Parameter selection of generalized fuzzy entropy-based thresholding method with quantum-behavior particle swarm optimization

Image thresholding method based on generalized fuzzy entropy segments the image using the principle that the membership degree of the threshold point is equal to m (0 < m < 1), better segmentation result can be obtained than that of traditional fuzzy entropy method, especially for images with bad illumination. The main problem of this method is how to determine the parameter m effectively. In this paper, based on the advantages of Quantum-Behavior Particle Swarm Optimization(QPSO) in few parameters and guaranteeing global convergence, we proposed an algorithm to select the parameters of generalized fuzzy entropy. Using an image segmentation quality evaluation criterion and the maximum fuzzy entropy criterion, the optimal parameter m and the membership function parameters (a, b, d) are automatically determined respectively by QPSO, realizing the aim of automatic selection the threshold by generalized fuzzy entropy-based image segmentation method. Experiment results show that our method can obtain better segmentation results than that of traditional fuzzy entropy-based method.