Title:
FuzzyEntropyBasedPost-ProcessingMethodforC-MeanClu
Download
Description: This paper proposes a combination of C2 average clustering algorithm and fuzzy entropy image segmentation method, this method first USES the C2 average clustering algorithm to initial segmentation of noise images, using the fuzzy entropy criterion for subsequent processing. The method on the one hand able to inherit the advantages of C2 average clustering algorithm, based on the flexibility to use more features and more in threshold image segmentation, on the other hand, give full consideration to the regional information of source image, as the minimum fuzzy entropy criterion, the C2 average clustering algorithm of the segmentation results of a preliminary classification of fault point for further processing, overcome the C2 average clustering algorithm is sensitive to noise shortcomings. The experimental results show that the method is only 4 ~ 6 s more than the C2 mean clustering algorithm, and the image of low signal-to-noise ratio can achieve better segmentation than the C2 mean clustering algorithm.
To Search:
File list (Check if you may need any files):
