Advances in Electrical and Computer Engineering | |
Application of Rosette Pattern for Clustering and Determining the Number of Cluster | |
SADR, A. ; MOMTAZ, A. K.. | |
关键词: clustering; Fuzzy C-means (FCM); pattern recognition; Rosette Pattern; validity index; | |
DOI : 10.4316/AECE.2011.03013 | |
学科分类:计算机科学(综合) | |
来源: Universitatea "Stefan cel Mare" din Suceava | |
【 摘 要 】
Clustering is one of the most important research topics which has many practical applications such as medical imaging and Non-Destructive Testing (NDT). Most clustering algorithms like K-means, fuzzy C-Means (FCM) and their derivatives require the number of clusters as one of the initializing parameters. This paper proposes an algorithm for image clustering with no need to any initializing parameter. In this state-of-the-art, an image is sampled based on a rosette pattern and according to the pattern characteristics, the extracted samples are clustered and then the number of clusters is determined. The centroids of classes are computed by means of a method based on calculation of distribution function. Based on different data sets, the results show that the algorithm improves the capability of the clustering by a minimum of 62.26% and 87.62% in comparison with FCM and K-means algorithms, respectively. Moreover, in dealing with high resolution data sets, the efficiency of the algorithm in clusters detection and run time improvement increases considerably.
【 授权许可】
Unknown
【 预 览 】
Files | Size | Format | View |
---|---|---|---|
RO201902181761996ZK.pdf | 1242KB | download |