In view of the continuous development of manufacturing industry,high accuracy,high speed and low error are required for the detection technology.A detection algorithm for workpiece size and scratch in industrial manufacturing was presented.The algorithm was based on the combination of OpenCV vision library and industrial camera,and operated under the appropriate light source platform to extract the outline size of the workpiece and surface scratches.Then camera calibration parameters were used to solve the actual size of the workpiece.The detection results show that the proposed detection algorithm has a smaller error than the traditional Hough transform detection algorithm,and the scratch detection is very accurate.