A new edge detection algorithm based on a statistical approach
Mehdi N. Fesharaki, G.R. Hellestrand · 2002
A method of generating edge maps thereby segmenting an image into regions based on the Student t-test is presented. By applying this test to compare the distribution functions of the intensities in the neighborhood of a given pixel, the pixel can be accurately classified as either an edge pixel on the boundary between regions, or as a pixel of a particular type of regions. The results show that using 5/spl times/5 window, this method robustly segments both synthetic and natural images, and maintains this performance in the presence of additive noise.>