Typhoon image segmentation by combining curvelet transform with continuous wavelet transform

Abstract

Discrete curvelet transform (DCT) is done to the original typhoon cloud image. The global contrast is enhanced in the low frequency components based on in-complete Beta transform and genetic algorithm (GA) in the DCT domain. Then the enhanced typhoon cloud image is used to segment the main body of the typhoon from the typhoon cloud image. In order to reduce the effect of noise, the gray-level histogram is smoothed by Bezier curve and Bezier histogram is obtained. One dimension continuous wavelet transform is done to the curvature curve of Bezier histogram. The gray level values, which are located in adjusted peak values at all scales, is considered as segmentation thresholds. The gray level values of valley between peaks are considered as quantity gray levels. Finally, optimal segmentation scale is obtained by a cost criterion. Experimental results show that the main body of a typhoon can be segmented effectively from complex background in the typhoon cloud image by proposed approach.