期刊名称：International Journal of Signal Processing, Image Processing and Pattern Recognition
印刷版ISSN：2005-4254
出版年度：2015
卷号：8
期号：8
页码：89-98
DOI：10.14257/ijsip.2015.8.8.10
出版社：SERSC
摘要：This paper presents a novel image segmentation method that performs histogram thresholding based on the conception of Tsallis generalized divergence. Firstly, to fit the image segmentation task, the original formula of Tsallis divergence was simplified, and then the symmetrical version was constructed. After that, the criterion of divergence sum of the objective and background between original and thresholded image was set up based on the symmetrical version of the Tsallis divergence. The optimal threshold obtained by minimizing the criterion of divergence sum. Finally, the proposed method was tested on different gray level images, and the performance was evaluated using uniformity measure, shape measure, and CPU run time. Experimental results indicate the effectiveness of the proposed method
