A Fuzzy Clustering Algorithm Based on the Splitting and Lumping Method for Image Segmentation

作者：Liu, Wenping^*; Hung, Chih-Cheng; Chen, Shihong; Cui, Tianyi

来源：Information-An International Interdisciplinary Journal, 2012, 15(8): 3499-3509.

• Image Segmentation
• Pixels Clustering
• Fuzzy C-Means
• Splitting and Lumping

This paper proposes an adaptive fuzzy clustering algorithm (A-FCM), which integrates the fuzzy concept from the fuzzy C-Means algorithm (FCM) with the mechanism of splitting and lumping from the ISODATA. The proposed algorithm provides an adaptive capability for image segmentation. In other words, unlike the FCM, the number of clusters need not be known a priori in the proposed algorithm. Experimental results demonstrate that the A-FCM has better performance than those of the Possibilistic C-Means, ISODATA, and FCM algorithms.

• 出版日期2012-8
• 单位北京林业大学; 北京联合大学