Digital Image Measurement Of Gear Parameters

The gear is an important transmission part that is widely used in machinery, and its precision of has a direct impact on the work performance and service life of the whole machine. At present, there are several kinds of gear measuring equipments, for example gear inspection centers, CMMs, gear testers and so on. However, these equipments are expensive and have strong technology requirements for users and maintainers, and a lot of SMEs can not afford them. With the rapid development of digital image processing technology, the measurement based on image processing, which have advantages of non-contact, high speed, dynamic range, informative and relatively low-cost, can be widely applied to various real-time, on-line and high precision testing. An efficient and effective new method for gear measurement based on image processing is explored in this thesis.In Chapter II, the principle of the image measurement and the software and hardware structure of image measuring instrument are described; We present how to test the gear parameters and how to fit an involute in the Chapter III. The base circle radius, as well as the angle between +Y direction and the line segment connecting the involute starting point and the center of the base circle, is computed through fitting an involute to the points, and addendum circle and tooth root radius are computed using the least-squares fitting to the tooth tip and tooth root circle points. Other parameters and machining errors are computed with respect to the base circle radius, the involute initial angle, the addendum circle radius and the tooth-root radius. Finally, the system is calibrated to get the actual value of each parameter of the gear.An introduction of digital image processing and edge detection algorithms is given in Chapter IV. In gear measurement, the image of a gear is obtained using the CCD camera of a three-dimensional imaging instrument. The noise and interference signals in images are filtered out using median filtering, and the gear boundary with sub-pixel precision is extracted using Zernike moments operator to acquire the edge point coordinates of the tooth profile.In Chapter V, a gear picture is used to explain the data processing and the parameters and machining errors of the gear. A software system has been developed using MATLAB7.0, and it has been tested with a standard involutes spur gear. The experimental results show that it is feasible to measure the gear parameters using image processing methods.Finally, the feasibility of this method is analyzed and open issues for future works are identified in Chapter VI as a conclusion of the thesis.