Improved Grey Wolf Algorithm And Its Application In Face Recognition

With the rapid development of the swarm intelligence algorithm in the past20 years,it is widely used in different fields due to its strong robustness and simplicity.In 2014,Mirjalili et al.proposed a new swarm intelligence bionic algorithm,Grey Wolf Optimizer(GWO),which is implemented by simulating the leadership level and hunting mechanism of the wolves,compared to other intelligence.Optimization algorithms,such as particle swarm optimization(PSO)and differential evolution(DE),have the advantages of less parameters and high convergence precision.However,the algorithm still has the shortcomings of traditional intelligent optimization algorithms,and it is easy to fall into the early local optimal solution and convergence.The problem is slower,so the gray wolf algorithm is still in the research stage of improvement,and the application range is not wide.In summary,this paper makes improvements to the shortcomings of the algorithm,and applies the improved algorithm to the face recognition by optimizing the kernel parameters and penalty parameters of the support vector machine.The main research contents of the thesis are as follows:In order to solve the shortcomings of grey wolf algorithm,this paper proposes an improved grey wolf algorithm.Firstly,the parameters of exploration ability and development capability are improved in grey wolf algorithm,and the nonlinear dynamic convergence function is used to replace the linear convergence function in the original algorithm.The local search ability of the algorithm is improved,and then the dynamic global value strategy is used to accelerate the global search ability of the algorithm and the convergence of the local search ability through dynamic balance adjustment,so as to seek the global optimal solution.After comparison of different benchmark function types,the experimental results show that the improved gray wolf algorithm has the advantages of higher precision and faster convergence than the comparison algorithm.In terms of face recognition,this paper first uses two-dimensional principal component analysis(2DPCA)and principal component analysis(PCA)to extract features,and then puts the extracted feature matrix into the improved gray wolf algorithm optimized classifier.Support vector machine(SVM),classification processing,through the ORL face database and Yale face database experimentalcomparison,prove that the optimized support vector machine is better than other comparison algorithms in classification performance.