基于视觉注意模型和进化规划的感兴趣区检测方法

张菁; 沈兰荪; 高静静

doi:10.3724/SP.J.1146.2008.00739

基于视觉注意模型和进化规划的感兴趣区检测方法

doi: 10.3724/SP.J.1146.2008.00739

基金项目:

国家自然科学基金(60772069，60402036)和北京市自然科学基金(4062006)资助课题

计量
- 文章访问数: 3456
- HTML全文浏览量: 146
- PDF下载量: 1928
- 被引次数: 0
出版历程
- 收稿日期: 2008-06-03
- 修回日期: 2009-02-27
- 刊出日期: 2009-07-19

Region of Interest Detection Based on Visual Attention Model and Evolutionary Programming

摘要

摘要: 根据生物注意机制，该文提出了一种基于视觉注意模型和进化规划的感兴趣区检测方法。采用进化规划方法分割图像候选区域；区域兴趣度由视觉注意模型产生的局部显著和进化规划计算的全局显著共同度量。在视觉注意模型中，图像经过小波多尺度变换和计算中央周边差得到局部显著度。注意焦点在显著度增强因子的作用下，选取候选区域得到感兴趣区。实验结果表明，所提方法检测的感兴趣区更接近人眼的视觉注意机制，并取得了较为满意的对象检测和兴趣度量结果。
- 计算机视觉;视觉注意模型;感兴趣区;进化规划;显著度;注意焦点
Abstract: According to biological attention mechanism, a region of interest detection based on visual attention model and evolutionary programming is proposed in this paper. Candidate image regions are segmented with evolutionary programming algorithm. The interest of a region is measured with local saliency produced by visual attention model and global saliency based on evolutionary programming. After wavelet multiscale transform, local saliency in visual attention model is computed by center-surround differences. Under the action of saliency enhancement factors, focus of attention selects region of interest from candidate regions. The experimental results show the proposed approach is closer to human attention mechanism and performs well for object detection and interest measurement.

HTML全文

参考文献(1)

沈兰荪, 卓力. 小波编码与网络视频传输[M]. 北京: 科学出版社, 2005: 9-11.Shen L S and Zhuo L. Wavelet Coding and Network Videotransmission [M]. Beijing: Science Press, 2005: 9-11.[2]Khanh V, Hua K A, and Tavanapong W. Image retrievalbased on regions of interest [J].IEEE Transactions on.Knowledge and Data Engineering.2003, 15(4):1045-1049[3]Bulthoff H H, Lee S W, and Poggio T, et al.. BiologicallyMotivated Computer Vision [M]. New York: SpringerPublishing Company, 2003: 150-159.[4]Huang L and Pashler H. Working memory and the guidanceof visual attention: Consonance-driven orienting [J].Psychonomic Bulletin Review.2007, 14(1):148-153[5]张菁, 沈兰荪, 等. 基于视觉感知的图像检索的研究[J]. 电子学报, 2008, 36(3): 494-499.Zhang J and Shen L S, et al.. A Survey of image retrievalbased on visual perception [J]. Acta Electronica Sinica, 2008,36(3): 494-499.[6]Itti L and Koch C. Computational modeling of visualattention [J].Nature Reviews Neuroscience.2001, 2(3):194-203[7]Fogel L J. Toward inductive inference automata [C]. InProceedings of the International Federation for InformationProcessing Congress. Munich, Germany, August 27-September 1, 1962: 395-399.[8]周方俊, 王向军, 张民. 基于t 分布变异的进化规划[J]. 电子学报, 2008, 36(4): 667-671.Zhou F J, Wang X J, and Zhang M. Evolutionaryprogramming using mutations based on the t probabilitydistribution [J]. Acta Electronica Sinica, 2008, 36(4): 667-671.[9]Stentiford F W M. Attention based similarity [J]. PatternRecognition, 2007, 40(3): 771-783.[10]Lai C C and Chang C Y. A hierarchical evolutionaryalgorithm for automatic medical image segmentation [J].Expert Systems with Applications.2009, 36(1):248-259[11]Richards W, Seung H S, and Pickard G. Neural votingmachines [J].Neural Networks.2006, 19(8):1161-1167[12]张鹏, 王润生. 基于视点转移和视区追踪的图像显著区域检测[J]. 软件学报, 2004, 15(6): 891-898.Zhang P and Wang R S. Detecting salient regions based onlocation shift and extent trace [J]. Journal of Software, 2004,15(6): 891-898.[13]He D J, Shao J M, and Gen N, et al.. A model for imagecategorisation based on a biological visual mechanism [J].New Zealand Journal of Agricultural Research.2007, 50(5):781-787

施引文献

资源附件(0)

访问统计