Edge Detection In Ficus Carica Tree Images Using Fuzzy Logic

Edge detection is an essential part of image processing, computer and machine vision. Numerous edge detection methods have been developed in the last years that can be summarized into two basic categories: gradient based and zero-crossing. All classical operators identify a pixel as a particular class by carrying out some series of operations within a mask centered on the pixel under observation. Recent researches have concentrated on the most accurate classification methods that include fuzzy logic, artificial neural networks, etc. This study shows how to detect edges in ficuscarica tree images based on fuzzy set theory. The fuzzy logic for edge detection using membership functions define the degree to which a pixel belongs to an edge or uniform region. The results are compared to classical operators. The proposed fuzzy image-processing algorithm has shown greater accuracy compared to other edge detection techniques, and avoids obtaining double edges.