Title :
Determining edgeness using homogeneity of templates
Author_Institution :
Dept. of Technol., Orebro Univ., Sweden
Abstract :
A fuzzy IF-THEN rule-based method for determining the edgeness of pixels in grayscale images is presented. It is based on the evaluation of template homogeneities and the estimation of the gradient in a local neighborhood. By taking the spatial distribution of the pixels into account, the influence of the image noise is reduced, as shown in a couple of experiments. The complexity of the operations involved is kept simple and the regularity is high, in order to facilitate a hardware implementation of the algorithm.
Keywords :
computational complexity; edge detection; fuzzy set theory; image processing; knowledge based systems; complexity; edgeness; fuzzy IF-THEN rule based method; gradient estimation; grayscale images; hardware implementation; pixels edgeness; reduced image noise; spatial distribution; templates homogeneity; Biomedical imaging; Fuzzy systems; Gray-scale; Hardware; Image edge detection; Noise reduction; Object detection; Object recognition; Pixel; Sensor systems;
Conference_Titel :
Fuzzy Systems, 2003. FUZZ '03. The 12th IEEE International Conference on
Print_ISBN :
0-7803-7810-5
DOI :
10.1109/FUZZ.2003.1206559
