling and compression method based on the sparsity of signal under a certain transformation, is widely used in various fields. Applying to the image fusion applications, only a part of sparse coefficients are needed to be fused. Furthermore, the fused sparse coefficients can be used to accurately reconstruct the fused image. The CS-based fusion approach can greatly reduce the computational complexity and simultaneously enhance the quality of the fused image. In this paper, an improved image fusion scheme based on compressive sensing is presented. This proposed approach can preserve more detail information, such as edges, lines and contours in comparison to the conventional transform-based image fusion approaches. In the proposed approach, the sparse coefficients of the source images are obtained by discrete wavelet transform. The low and high coefficients of infrared and visible images are fused by an improved entropy weighted fusion rule and a max-abs-based fusion rule, respectively. The fused image is reconstructed by a compressive sampling matched pursuit algorithm after local linear projection using a random Gaussian matrix. Several comparative experiments are conducted. The experimental results show that the proposed image fusion scheme can achieve better image fusion quality than the existing state-of-the-art methods. © 2014 Elsevier B.V.


He W.,Chongqing University | He W.,Key Laboratory of Dependable Service Computing in Cyber Physical Society | Sun D.-H.,Chongqing University | Sun D.-H.,Key Laboratory of Dependable Service Computing in Cyber Physical Society
International Journal of Advanced Manufacturing Technology | Year: 2013

In this paper, flexible job shop scheduling problem with machine breakdown is of concern. Considering that there is a limitation in improving robust and stable performance of rescheduling with a single strategy, an approach with multi-strategies is proposed to make the scheduling more robust and stable. First, in prescheduling, a new idle time insertion strategy is put forward. In this new policy, idle time equal to repair time is inserted into an appropriate position of each machine according to the machine's breakdown nature. Second, route changing strategy combined with right-shift policy is proposed to keep the rescheduling as stable and robust as possible. In this policy, whether to right shift or route change is dependent on the cost of archiving robustness and stability. Based on the two strategies, new algorithms dealing with idle time insertion, right-shift scheduling, and route changing scheduling are designed. The computational results show the effectiveness of the new strategies and new algorithms compared with other strategies. © 2012 Springer-Verlag London Limited.


Liong V.E.,Advanced Digital science CenterSingapore | Lu J.,Advanced Digital science CenterSingapore | Ge Y.,Key Laboratory of Dependable Service Computing in Cyber Physical Society | Ge Y.,Chongqing University
Pattern Recognition Letters | Year: 2015

In this paper, we propose a regularized local metric learning (RLML) method for person re-identification. Unlike existing metric learning based person re-identification methods which learn a single distance metric to measure the similarity of each pair of human body images, our method combines global and local metrics to represent the within-class and between-class variances. By doing so, we utilize the local distribution of the training data to avoid the overfitting problem. In addition, to address the lacking of training samples in most person re-identification systems, our method also regulates the covariance matrices in a parametric manner, so that discriminative information can be better exploited. Experimental results on four widely used datasets demonstrate the advantage of our proposed RLML over both existing metric learning and state-of-the-art person re-identification methods. © 2015 Elsevier B.V.

Loading Key Laboratory of Dependable Service Computing in Cyber Physical Society collaborators
Loading Key Laboratory of Dependable Service Computing in Cyber Physical Society collaborators