Truck Size Measurement System Based on Computer Vision

According to the application background of the size measurement of trucks in cement plants, the car size measurement system was developed according to the monocular vision principle. The key frame is extracted by the image difference method, then sub-pixel fine positioning is performed by the Zernike moment method. The edge connection is performed by the Progressive Probabilistic Hough Transform algorithm, and the internal and external parameter matrix obtained by the camera calibration is combined to complete the automatic measurement of the truck’s volume. By comparing with the measured data, the measurement accuracy meets the requirements of the factory. This also verifies the feasibility and accuracy of this measurement method.