Using local structure for the reliable removal of noise from the output of the log edge detector

In this paper a method is suggested for enhancing the performance of the Laplacian-of-Gaussian edge detector by reliably removing false edges that are created by noise from its output. The discrimination between false and valid edges is based on the distance between successive edge contours. Statistical analysis of the performance and simulation results are, also, provided.

[1]  D Marr,et al.  Theory of edge detection , 1979, Proceedings of the Royal Society of London. Series B. Biological Sciences.

[2]  S. M. Cobb,et al.  The distribution of intervals between zero crossings of sine wave plus random noise and allied topics , 1965, IEEE Trans. Inf. Theory.

[3]  Steven W. Zucker,et al.  The Local Structure of Image Discontinuities in One Dimension , 1987, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4]  John F. Canny,et al.  A Computational Approach to Edge Detection , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[5]  Ramesh C. Jain,et al.  Reasoning About Edges in Scale Space , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[6]  James J. Clark.  Singularity Theory and Phantom Edges in Scale Space , 1988, IEEE Trans. Pattern Anal. Mach. Intell..

[7]  K. Boyer,et al.  Tissue boundary refinement in magnetic resonance images using contour-based scale space matching. , 1991, IEEE transactions on medical imaging.

[8]  J. A. McFadden.  The axis-crossing intervals of random functions-II , 1958, IRE Trans. Inf. Theory.

[9]  Hong Jeong,et al.  Adaptive Determination of Filter Scales for Edge Detection , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[10]  Hemant D. Tagare,et al.  Localization performance measure and optimal edge detection , 1990, Other Conferences.