Research on Measurement Method of Tread Pattern Groove Depth Based on Computer Vision and Structured Light Measurement Technology

This paper presents a fast and convenient method to detect the tread pattern groove depths, which is based on computer vision and structured light measurement technology. Firstly, the intrinsic parameter matrix of the camera needs to be calibrated. Secondly, the centerline pixel coordinates that the centerline of the structured light stripe on the calibration target are extracted by irradiating the structured light to the calibration target. Thirdly, the camera coordinates are calculated by extrinsic parameters. The RANSAC algorithm is used to fit the a structured light plane in the camera coordinate system. Finally, the groove point is located by calculating the gradient change in the neighborhood of the pixel point, and the groove depth is calculated by the distance between the groove point and the calculation baseline. The experiment results show that the precision of this method meets the real-time measurement requirements.