基于克隆选择和多重空间构造的彩色图像分割

邓晓政; 焦李成; 杨淑媛; 吴秋逸

doi:10.3724/SP.J.1146.2009.00922

基于克隆选择和多重空间构造的彩色图像分割

doi: 10.3724/SP.J.1146.2009.00922

基金项目:

国家863计划项目(2008AA01Z125，2009AA12Z210)，国家自然科学基金(60703107)，高等学校科技创新工程重大项目培育资金项目(706053)和国家部委科技项目(XADZ2008159，51307040103)资助课题

计量
- 文章访问数: 3295
- HTML全文浏览量: 89
- PDF下载量: 724
- 被引次数: 0
出版历程
- 收稿日期: 2009-06-26
- 修回日期: 2010-05-13
- 刊出日期: 2010-08-19

Color Image Segmentation in a Multidimensional Space Based on Clonal Selection Algorithm

摘要

摘要: 该文提出一种基于克隆选择和多重空间构造的彩色图像分割方法。该方法首先将一些常见的颜色分量分为亮度成分、单频光谱成分、双频光谱成分、多频光谱成分，利用主分量分析技术(PCA)分别对其进行计算，得到最具有识别能力的颜色分量来构造多重颜色空间；然后利用克隆选择算法对样本进行自学习，得到全局最优的聚类中心；最后用得到的聚类中心对图像进行分割。由于该文的方法结合了克隆选择算法的非线性分类能力，通过对彩色图像自适应的构造最佳多重空间，能够快速准确地得到分割结果，克服了传统分割方法使用固定颜色空间以及容易陷入局部最优的缺点。通过实验表明，该方法对彩色图像同质区的分割均匀，边缘保持度好。此外，新方法对彩色图像的亮度和纹理变化不敏感，鲁棒性较强。
- 彩色图像分割 /
- 多重空间 /
- 克隆选择 /
- 主分量分析
Abstract: A novel color image segmentation method is proposed in this paper. Multidimensional space is defined by using the PCA technique to computing the most discriminating color components for a given color image among a set of conventional color spaces. Then, training samples for every region in the color image is selected and these samples is trained by clonal selection algorithm to obtain clustering center of every region. Finally, output the segmentation result according to these clustering centers. Due to the nonlinear classification property of clonal selection algorithm and adaptive definition of a multidimensional space for a given color image, the segmentation result can be obtained accurately and quickly. In experiments, different color images are used to test the performance of the suggested method. The result indicated that this method performs more robustness and adaptability.
- Color image segmentation /
- Multidensional space /
- Clonal selection /
- Principal Component Analysis (PCA)

HTML全文

参考文献(1)

Cheng H D, Jiang X H, Sun Y, and Wang J. Color imagesegmentation: aadvances and prospects[J]. PatternRecognition, 2001, 34(12): 2259-2281.[2]Trussel H J, Saber E, and Vrhel M. Color image processing:basics and special issue overview[J]. IEEE Signal ProcessingMagazine, 2005, 22(1): 14-22.[3]Kurugollu F, Sankur B, and Harmanci A E. Color imagesegmentation using histogram multithresholding andfusion[J].Image and Vision Computing.2001, 19(13):915-928[4]Cheng F Y and Shou-Xian. Automatic seeded region growingfor color image segmentation[J].Image and Vsion Computing.2005, 23(10):877-886[5]Koschan A and Abidi M. Detection and classification of edgesin color images[J].IEEE Signal Processing Magazine.2005,22(1):64-73[6]Hung Wen-liang, Yang Min-shen, and Chen De-hua.Bootstrapping approach to feature-weight selection in fuzzyc-means algorithms with an application in color imagesegmentation[J].Patter Recognition Letters.2008, 29(9):1317-1325[7]马文萍, 焦李成, 尚荣华. 免疫克隆SAR图像分割算法[J].电子与信息学报.2009, 31(7):1749-1752浏览[8]Alain Tremeau. Color classification in a multidimensionalcolor space[C]. IEEE International Symposium on SignalProcessing and Information Technology, Cairo, Egypt, Dec.15-18, 2007: 819-824.[9]De Castro L N and Timmis J. Artificial immune systems: ANew Computational Intelligence Approach[M]. Berlin:Speringer-Verlag, 2002: 15-16.[10]Forrest S, Perelson A S, and Alledn L, et al.. Self-nonselfdiscrimination in a computer[C]. Proceedings of IEEESymposium on Research in Security and Privacy, Oakland,CA, USA, May 16-18, 1994: 202-212.[11]De Castro L N and Von Zuben F J. Learning andoptimization using the clonal selection principle[J].IEEETransactions on Evolutionary Computation.2002, 6(3):239-251[12]Farmer J D, Packard N H, and Perelson A S. The immunesystem, adaptation, and machine learning[J]. Physical D,1986, 2(1): 187-204.[13]焦李成, 杜海峰, 刘芳. 免疫优化计算、学习与识别[M]. 北京:科学出版社, 2006: 12-34.Jiao Li-cheng, Du Hai-feng, and Liu Fang. ImmunologicalComputation for Optimization, Learning and Recognition[M].Beijing: Science Press, 2006: 12-34.

施引文献

资源附件(0)

访问统计