An image segmentation method based on gray stretch in wavelet domain

In this paper, an automatic wavelet domain-based image segmentation method that combines the discrete wavelet transform with the principle of the largest between-class variance is proposed. It not only takes into full account the spatial distribution and contextual relationship between the foreground and background pixels, but also enhances the image contrast and edges, contributing to better target segmentation effect. Experimental comparisons between the proposed method and Otsu´s method show that the proposed approach can better suppress interference and extract the target.