期刊:
The Journal of Information and Computational Science,2015年12(12):4689-4695 ISSN:1548-7741
通讯作者:
Chen, Yuantao
作者机构:
[Chen, Yuantao] School of Computer and Communication Engineering, Changsha University of Science &, Technology, Changsha, China;[Wang, Zhongyuan; Zuo, Jingwen; She, Kang; Xiang, Zhiwu] College of Chengnan, Changsha University of Science &, Technology, Changsha, China
通讯机构:
School of Computer and Communication Engineering, Changsha University of Science & Technology, Changsha, China
关键词:
Local gradient ratio feature measurement;Object recognition;SAR image;Similarity calculation
期刊:
Geotechnical Special Publication,2014年(248 GSP):83-91 ISSN:0895-0563
作者机构:
[Yuan, Jianbo] ChengNan College, Changsha University of Science and Technology, Changsha, Hunan, 410076, China;[He, Ying] Hydrochina Zhongnan Engineering Corporation, Construction and Traffic Engineering Division, Changsha, Hunan, 410014, China
会议名称:
Geo-Hubei 2014 International Conference on Sustainable Infrastructure: Characterization, Modeling, and Evaluation of Geotechnical Engineering Systems
会议时间:
July 20, 2014 - July 22, 2014
会议地点:
Yichang, Hubei, China
会议论文集名称:
Characterization, Modeling, and Evaluation of Geotechnical Engineering Systems
期刊:
Journal of Networks,2014年9(1):223-230 ISSN:1796-2056
通讯作者:
Wu, X.(252167977@qq.com)
作者机构:
[Wu, Xianwen; Xue, Zhiliang] Department of Information Engineering, Hunan Railway Professional Technology College, Zhuzhou 421001, China;[Zuo, Jingwen] Computer Center, College of ChengNan, Changsha University of Science and Technology, Changsha 410076, China
期刊:
Journal of Computational Information Systems,2014年10(22):9547-9554 ISSN:1553-9105
通讯作者:
Chen, Yuantao
作者机构:
[Chen, Yuantao; Gui, Yan] School of Computer and Communication Engineering, Changsha University of Science &, Technology, Changsha , China;[Zuo, Jingwen; She, Kang; Xiang, Zhiwu] College of Chengnan, Changsha University of Science &, Technology, Changsha , China
通讯机构:
School of Computer and Communication Engineering, Changsha University of Science & Technology, Changsha, China
摘要:
In view of the traditional fuzzy C-means (FCM) clustering algorithm, it is difficult to determine the number of clustering procedure. The iterative speed of FCM is slow than many improved method. The weakness of FCM is easy to fall into local optimum and the initial value of clustering centers is sensitive. The paper has presented a fusion of global search algorithm and fuzzy C-means (GS-FCM) clustering algorithm. Firstly, using the global search (GS) algorithm is globally and robustness for the initial clustering center and the clustering number. The traditional FCM clustering algorithm has the initial cluster centers and the number of clusters. Secondly, poly category standard function is proposed by the novel fuzzy. The image pixel neighborhood dependence is taken into account, and the common role of pixel information is to enhance the continuity of the segmentation results of space. In addition, we also employs a novel distance instead of Euclidean distance formula and robustness of the new algorithm for noise. The simulation results show that the new algorithm can effectively avoid the traditional FCM algorithm for the initial clustering center. It is sensitive and convergence to local optimal solution. The clustering speed and robustness, accuracy is improved than the traditional FCM algorithm. The images with different characteristics of the division has achieved good results.
摘要:
For spectral clustering is applied to image segmentation is difficult to calculate the spectral weight matrix of the actual problem, we have defined the pixel distance between the point and the class is given a sampling theorem, the design of a hierarchical image segmentation algorithm in the use of this algorithm for image segmentation. By adjusting the scaling factor to merge or split a large class of smaller classes, so the image segmentation both randomness but also has multi-scale feature, called spectral clustering based on multi-scale stochastic tree image segmentation (SCMSTIS). The experimental results show that the algorithm is effective.
作者机构:
[Y.T.Chen; J.W.Zuo; Z.W.Xiang; K.She] School of Computer and Communication Engineering,Changsha University of Science and Technology;[Y.T.Chen; J.W.Zuo; Z.W.Xiang; K.She] College of Chengnan, Changsha University of Science and Technology
会议名称:
2014 International Conference on Artificial Intelligence and Industrial Application (AIIA2014)
会议时间:
2014-09-21
会议地点:
中国香港
会议论文集名称:
Proceedings of 2014 International Conference on Artificial Intelligence and Industrial Application (AIIA2014)
关键词:
support vector machine;feature model;texture features;highresolution remote sensing image
摘要:
A river is an important geographical structure, and has very typical significance in both military and civilian circumstances. For detection and identification of a river, a support vector machine(SVM