Rapid vehicle distance measurement optimization method for vehicle-mounted binocular system

Abstract

The invention discloses a vehicle distance measurement method based on a vehicle-mounted binocular camera. The method is characterized by comprising steps as follows: S1, vehicle detection is performed at the angle of view on one side of the binocular camera according to target vehicle texture information and prior road information, the position of a vehicle and the distance away from the vehicleare determined; S2, similarity matching of the target vehicle at the angle of view on the other side of the binocular camera is performed, a matching position with higher similarity is selected, the center offset of the target vehicle at two angles of view is determined, and the actual position of the target vehicle and the distance away from the vehicle are determined according to the center offset; S3, a confidence interval of the binocular camera is determined according to parameters of the binocular camera, the actual position of the target vehicle and the distance away from the vehicle, which are determined according to the center offset, are output in the confidence interval of the binocular camera, and the vehicle detection reliability is verified through an offline training classifier outside the confidence interval. With the adoption of the method, the vehicle distance measurement accuracy is improved.

Description

technical field [0001] The invention relates to a vehicle distance measurement method, in particular to a rapid vehicle distance measurement optimization method for a vehicle-mounted binocular system. Background technique [0002] Advanced Driver Assistance System (ADAS) is one of the important technologies for the development of vehicle braking safety, and has gradually become an important safety reference for NCAP in various countries to evaluate vehicle safety levels. ADAS functions such as Forward Collision Warning (FCW) and Lane Departure Warning (LDW) have become the standard configuration of existing mid-to-high-end models. For ADAS systems such as FCW and Automatic Emergency Braking (AEB) for vehicles in front, not only require The perception system can detect the vehicle ahead in a timely and accurate manner, and it is also required that the system can accurately predict the movement state of the vehicle in front relative to the vehicle in order to ensure the effect...

AI Technical Summary

Problems solved by technology

[0003] The existing ranging method based on binocular vision system is mainly based on the parallax calculation between two cameras to estimate the distance of obstacles in the scene. The axes are parallel, the acquisition is synchronized and the exposure parameters are consistent, etc., and the practical application has great limitations

Images