SAR Image Segmentation Based on SWT and Improved AFSA

In order to speed up the segmentation procedure and solve the problem of noise-sensibility in image segmentation, the paper suggests a fast SAR (Synthetic Aperture Radar image) image segmentation method, which integrates SWT (Stationary Wavelet Transform) and AFSA (Artificial Fish Swarm Algorithm). In the method, an original image is decomposed by multilevel SWT firstly. And then, approximation coefficients at the highest level are used to reconstruct the original image. Next, two-dimensional histogram of the reconstructed image and its mean-filtered image is produced, whose trace of the between-class scatter matrix is taken as the fitness function of our improved AFSA. Additionally, the method gets a faster convergence successfully by adjusting the strategy of keeping the best fish. Experimental results indicate that the proposed method has obvious improvement on segmenting speed and segmented effect.