Abstract:To overcome the over segmentation phenomenon of edgeflow-driven anisotropic diffusion (EFD) and the high computational complexity of normalized cut (NCut) , a color image segmentation algorithm based on EFD and NCut is presented. EFD is applied to the image to get a preliminary result. Then, the segmented regions are taken as nodes to construct a weighted undirected graph G, and the NCut is applied to perform globally optimized clustering. Segmentation results are achieved after proper post-process. The graph structure is based on segmented regions instead of image pixels, and thus the proposed algorithm requires lower computational complexity. In addition, EFD focuses on local detail while NCut captures global property, so this algorithm combines both advantages. Experimental results show that this algorithm can get appropriate segmentation results.
[1] Geraud T, Strub P Y, Darbon J. Color Image Segmentation Based on Automatic Morphological Clustering // Proc of the IEEE International Conference on Image Processing. Thessaloniki, Greece, 2001, Ⅲ: 70-73 [2] Ye Qixiang, Gao Wen, Zeng Wei. Color Image Segmentation Using Density-Based Clustering // Proc of the IEEE International Conference on Acoustics, Speech and Signal Processing. Hongkong, China, 2003, Ⅲ: 345-348 [3] Wan S Y, Higgins W E. Symmetric Region Growing. IEEE Trans on Image Processing, 2003, 12(9): 1007-1015 [4] Falco A X, Udupa J K, Samarasekera S, et al. User-Steered Image Segmentation Paradigms Live Wire and Live Lane. Graphic Models and Image Processing, 1998, 60(4): 233-260 [5] Wu Yong, He Yuanjun, Cai Hongming. Optimal Threshold Selection Algorithm in Edge Detection Based on Wavelet Transform. Image and Vision Computing, 2005, 23(13): 1159-1169 [6] Weszka J S, Rosenfeld A. Histogram Modification for Threshold Selection. IEEE Trans on System, Man and Cybernetics, 1979, 9(1): 38-52 [7] Dong Guo, Xie Ming. Color Clustering and Learning for Image Segmentation Based on Neural Networks. IEEE Trans on Neural Networks, 2005, 16(4): 925-936 [8] Zhu Hongwei, Basir O A. Fuzzy Sets Theory Based Region Merging for Robust Image Segmentation // Proc of the International Conference on Fuzzy Systems and Knowledge Discovery. Changsha, China, 2005, Ⅱ: 426-435 [9] Mumford D, Shah J. Optimal Approximations by Piecewise Smooth Functions and Associated Variational Problems. Communications on Pure and Applied Mathematics, 1989, 42(5): 577-585 [10] Sumengen B, Manjunath B S. Edgeflow-Driven Variational Image Segmentation: Theory and Performance Evaluation [EB/OL].[2009-06-06]. http://barissumengen.com/documents/variational_eval.pdf [11] Ma Weiying, Manjunath B S. Edgeflow: A Technique for Boundary Detection and Image Segmentation. IEEE Trans on Image Processing, 2000, 9(8):1375-1388 [12] Soundararajan P, Sarkar S. Analysis of Mincut, Average Cut, and Normalized Cut Measures // Proc of the 3rd Workshop on Perceptual Organization in Computer Vision. Vancouver, Canada, 2001: 1-4 [13] Shi Jiaobo, Malik J. Normalized Cuts and Image Segmentation. IEEE Trans on Pattern Analysis and Machine Intelligence, 2000, 22(8): 888-905 [14] Arbelaez P, Fowlkes C, Martin D. The Berkeley Segmentation Dataset and Benchmark [EB/OL]. [2009-01-31]. http://www.eecs.berkeley.edu/Research/Projects/CS/vision/grouping/segbench/ [15] Cors T, Yu S, J.Shi. Normalized Cuts Matlab Code [EB/OL]. [2009-06-09]. http://www.cis.upenn.edu/~jshi/software [16] Yang Weili, Guo Lei. Image Segmentation Method Based on Watersheds and Graph Theory. Computer Engineering and Applications, 2007, 43(7): 28-30 (in Chinese) (杨卫莉,郭 雷.基于分水岭算法和图论的图像分割.计算机工程与应用, 2007, 43(7): 28-30)