Multi-threshold image segmentation with improved quantum-inspired genetic algorithm

In this paper, a method of multi-threshold image segmentation was proposed using the principle of maximum entropy and an improved quantum-inspired genetic algorithm (IQGA). With the increase number of multi-threshold, it is unrealistic to compute the entropy of all possible combinations and find the maximum entropy in all the multi-threshold combinations for images segmentation. Quantum-inspired genetic algorithm (QGA) has a better characteristic of population diversity, rapid convergence and global search capability than that of the conventional genetic algorithm (CGA). However, the solutions of QGAs may diverge or have a premature convergence to a local optimum due to the selection of the rotation angle in searching the maximum value of a function. Therefore, IQGA is put forward which joins the optimal selection and catastrophe operations, and defines an adaptive rotation angle of quantum gate during quantum chromosomes update procedure. Experimental results demonstrated that the proposed method has a good performance.