摘要:
Image classification is a popular and challenging topic in the computer vision field. On the basis of advances in neuroscience, this paper proposes a sparse-based neural response feature extraction method for image classification. The approach extracts discriminative and invariant representations of images by alternating between non-negative sparse coding and maximum pooling operation with effectiveness. Additionally, effective template selection methods are proposed to further enhance the performance of the algorithm. In comparison with traditional hierarchical methods, our proposed model accounts for the neural processing of visual cortex in human brain, which appears to gain more beneficial discriminative and robust properties for image classification tasks. A variety of benchmarks are used to evaluate the algorithm. The experiment results demonstrate that our proposed algorithm achieves quite excellent or state-of-the-art performance compared with other popular methods. Image classification is a popular and challenging topic in the computer vision field. On the basis of advances in neuroscience, this paper proposes a sparse-based neural response feature extraction method for image classification. The approach extracts discriminative and invariant representations of images by alternating between non-negative sparse coding and maximum pooling operation with effectiveness. Additionally, effective template selection methods are proposed to further enhance the performance of the algorithm. In comparison with traditional hierarchical methods, our proposed model accounts for the neural processing of visual cortex in human brain, which appears to gain more beneficial discriminative and robust properties for image classification tasks. A variety of benchmarks are used to evaluate the algorithm. The experiment results demonstrate that our proposed algorithm achieves quite excellent or state-of-the-art performance compared with other popular methods.
摘要:
In this paper, a model of network utility maximization (NUM) is presented for random access control in multi-hop wireless networks. Different from the classical NUM framework, our model considers the queueing stability. We propose a distributed iterative prices and link probabilities adaption algorithm by using dual decomposition techniques, which only requires limited message passing, but converges to the global optimum of the total network utility. Numerical results and simulation comparison validate our conclusion.
摘要:
A new approach for feature extraction using neural response has been developed in this paper through combining the hierarchical architectures with the sparse coding technique. As far as proposed layered model, at each layer of hierarchy, it concerned two components that were used are sparse coding and pooling operation. While the sparse coding was used to solve increasingly complex sparse feature representations, the pooling operation by comparing sparse outputs was used to measure the match between a stored prototype and the input sub-image. It is recommended that value of the best matching should be kept and discarding the others. The proposed model is implemented and tested taking into account two ranges of recognition tasks i.e. image recognition and speech recognition (on isolated word vocabulary). Experimental results with various parameters demonstrate that proposed scheme leads to extract more efficient features than other methods.
期刊:
International Journal of Modelling, Identification and Control,2013年20(2):190-198 ISSN:1746-6172
通讯作者:
Qu, S.(qushaocheng@mail.ccnu.edu.cn)
作者机构:
[Wang, Li] Network Centre, Wuhan Conservatory of Music, Wuhan city, Hubei Province, 430060, China;[Zhu, Xiaokun] Department of Journal, Central China Normal University, Wuhan City, Hubei Province, 430079, China;[Zhou, Bing; Qu, Shaocheng] Department of Information and Technology, Central China Normal University, Wuhan City, Hubei Province, 430079, China
通讯机构:
Department of Information and Technology, Central China Normal University, China
关键词:
human dynamics;power law;interval time distribution;exercise behaviour;habit;human behaviour;behaviour dynamics;fitness clubs;club membership;dynamic modelling.
期刊:
Proceedings of SPIE - The International Society for Optical Engineering,2013年8783 ISSN:0277-786X
通讯作者:
Deng, He
作者机构:
[Deng, He; Liu, Qingtang] Cent China Normal Univ, Dept Informat Technol, Wuhan 430079, Peoples R China.;[Cheng, Lifang] Air Force Radar Acad, Wuhan 430019, Peoples R China.
通讯机构:
[Deng, He] C;Cent China Normal Univ, Dept Informat Technol, Wuhan 430079, Peoples R China.
会议名称:
5th International Conference on Machine Vision (ICMV) - Computer Vision, Image Analysis and Processing
会议时间:
OCT 20-21, 2012
会议地点:
Wuhan, PEOPLES R CHINA
会议主办单位:
[Deng, He;Liu, Qingtang] Cent China Normal Univ, Dept Informat Technol, Wuhan 430079, Peoples R China.^[Cheng, Lifang] Air Force Radar Acad, Wuhan 430019, Peoples R China.
会议论文集名称:
Proceedings of SPIE
关键词:
Small target detection;local entropy;local reverse entropy
摘要:
The detection of small targets that submerges in the homogeneous image background is a difficult problem in the design of small targets detection algorithm. The concept of local reverse entropy combining the concepts of local entropy with reverse entropy is proposed in this paper, which is to solve that problem by enhancing small targets. The norms of local signal-to-background ratio and elapsed time are used to demonstrate the performance of local reverse entropy map. Both quantitative analysis and qualitative comparison confirm the validity and efficiency of the presented approach.
