HMT Model and B-Spine Wavelet Based Intelligent Medical Image Edge Extraction Algorithm

A new medical image edge extraction algorithm is proposed to extract the significant edge of the medical image to avoid the interference of the internal noise and the external noise. The information measure combined with simulated annealing algorithm is used to remove the internal noise; then wavelet domain HMT model is used to denoise; at last we design quadratic B-spline wavelet to extract the edge feature information. The quadratic B-spline smoothing filter performing multi-scale filtering is designed using the characteristics of wavelet transform. The obtained multi-scale wavelet transform is combined to extract image edge at different scales. The experiment results of computer simulation show that the proposed edge extraction algorithm provides the best compromise between noise rejection and accurate edge localization.