Investigation On The In-site Measurement On The Parameters Of A Spur Gear Based On Machine Vision

Gear is one of the most important parts in mechanical transmission.Improving the measurement technology level of gear is one of the key factors to improve the manufacturing quality of gear production.It is the essential requirement of intelligent production to measure the errors of gears in mass production on line quickly.It is of great significance and feasibility to measure the machining errors of spur gears in non-contact way with machine vision technology.In this paper,the image of spur gear is collected by industrial camera,and the image preprocessing method,morphological operation method and threshold segmentation method are introduced in detail.The classical pixel level edge detection operators such as Sobel operator,Prewitt operator,Laplace operator and Canny operator are used to detect the edge of the preprocessed gear image.The results show that the pixel level edge detection method can not meet the accuracy requirements of gear measurement,so the paper uses the Zernike moment subpixel edge detection algorithm for reference,and combines the maximum class variance method of image histogram to process the best global threshold value of the online straight gear image,which improves the efficiency of gear edge detection and reduces the error at the same time.On this basis,a series of algorithms are designed and developed,and the basic parameters and pitch deviation of gears are obtained,and the error analysis is carried out.The tooth profile deviation is an important parameter that affects the stability of gear transmission.The paper uses the idea of reverse engineering and the principle of computer graphics to establish the standard gear model,and obtains the reference point cloud of gear,and realizes the registration of the target point cloud and the reference point cloud of tooth profile with the classical ICP algorithm,collects the data after registration and carries out curve fitting,obtains the tooth profile deviation and analyzes the error.Through the comparison between the calculated results and the measured values,it is proved that the algorithm proposed in this paper is of high accuracy and can meet the needs of the on-line measurement of spur gears in the actual production process.