Image Sensitivity Detection Algorithm For Intelligent Vehicle Visual Perception System

With the development of science and technology,the information age has come to the human side,in the rapid development of scientific and technological information,people have been in the process of information transmission using the image as the main information carrier,in the complex information,how can we quickly and efficiently To obtain the value of the image information,has become one of the focus of the current research.In many living scenes,the traffic environment belongs to the multi-faceted,complex and real-time changes.The main contents of this paper are based on the knowledge of human visual perception theory to analyze the image of the traffic environment,According to its significance and superiority in image feature recognition,this paper combines the selective attention mechanism and the control strategy in the advanced cognitive process,and applies it to intelligent analysis of traffic scene.The safe driving of smart cars not only needs to recognize and perceive various elements such as traffic roads,traffic flow,traffic signs,traffic facilities,both sides of the environment,and emergencies,but also further carry out these elements from a cognitive perspective.Purposeful choice and cognitive integration in order to take appropriate measures to make correct judgments.This dissertation is based on the theoretical research of intelligent vehicle detection algorithm for road traffic image in traffic environment.It mainly studies and analyzes the saliency map model proposed by ITTI,and integrates the traffic image based on the analysis of the advantages and disadvantages of ITTI algorithm.Features,improved and optimized ITTI algorithm to make it more efficient to deal with the significant image of traffic environment.By using the traffic image for simulation experiments and comparing the experimental results,it can be seen that the improved algorithm has a clearer effect,and that the improved algorithm has a better matching degree through the Roc curve.Finally,the improved algorithm is compared with the common excellent algorithms SR,RC,GB,MZ,HC and CA.According to the experimental results,it can be seen that the edge contours of the salient images obtained by using this method are clearer and the internal information is more abundant.Better integrity.This provides a good foundation for the application of visual perception systems in the future.