Abstract
Auto gearbox is charged with Torque and achieved reverse and cut off the task of power during shutting off the engine temporarily, so that cars can adapt to all kinds of driving conditions. Therefore, the relative motion of the gear pair of auto gearbox is very frequent , the loss of the tooth surface is inevitable as the miles add to the engine. The tooth surface may lead to different damage forms during long-time operation of the gears. In the different damage forms, geometric parameters of the gears will modify when emergence and enlargement of the pitting corrosion which leads to loss of metal materials of tooth surface. Thus influence the life of the gears. Tooth surface pitting area rate is the basis to judge whether or not the gear pitting failure. According to the specification, the tooth surface pitting area shall not be greater than 1/4 of the tooth surface area. This paper presents mechanism and characteristics of gear pitting in auto gearbox, by means of a general way of the digital image processing and pattern recognition technology, with the application of MATLAB software to analyse the gear pitting of the CCD image, showed methods and steps to image acquiring, processing and analyzing, including image cropping, scaling, rotating, grayscaling, binarying and area calculating, etc.; developed a corresponding software codes and GUI interface. In short, giving an area recognition on gear pitting of auto gearbox early and appropritiate measures for improving the service life of gears and equipment operation and the safety of the people in car has a great significance.