Research On The Contour Information Extracting Algorithm Of Small Module Gears In Machine Vision Measurement

The small module gear plays an important role in precision instruments industry, which makes its geometric measurement of significance importance. However, the conventional contact measuring method and equipment can’t meet the quality testing requirements on precision, speed and non-destruction. In order to resolve the problem, the machine-vision measurement has been developed in the field of industrial parts detection. In this dissertation, equipment applying machine-vision measurement techniques to the small gear measurement is introduced, with particular attentions to the research of edge detection of small gears and coordinates transformation of the gear’s tooth profile.The main contents of this dissertation are:(1) A detection system of small module gear based on machine vision is designed. With it, a precision detection platform of small module gear is build through the requirement analysis on modules, and the selection on hardware equipments considering the key parameters.(2) The image preprocessing algorithms suitable for the current experimental environment are studied, in order to move the noise caused by impurity collecting equipment and the environment impacts, so that the useful information of gear images can be retained to the maximum extent and the tooth profile can be extracted easily.(3) The image mosaic algorithm is studied. An algorithm using the triangular geometric constraint relation and the partial autocorrelation analysis of the feature points is proposed for realizing the mosaic of high resolution images of small module gears.(4) A software detection system is developed in Visual C++ 6.0. It realizes the functions of image acquisition, image processing, tooth profile extraction and parameter evaluation. Experiments are carried out on gears with modules of 0.3mm, 0.5mm and 0.8mm. The results show the accuracy of the test in the expected set of ±5μm range.