Fuzzy Based Image Edge Detection Using Improved Cuckoo Search Optimization Algorithms

Abstract

Evolutionary algorithms (EAs) are designed to find the best solution by removing the solutions with the lowest fitness. As a result, this research provides an evolutionary algorithm-based edge detection approach. The best filter coefficients and thresholding procedure are obtained using a training dataset consisting of simple pictures and their related optimal edge characteristics. The Cuckoo Search optimization algorithm is an effective mechanism inspired by the brood behavior of cuckoo birds. In this research work, an enhanced cuckoo search algorithm is proposed which improvises the search capability of identifying multi-threshold values for edge detection. An efficient Fuzzy Logic technique is also applied to segment the image based on fuzzy membership functions. The proposed model evaluation shows the significance in different images.