Layer extraction technique applied to edge detection of MR images with truncation artifacts

In this paper, we propose a new edge detection technique to fit to MR images with truncation artifacts. According to singularity function analysis theory, any truncation artifact digital signal can be expressed as the weighted sum of truncated Singularity function related to the signal singularity locations, and the singularity locations and weighted coefficients can be determined by this signal. Using layer extraction technique, singularity locations in MR images with truncation artifacts can be detected one by one in order of their singularity degrees from big to small. Edges of MP images with truncation artifacts can be constructed by these singularity locations. False edges caused by truncation artifacts can be removed using this method. Experiment and simulation results show that this method gives much better performance than other available methods