Tóm tắt:

This study aims at developing a metaheuristic based image thresholding method via the employments of the History-Based Adaptive Differential Evolution with Linear population size reduction algorithm (LSHADE) and the Otsu objective function. The LSHADE algorithm is used to determine an appropriate set of threshold levels that maximizes the separation between clusters of pixels. Experimental results with four applications demonstrate the usefulness of the newly developed tool.