Research On Road Traffic Sign Recognition Method Based On Support Vector Machine

In the process of human history development,tools play a revolutionary role,and the birth of transportation has truly made the earth a global village.Since the invention of the car,it has undergone a step-by-step development.People hope that the car can be as intelligent as human beings,that is,automatically judge the road situation with less human intervention.Based on this,the researchers conducted intelligent research on the road environment in which vehicles and vehicles are located.At this time,Intelligent Traffic Systems(ITS)came into being.The identification of road traffic signs in intelligent transportation systems is a key part of vehicle intelligence.In this paper,the method of road traffic sign recognition based on support vector machine is studied.The research is divided into two parts.The first part is to perform image preprocessing on the acquired road surface image,image image area determination and image edge detection to obtain the target image containing the road traffic sign.The second part consists of three stages.The first stage is to extract the feature of the target image.Firstly,the Hu invariant feature extraction algorithm and the PHOG feature extraction algorithm are studied,and then the Hu invariant moment and PHOG feature extraction method are used.Based on the feature extraction method of Hu invariant moment and low-dimensional PHOG fusion is proposed.Finally,the fused features are taken as the extraction features of the road traffic sign image.The second stage is to build a multi-class support vector machine classifier using a one-to-one method based on the two-class support vector machine.In the process of constructing multi-class support vector machine,the kernel function of support vector machine is first selected.Then5 the penalty parameter C of the support vector machine and the parameter p of the kernel fiunetion are used to optimize the parameters.Finally,based on the selection result of kernel function and the optimization result of support vector machine parameters,the multi-class support vector machine classifier is constructed.In the third stage,this paper constructs a road traffic sign recognition system based on support vector machine and designs simulation experiments to test the recognition accuracy and recognition rate of the system.The test result data shows that the road traffic sign recognition system based on support vector machine constructed in this paper can quickly identify road traffic signs and has higher recognition accuracy.