Automatic grayscale image segmentation based on Affinity Propagation clustering

• Published in: Pattern analysis and applications : PAA Vol. 23; no. 1; pp. 331 - 348
• Main Authors: Zhou, Shibing, Xu, Zhenyuan
• Format: Journal Article
• Language: English
• Published: London Springer London 01.02.2020; Springer Nature B.V
• Subjects: Affinity; Algorithms; Clustering; Computer Science; Gray scale; Image analysis; Image enhancement; Image segmentation; Object recognition; Pattern analysis; Pattern Recognition; Propagation; Theoretical Advances; Cluster analysis; Image segmentation; Affinity Propagation; Clustering validity index
• ISSN: 1433-7541; 1433-755X
• DOI: 10.1007/s10044-019-00785-4

Image segmentation is an important research subject in the field of image analysis and pattern recognition. Based on the affinity propagation (AP) clustering algorithm, an automatic segmentation method is proposed for grayscale images. The AP algorithm is used for image segmentation to avoid the choice of initial clustering centers and enhance the stability of the segmentation results, and a new index is proposed to analyze the validity of segmentation results and determine the optimal number of segments. Moreover, gray values instead of pixels in gray space are clustered as the dataset to decrease the time complexity of the similarity matrix and validity analysis. Theoretical analysis and experimental results demonstrate the effectiveness of the proposed index and method.