원문정보
초록
영어
FCM is one of a conventional clustering method and has been generally applied for medical image segmentation. On the other hand, conventional FCM at all times suffers from noise in the images. Even though the unique FCM algorithm yields good results for segmenting noise free images, it fails to segment images corrupted by noise, outliers and other imaging artifact. The most important shortcoming of standard FCM and FPCM algorithms are that the objective function does not think about the spatial dependence therefore it deal with image as the same as separate points. In order to decrease the noise effect during image segmentation, the proposed method incorporates spatial information into the FPCM cluster algorithm. The proposed algorithm is applied to both artificial synthesized image and real image. Segmentation results demonstrate that the presented algorithm performs more robust to noise than the standard FCM and FPCM algorithm.
목차
1. Introduction
2. Algorithms
A. Traditional Fuzzy C-Means
B. Fuzzy possibilistic c-means algorithm
C. Spatial Fuzzy C-Means (SFCM)
3. A modified fuzzy possibilistic c-means algorithm with spatial information
4. Experimental Result
A. Cluster validity functions
B. Experimental results on synthetic image
C. Experimental result on real image
5. Conclusions
References