Abstract:With the rapid development of artificial intelligence technology,intelligent vehicle assistant driving system has gradually entered people′s life.Although the intelligent vehicle assistant driving system based on visible light imaging is becoming more and more mature,it is limited by the visible light imaging conditions.In the dark,rain and fog circumstances,the effect of utilization may be greatly reduced or even unable to use.Infrared intelligent vehicle assistant driving system is widely concerned,because it relies on infrared imaging equipment,which makes it an important means to solve this problem.However,in some special cases,such as sudden acceleration of the vehicle and great changes in the application scene,the real-time performance of the detection and imaging of the infrared intelligent vehicle auxiliary system may be affected,which may cause the display picture to jam.The above situation will seriously affect the driver′s judgment and even driving safety.In view of the above situation,this paper proposes an adaptive frame detection algorithm,which can adjust the number of detection intervals to ensure the real-time performance of detection and imaging.Experimental results show that the algorithm is effective and practical for improving the speed of video detection.