A Measurement Method For Gear Chamfering Based On Binocular Vision

Binocular stereo machine vision measurement system can obtain the depth information of the measured object. The system also can recover the 3D information of the measured object in the scene by simulating the human visual system. The binocular stereo camera may achieve non-contact measurement by capturing the image information of the gear chamfer. Comparing with mechanical contact measurement, Non-contact measurement has high efficiency, not damage to the measured object and so on. Therefore,the study of binocular machine vision gear chamfering measurement has the theoretical and practical value.Based on the principle of binocular machine vision, this paper studies the measurement method of gear chamfering by moving monocular camera:(1)Using two different cameras and considering the impact of the camera during the imaging process distortion, the camera imaging model is established. The internal and external parameters of the two cameras are obtained by Zhengyou Zhang's calibration method. Binocular calibration results can be used for the epipolar rectification process.Then it is transformed into a parallel binocular vision model.(2)Utilizing a movable platform and a monocular camera realizes a binocular stereo vision measurement system as two cameras. Binocular machine vision measurement method based on monocular camera realizes the parallel binocular machine vision by using a mechanical device instead of software. In order to prove the measurement method,it makes the experiments. The experiments include the stability of the experimental device,the moving center distance of the camera and the experiment of measurement and so on.The measurement is validated based on the analysis of experimental data.(3)Based on the feature point matching model, this paper proposes a new method of edge detection, which is based on the study of multiple edge detection algorithms and the actual situation. The depth information of the measured object and the spatial coordinates of the feature points are obtained by matching the feature points. It realizes the requirement of measurement.