Adaptive fuzzy edge detector for image enhancement
Chang-Shing Lee, Yau-Hwang Kuo · 2002
A novel adaptive fuzzy edge detector for image enhancement, which can work well in full range of random impulse noise probability and perform efficiently in the environment of mixed Gaussian impulse noise, is proposed. It is an extended adaptive weighted fuzzy mean (EAWFM) filter, which combines adaptive weighted fuzzy mean filter and fuzzy normed inference system to efficiently perform edge detection in smeared images. The membership functions of all fuzzy sets used in EAWFM can be adaptively determined for different images, and EAWFM filter is capable of converting blurred edges to clear ones and suppressing noise at the same time. The important properties of EAWFM filter are analyzed and some experimental results are presented to show its excellent performance.