Paper Info
Title
A robust fuzzy c-means algorithm based on diffusion equation for sar image segmentation
Abstract
Fuzzy c-means (FCM) algorithm and many modified ones have been widely used in image segmentation. But these methods are not adaptable to SAR images owing to the intrinsic speckle noise. In order to improve the noise-resistibility and the detail-preserving in SAR image segmentation, we propose a robust FCM algorithm based on diffusion equation (FCM DE). Firstly, the SAR image is diffused based on the partial differential equation to generate an auxiliary image, which is robust to speckles. Secondly, to make the algorithm more robust to outlier, the maximum probability of the local gray-level histogram is used to design an adaptive factor to adjust the effect of the diffusion term automatically. Moreover, this method can be extend to other PDE-based noise removal approaches and applied to other kinds of images, such as MR images and optical remote sensing images. Experiments on the simulated and real SAR images demonstrate the efficiency of FCM DE compared with other five fuzzy clustering algorithms in SAR images segmentation.
Year
DOI
Venue
2017
10.1109/CISP-BMEI.2017.8301976
2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)
Keywords
Field
DocType
SAR,segmentation,diffusion equation,fcm
Fuzzy clustering,Histogram,Synthetic aperture radar,Computer science,Image segmentation,Robustness (computer science),Artificial intelligence,Speckle noise,Cluster analysis,Computer vision,Pattern recognition,Segmentation,Algorithm
Conference
ISBN
Citations 
PageRank 
978-1-5386-1938-4
0
0.34
References 
Authors
7
3
Name
Order
Citations
PageRank
Ling Wan141.47
Tao Zhang242.82
Hongjian You310317.44