A Computational Approach to Edge Detection
IEEE Transactions on Pattern Analysis and Machine Intelligence
Characterization of Signals from Multiscale Edges
IEEE Transactions on Pattern Analysis and Machine Intelligence
Construction of wavelets for width-invariant characterization of curves
Pattern Recognition Letters
Hi-index | 0.00 |
We offer a general framework for multiscale edge detection including both the extrema detection and the zero-crossing detection based on B-spline theory. Fast algorithms for multiscale decomposition and reconstruction are given. Especially the impulse and frequency responses of the corresponding decomposition and reconstruction filters are derived explicitly. These results also make contribution to the further application of stereo vision matching, denoising, etc.