A multivariate nonlinear dynamic programming of female figures in Qing dynasty literature

Abstract

Abstract To improve the recognition of female images in Qing Dynasty literature, this paper designs a multivariate nonlinear-based image dynamic programming model. Data-driven typical correlation analysis is used to extract the key controllable variables with the strongest correlation with female image features and use them as input variables for modeling so as to make the maximum correlation between different image features and improve the modeling efficiency. The nonlinear spatial state model is established by using the subspace identification method, and the female image planning model is substituted into the multivariate nonlinear analysis, and the parameters of the nonlinear fitting function about the image planning can be obtained. The simulation analysis of the image dynamic planning model based on multivariate nonlinearity is carried out, and the results show that the average annual failure time of the model designed in this paper is 12.6 hours, and the classification accuracy in female image classification is 96.3%, and the classification time is stable at 1.162 s. The results show that the multivariate nonlinear features of the model can carry out dynamic planning of female images in Qing dynasty literature and improve the classification of image feature recognition.