Image edge detection based on fuzzy weighted morphological filter

A novel fuzzy weighted morphological filter is proposed for image edge detection, which based on weighted morphological operators and fuzzy morphological structuring elements (SE). It divided the filter window into four sub-windows, and selects the least variance sub-window and its mean of pixels is given. If the difference value between the pixel value and the mean is small, a big weight factors in SE is put; if the difference value is big, a small weight factors in SE is put. Experimental results prove that the new operators' performance dominates those of classical operators for images in edge detection, and obtains superbly detail edges.