Color multilevel thresholding using gray-level co-occurrence matrix and differential evolution algorithm
Shreya Pare, Anil Kumar, Girish Kumar Singh · 2017
Color images such as satellite images (remote sensing images) and some natural images exhibit poor fuzziness in boundaries, complex background, and non-uniform illumination. This article presents a new approach intended to produce more sophisticated and reliable segmentation of such color images, which are based on the gray level co-occurrence matrix (GLCM) and differential evolution (DE) algorithm that upholds optimum result during iterations. The proposed methodology augments the robustness and effectiveness of the traditional GLCM algorithm. Performance of the presented technique is evaluated through different tests conducted over various natural and satellite colored images. The simulation results validate the overall accuracy obtained by the proposed GLCM-DE approach, and the methodology was found to be superior to other optimization algorithm based techniques.