Researching On Pattern Recognition Problems Based On Wu Method

Wu Method solves a system of polynomial equations, which is well applied in Mathematic and nonlinear engineering problems,and makes our country keep ahead in the research on Auto Reasoning and Mathematic Mechanics since its creation.Key issues of Pattern Recognition are investigated. Aiming at these issues, the way of applying Wu Method to Pattern Recognition is discussed in theory, and the solution of specific classification problems is also discussed, which provide a new way to deal with nonlinear pattern classification problems.The main work of the dissertation is that a new pattern classification method called Wu Classification Method based on Wu Method is proposed. The method complies with the Structural Risk Minimization principle and constructs a uniform structure of classifier functions, whose forms and coefficients can be regulated automatically to get the best one. Finally experiments are provided to validate the successful application of Wu Classification Method in nonlinear Pattern Classification problems, and the superiority of this method is proved by the results.According to the dissertation, Wu Classification Method based on the principle of Structural Risk Minimization is an effective classification way. By choosing classification form automatically, the method can solve a certain nonlinear pattern classification problem, which is provided with a bright prospect with applications and developments.