期刊:
International Journal of Future Generation Communication and Networking,2013年6(3):1-12 ISSN:2233-7857
作者机构:
[Duan, Jia-Qi; Li, Shining] Shaanxi Key Lab of Embedded System Technology, School of Computer Science, Northwestern Polytechnical University, Xi'an, 710129, China;[Ning, Guoqin] Department of Information Technology, Huazhong Normal University, Wuhan, 430079, China
摘要:
Cognitive radio enabled vehicular networks (CR-VNETs) is a new communication paradigm that enables moving vehicles to identify spectrum opportunities along busy streets and freeways. This detected spectrum may possibly lie in licensed frequency bands, and can be used for emergency communications, such as by primary responders during crises events. Spectrum sensing ensures that this spectrum is not currently occupied by licensed users, who have priority access rights. However, as the vehicles are in motion, the spectrum sensing at a given location must be completed with minimum delay, a challenge for classical energy and feature based detection schemes. This paper presents a new distributed compressive sampling technique that allows individual vehicles to report partial information to a centralized base station (BS), with an overhead of only few bytes. Thus, we tradeoff reporting time with processing complexity at the BS, which is tasked with re-constructing the overall spectrum utilization from these portions. Simulation results reveal significant improvements in detection time and accuracy, making our approach suitable for CR-VNETs.
摘要:
Background suppression is vitally important for the small target detection, which aims to enhance targets and improve the signal-to-noise ratio of small target images. Consequently, the study proposes a background suppression approach based on the fast local reverse entropy operator, which is designed according to the fact that the appearance of a small target could result in the great change of the value of local reverse entropy in the local region. The operator is adopted to suppress complex backgrounds of small target images in order to enhance small targets, and then bring about high probabilities of detection and low probabilities of false alarm in the small target detection. Both quantitative and qualitative analyses contribute to confirm the validity and efficiency of the proposed approach.
作者机构:
[Chen, Dan; Wang, Lizhe] China Univ Geosci, Sch Comp, Wuhan 430074, Peoples R China.;[Wang, Lizhe] Chinese Acad Sci, Ctr Earth Observat & Digital Earth, Beijing 100864, Peoples R China.;[Streit, Achim; Tao, Jie; Marten, Holger] Karlsruhe Inst Technol, Steinbuch Ctr Comp, D-76021 Karlsruhe, Germany.;[Ranjan, Rajiv] CSIRO, ICT Ctr, Informat Engn Lab, Canberra, ACT, Australia.;[Chen, Jingying] Cent China Normal Univ, Natl Engn Ctr E Learning, Beijing, Peoples R China.
通讯机构:
[Wang, Lizhe] C;Chinese Acad Sci, Ctr Earth Observat & Digital Earth, Beijing 100864, Peoples R China.
关键词:
Cloud computing;Data-intensive computing;Hadoop;MapReduce;Massive data processing
摘要:
Recently, the computational requirements for large-scale data-intensive analysis of scientific data have grown significantly. In High Energy Physics (HEP) for example, the Large Hadron Collider (LHC) produced 13 petabytes of data in 2010. This huge amount of data is processed on more than 140 computing centers distributed across 34 countries. The MapReduce paradigm has emerged as a highly successful programming model for large-scale data-intensive computing applications. However, current MapReduce implementations are developed to operate on single cluster environments and cannot be leveraged for large-scale distributed data processing across multiple clusters. On the other hand, workflow systems are used for distributed data processing across data centers. It has been reported that the workflow paradigm has some limitations for distributed data processing, such as reliability and efficiency. In this paper, we present the design and implementation of G-Hadoop, a MapReduce framework that aims to enable large-scale distributed computing across multiple clusters.
期刊:
Information Technology Journal,2013年12(3):434-438 ISSN:1812-5638
通讯作者:
Wang, F.
作者机构:
[Wang, Feng; Xu, Shun] School of Educational Science and Technology, Huang Gang Normal University, Huang Gang, China;[Xu, Shun] Department of Information Technology, Central China Normal University, China
通讯机构:
[Wang, F.] S;School of Educational Science and Technology, Huang Gang Normal University, China
期刊:
Proceedings of SPIE - The International Society for Optical Engineering,2013年8783 ISSN:0277-786X
通讯作者:
Deng, He
作者机构:
[Deng, He] Cent China Normal Univ, Dept Informat Technol, Wuhan 430079, Peoples R China.;[Cheng, Lifang] Air Force Radar Acad, Wuhan 430019, Peoples R China.
通讯机构:
[Deng, He] C;Cent China Normal Univ, Dept Informat Technol, Wuhan 430079, Peoples R China.
