Research On Pulse Coupled Neural Network In Image Segmentation Based On Quantum Genetic Algorithm

As a new generation of neural network,Pulse coupled neural network (PCNN) has its unique pulse properties and has been studied by many scholars. Because of the values of parameters in model have a great effect on image segmentation, research on parameters is very necessary.Firstly, the PCNN is summarized based on image segmentation technique; secondly, the quantum rotation gate angle and quantum mutation of the quantum genetic algorithm is improved, while introducing the simulated annealing algorithm, the mixed improving double chains quantum genetic algorithm is proposed (SA-IDCQGA),it will be carried on the analysis and the proof of the convergence, through the optimization performance compared with the common quantum genetic algorithm and the double chains quantum genetic algorithm, the simulation results analysis shows that SA-IDCQGA has better searching ability and faster convergence speed, and the robustness and practicability is good. Finally,SA-IDCQGA algorithm is proposed in this paper to be applied to the segmentation of PCNN images, by optimizing the parameters in the model to achieve the purpose of automatic segmentation, omits the artificial set parameters of the complex, the simulation results of MATLAB,and the maximum entropy image segmentation and OTSU image segmentation method, the effect of the optimization of SA-IDCQGA PCNN in image segmentation is better than the other two methods.