Subpixel edge detection method based on low-frequency filtering

The paper presents a method of edge detection in images that based on low-frequency filtering. The method uses polynomial interpolation in order to determine the coordinates of the edge point with sub pixel accuracy. Comparing to other edge detection methods, it does not use numerical differentiation. The method is based on finding of a shared point of intersection of two boundary curves received as a result of low-frequency filtering. The low-frequency filtering is performed with the help of Gaussian filter with specific weighting coefficients. The presented algorithm was assessed experimentally.

Author: Andrzej Kotyra
Conference: Title