Research On The Consideration Of Dimension Of Gear Based On Machine Vision

Based on the theory of machine vision, combined with the related knowledge of image processing technology, the main focus of this thesis was to design and apply image processing algorithm, and investigated the realization of the automatic measuring method for gear dimension measurement. The main content and achievements can be summed up as follows:(1) The development of machine vision, the technology of gear measurement was summarized, respectively. The important scientific significance and practical value of this thesis were introduced too.(2) A non-contact measuring plan based on machine vision for spur gear was proposed, and then the hardware system was built with fulfilling requirement for measurement.(3) Adopting the method of the maximum between-cluster variance to give the gray threshold; after comparing the advantages and disadvantages of typical edge detection operators, the mathematical morphological method was employed to detect edge, the outline of the edge was repaired according to the toward of the image.(4) The calibration of the camera was completed by the MATLAB calibration toolbox through the research on the model of the pinhole camera and camera calibration method, and then using the camera intrinsic parameter, the image rectification transforms the image to compensate lens distortion.(5) The algorithm for the gear parameters measurement was put forward, especially, the algorithm of the keyway depth and width is introduced, and the experimental results were given by the dimension drawing. By analyzing the main factors affecting the measurement accuracy of measurement system, makes the comprehensive evaluation.The technigue proposed in this thesis had significant science and value of application; it can greatly reduce the labor intensity, increase production efficiency. The results of this thesis were expected to replace the traditional detection technique with a broad application prospect in the market.