Research on Reversing Radar Based on Linear Structured Light

Abstract

Aiming at the problem that the existing reversing radar detection method can not accurately detect obstacles, a detection method based on structured light is proposed. Remove the noise point by median filtering the strip image, and the strip area is obtained by using the characteristics of the line structure light in the RGB space and the gray space. The gray center of gravity method is used to extract the initial center of the light strip. Based on this, the Hessian matrix is used to extract the sub-pixel center of the strip, which reduces the number of calculations of Gaussian convolution and improves the extraction speed of the sub-pixel center of the strip. The distance, width and angle information of the obstacle are determined by calculating the light bar parameters. The experimental results show that the reversing radar based on line structure light can meet the real-time and accuracy of detecting obstacles in reversing.