Fuzzy Logic Approaches in the Task of Object Edge Detection

The task of reducing the computational complexity contour detection in images is considered article. solution to achieved by modifying Canny detector and number passes through original image. In first case, two are excluded when determining adjacency central pixel with eight adjacent ones a frame size 3х3. second three excluded, as case third one necessary determine angle gradient direction. This passage provided combination fuzzy rules. goal work increase performance operations process detecting edges objects edge carried out some replacement most complex procedures. proposed methods, fuzzification input variables after its difference between cells Then base rules built. method, four pass depending on sixteen themselves set direction, while eliminating along cell makes it possible take into account shape distribution. Then, based center gravity resulting variable defuzzified. Further use a-cut binarize image selection object it. presented experimental results showed that noise level depends value parameters labels trapezoidal membership functions. software was developed evaluate methods. limitation methods piecewise-linear Experimental studies approaches have shown time method 18% faster compared 2% than method. However, during visual assessment, found better determines objects.