【作者】 王占江； 盛春冬； 孙惠杰；

【Author】 WANG Zhanjiang1,2,SHENG Chundong1,SUN Huijie1,3 (1. College of Science,Harbin Engineering University,Harbin 150001,China; 2. Unit 91796,Huludao 125000,China; 3. College of Computer Science and Information Engineering,Harbin Normal University,Harbin 150080,China)

【机构】 哈尔滨工程大学理学院； 辽宁省葫芦岛市91796部队； 哈尔滨师范大学计算机科学与信息工程学院；

【摘要】 为了克服基于FCM算法的图像分割技术对噪声敏感和运算效率低等缺点,利用KFCM算法对于噪声的鲁棒性质,以及模糊退火算法Metropolis准则和粒子群算法相结合的智能全局搜索能力,改进了图像分割技术。实验表明:该算法具有一定的降噪和全局搜索能力,提高了运算速度及图像分割效果。该算法对于丰富图像分割研究具有一定的参考价值和指导意义。更多还原

【Abstract】 A new fuzzy algorithm based on SAPSO and KFCM is presented in this paper,which overcomes the shortcomings of the local optima and sensitivity to initialization from fuzzy C-means algorithm (FCM). The new algorithm takes advantage of the robustness of KFCM,the capacity of global search in PSO algorithm and the ability of jumping out of the local optima in SA to improve the FCM method. An experiental study demonstrates that the algorithm presented avoids the local optima and increases the convergence speed. The study is a valuable reference and of significance for image segmentation.更多还原