Steel tape is a commonly used length-measuring tool, which needs to be verified before using. In this paper, a method based on image processing for steel tape verification, which replaces the traditional eye detection, is proposed. Firstly, the camera is calibrated to get the pixel equivalent. Secondly, the standard steel tape and steel tape to be measured are arranged side by side on the calibration table, and the image of the integral meter is collected by the CCD camera. Thirdly, the acquired image is divided into regions, then the fine regions are further divided, and the area where the integral meter scale line locates is selected by the frame. Finally, the coordinates of the scale line at the integral meter are extracted and calculated by the method of pixel traversal and gray center of gravity. The error of the steel tape can be obtained by multiplying the difference between them by the pixel equivalent. The verification process is efficient and the results are reliable. The resolution of the measurement system can reach 0.04 mm, which meets the requirements of the measurement task.
With the continuous improvement of equipment measurement accuracy and production efficiency requirements, calibration method with external reference standard can no longer meet the quality and efficiency requirements. In order to solve the online calibration problem and effectively improve the calibration efficiency of the Articulated Arm Coordinate Measuring Machines (AACMMs) in practical application, a self-calibration system of circular grating angle sensor which is applied to joints of the AACMMs was established. Based on the harmonic analysis of the angle measurement error, this paper deduces and analyzes the error suppression principle of the layout of the scanning heads on the calibration result, and establishes a non-uniform layout of the scanning heads to eliminate more and higher order harmonic errors. The simulation and test results show that the self-calibration method using this layout form of multiple reading heads can effectively reduce the measurement angle error without increasing the number of scanning heads, and improve the calibration efficiency and measurement accuracy of AACMMs.