Guohua Cao

A Study on Analysis Method of Motion Characteristics in the Crash Test Based on Computer Vision

The actual vehicle crash test is the most basic and effective method in the development and validation of comprehensive vehicle safety performance. In the actual crash test, the optical measurement system can significantly record the motion form change through images on each target point. The image after-treatment can be performed by using the photographic measurement technology based on computer vision. The accurate motion characteristic information of the target points can be obtained through bundle adjustment of a series of frames, 3D photogrammetry and target tracking, as well as related calculation. The experimental results conducted at Zhejiang Key Laboratory of Automobile Safety Technology indicate that this method can particularly describe the motion characteristics of bodywork and dummy in the crash test with high sensitivity, quick speed and more data acquisition, thus satisfying the development requirements of vehicle safety performance.