会议名称:
5th International Conference on Machine Vision (ICMV) - Computer Vision, Image Analysis and Processing
会议时间:
OCT 20-21, 2012
会议地点:
Wuhan, PEOPLES R CHINA
会议主办单位:
[Deng, He] Cent China Normal Univ, Dept Informat Technol, Wuhan 430079, Peoples R China.^[Cheng, Lifang] Air Force Radar Acad, Wuhan 430019, Peoples R China.
摘要:
The problems of detection, identification and tracking of moving target under the condition of moving camera are quite difficult in the application of computer vision. This paper proposes two different methods to the detection of moving target in the light of the size of the object. One approach integrates the image-interpolation technique and the active contour model, the other unites the image-interpolation technique, the modified local entropy algorithm and the region growing skill. The introduction of the image-interpolation technique is to estimate the global motion, i.e. background motion. The active contour model or the modified local entropy algorithm and region growing skill act upon the difference image, and then detect the moving target. It is demonstrated the reasonableness and efficiency of two detection approaches through experiments.<br/>
期刊:
International Journal of Digital Content Technology and its Applications,2012年6(18):442-448 ISSN:1975-9339
通讯作者:
Wan, L.(liyongwan20120@126.com)
作者机构:
[Wan, Liyong] National Engineering Research Center for E-Learning, Huazhong Normal University, Wuhan, China;[Zhao, Chengling] Department of Information and Technology, Huazhong Normal University, Wuhan, China;[Wan, Liyong] School of Media and Communication, Wuhan Textile University, Wuhan, China
摘要:
In adaptive e-learning system, the researchers focused on how to provide the learners with learning objects depending on their preferences and needs. This makes the process of filtering unwanted learning objects an increasing challenge for the learners. Content recommendation systems can suggest the users what they should learn. People have made great efforts on research about recommendation systems using content-based recommendation and collaboration-based recommendation. In this paper, the authors proposed a hybrid learning object recommendation method to improve the accuracy of the recommendation. The experiment result shows that recommendation precision of hybrid recommendation method is higher than content-based recommendation method and collaboration-base recommendation method.<br/>
期刊:
Proceedings of the 2012 24th Chinese Control and Decision Conference, CCDC 2012,2012年:208-211 ISSN:1948-9439
通讯作者:
Xia, Dan
作者机构:
[Yang, Fei] Wuhan Univ, Sch Power & Mech Engn, Dept Automat, Wuhan 430072, Peoples R China.;[Xia, Dan] Cent China Normal Univ, Dept Informat Technol, Wuhan, Peoples R China.;[Meng, Fanbao] People Liberat Army, Unit 75719, Wuhan, Peoples R China.
通讯机构:
[Xia, Dan] C;Cent China Normal Univ, Dept Informat Technol, Wuhan, Peoples R China.
会议名称:
The 24th Chinese Control and Decision Conference (第24届中国控制与决策学术年会 2012 CCDC)
会议时间:
2012-05-23
会议地点:
太原
会议主办单位:
[Xia, Dan] Cent China Normal Univ, Dept Informat Technol, Wuhan, Peoples R China.^[Yang, Fei] Wuhan Univ, Sch Power & Mech Engn, Dept Automat, Wuhan 430072, Peoples R China.^[Meng, Fanbao] People Liberat Army, Unit 75719, Wuhan, Peoples R China.
会议论文集名称:
The 24th Chinese Control and Decision Conference (第24届中国控制与决策学术年会 2012 CCDC)论文集
期刊:
The Journal of Information and Computational Science,2012年9(17):5225-5232 ISSN:1548-7741
通讯作者:
Wang, S.(shujuan841019@gmail.com)
作者机构:
[Wang, Shujuan; Yang, Zongkai] National Engineering Research Center for E-learning, Central China Normal University, Wuhan 430079, China;[Liu, Qingtang] Department of Information Technology, Central China Normal University, Wuhan 430079, China
通讯机构:
National Engineering Research Center for E-learning, Central China Normal University, China
关键词:
Buyer-seller protocol;Copyright violation arbitration;Digital rights management;Digital watermarking
关键词:
3D modeling;Image segmentation;3D image processing;Visualization;Visual process modeling;Cameras;Image resolution;3D image reconstruction;Reconstruction algorithms;Optical engineering
摘要:
Modeling the three-dimensional (3-D) shape of plant stems is important in the study of plant growth in precision agriculture. To construct a 3-D model of real plant stems from images quickly, a novel volumetric method based on line-based models is proposed. Line-based models are constructed on the coarse 3-D skeleton of the plant stems, then carved with respect to silhouette consistency. The surface points on the plant stems are calculated from line-based models. Finally, a mesh surface model can be extracted from the surface points. The proposed method can give precise results together with low time complexity and space complexity. Experiments based on both synthetic and real data are presented to evaluate the speed and preciseness of the proposed method. (C) 2012 Society of Photo-Optical Instrumentation Engineers (SPIE). [DOI: 10.1117/1.OE.51.2.021116]