Research on Intelligent Anti-collision System for Automobile Driving Based on Machine Vision

With the rapid development of road traffic and more and more attention to traffic safety, the research of vehicle active safety technology is becoming more and more active. Due to the advantages of low price and high fit with the driver, the auxiliary driving system based on machine vision has become an important research direction of vehicle active safety technology. This paper focuses on the design of early warning system in auxiliary driving system, focusing on vehicle detection and tracking algorithm, lane line detection algorithm and anti-collision early warning algorithm. The number of cars is increasing year after year, but the number of traffic accidents caused by vehicles is also rising. Reducing the number of traffic accidents and improving the level of road traffic safety have become the focus of social attention. At present, the measuring devices in the vehicle anti-collision early warning system basically adopt hardware systems such as ultrasonic, infrared, laser and millimeter wave radar. Some well-known automobile companies around the world use radar installed on the top of the vehicle body to measure the distance of the vehicle in front, but the hardware cost is relatively high and the software algorithm processing complexity is relatively high, It is not conducive to the integration of embedded products and the market promotion of ranging system This paper focuses on the moving target detection algorithm based on machine vision in the research of automobile active anti-collision technology, and compares and analyzes the particle filter algorithm.