An Image Segmentation Algorithm Based on Artificial Bee Colony and Fast Fuzzy C-means Clustering

摘要：

To prevent the clustering of fuzzy C-means from local convergence, a novel image segmentation algorithm based on Artificial Bee Colony and Fast Fuzzy C-means clustering is proposed. In the algorithm, the original image is decomposed with discrete wavelet transform, an approximation image is reconstructed with low-frequency coefficients and a filtered image is produced by performing a second noise suppression. Then, based on the histogram of the filtered image, we employ Fast Fuzzy C-means clustering to design a fitness function for Artificial Bee Colony algorithm. Finally, by the division of labor and information sharing of employed bees, onlookers and scouts, the optimal solution of image segmentation is discovered quickly. Our experimental results indicate that, when compared to some other methods, the proposed algorithm not only shortens segmentation time, but also improves the performance of segmentation.

关键词： Artificial Bee Colony algorithm Fuzzy C-mean clustering Image segmentation

作者: Miao Ma Hualei Guo Min Guo Jiao He Shengrong Ding

作者单位: School of Computer Science, Shaanxi Normal University, Xian,Shaanxi,China Xian Communication Institute, Xian,Shaanxi,China

会议类型: 国际会议

会议名称: 2010 International Conference on Circuit and Signal Processing(2010年电路与信号处理国际会议 ICCSP 2010)

会议地点: 上海

会议语种:英文

页码: 508-511

在线出版日期: 2010-12-25（万方平台首次上网日期，不代表论文的发表时间）