标题：Research of Edge Detection Based on B-Spline multi-scale Wavelet Transform
作者机构：[Zhang, X]Shandong Univ Technol, Zibo 255049, Peoples R China.
会议名称：International Conference on Information Electronic and Computer Science
来源：2010 INTERNATIONAL CONFERENCE ON INFORMATION, ELECTRONIC AND COMPUTER SCIENCE, VOLS 1-3
关键词：Wavelet Transform; B-Spline Wavelet; Multi-scale Analysis; Adaptive; Threshold
摘要：Multi-scale analysis of wavelet transform solves noise suppression and edge extraction of conflict between the details well. Wavelet transform has the ability to detect local mutations. And signal edge is where the greatest rate of change. So wavelet transform is a good tool for edge detection. B-Spline function is chosen as the wavelet smoothing filter operators. This approach excels in the filter will also remove some weak edges. In order to compensate for this deficiency, an improved B-Spline wavelet edge detection algorithm with adaptive threshold is proposed. According to the computer simulation, this algorithm has a good anti-noise performance and edge localization ability, maintains the weak edge together with noise elimination.