Measurement of Safe Driving Distance Based on Stereo Vision

Abstract

The measurement of safe driving distance based on stereo vision is proposed. The model of camera imaging is established using traditional camera calibration method firstly. Secondly, the projection matrix is deduced according to camera's internal and external parameter and used to calibrate the camera. The method of camera calibration based on two-dimensional target plane is adopted. Then the distortion parameters are calculated when the nonlinear geometric model of camera imaging is built. Moreover, the camera's internal and external parameters are optimized on the basis of the projection error' least squares criterion so that the un-distortion image can be obtained. The matching is done between the left image and the right image corresponding to angular point. The parallax error and the distance between the target vehicle and the camera can be calculated. The experimental results show that the measurement scheme is an effective one in a security vehicles spacing survey. The proposed system is convenient for driver to control in time and precisely. It is able to increase the security in intelligent transportation vehicles.