Wavelet Transform Based Coronary Blood Vessel Segmentation using Entropy Thresholding

This paper presents a novel method for automatic segmentation of blood vessel in coronary angiographic images. The proposed algorithm composed of two steps, coronary angiogram enhancement process and entropy based thresholding. In the first step, a set of directional basis filters based on dyadic wavelet transform is designed to enhance blood vessels. Gaussian wavelet is used to fix the blood vessel directional information and its associated changes in coronary angiogram and a gradient image is obtained. Secondly the thresholding approach that evaluates 2-D entropy based on gray level gradient co-occurrence matrix is used to segment blood vessel from the background. The result of this method promises the simpleness and flexibility in many image enhancement and segmentation applications.