Parameters Optimization of Region growing Segmentation Based on Differential Evolution algorithm

Image segmentation is the first step of GEOgraphical Based Image Analysis (GEOBIA). Because of no obvious mathematical relationship between the parameters of segmentation algorithm and the optimal segments, users select parameters by "trial and error" method, which is time-consuming and subjective. This paper used the differential evolution algorithm to automatically optimize the parameters of region growing algorithm. Because of differential evolution algorithm is a heuristic optimization method, use this method to select the parameters of segmentation for optimal segments is objective and efficient. The experiment result shown that use this method to select the optimal segmentation parameters can get the accuracy result and improve the efficiency.