Research On Gear Measurement System Based On Active Vision

Gear is the most widely used part in the field of mechanical transmission,the quality of gear directly affects the performance of the entire mechanical system.In the process of industrialization,gear has always played an important role.With the development of science and technology,all trades put forward higher requirements on the gear quality.Due to the price is too high,the use and maintenance requirements are high,Existing gear measuring instruments and equipment hasn't been popularized.Machine vision measurement as a non-contact measurement,has been gradually applied to the field of industrial testing,because it is a non-destructive dectection,the detection efficiency and degree of automation is high.However,there are still many problems in the visual measurement.In this paper,a gear measurement method based on active vision is proposed for the problem that the measurement accuracy is low due to the low resolution of the camera and the distortion of the camera.Firstly,design the active vision measurement system,including the hardware part and the software part.The hardware design part includes the three-dimensional mobile platform,the motor driver,and the design and selection of the camera,lens and light source.The software design part includes the driver software design of the three-dimensional mobile platform of the lower computer,the software design of the host computer and the communication design of the upper and lower computer.By studying the image processing algorithms of the visual measurement,the preprocessing methods and processes of the gear images are designed.The median filtering is performed first,and the gears are segmented from the background by the adaptive threshold segmentation method.Then,use Canny edge detection algorithm to detect the edge.Then,analyze and study the ideal imaging model and distortion model of the camera.The mapping relationship between the world coordinate system and the image coordinate system is established.The process of solving the camera parameters and the method of calibrating the camera with the checkerboard are described.The calibrated camera parameters are used to correct the acquired gear image.The method of active vision measuring gear is described.Firstly,Analyze the complete image of the measured gear and calculate the rough parameters.Then the system's attention focus on each part of the gear to analyze and measure by high-resolution.And then the data of the whole parts is analyzed together.The standard gauge block is used to calibrate the active vision system to obtain the pixel equivalent of the system.In the process of fitting the straight edge of the block,it is found that the accuracy of the Hough transform is low and the robustness is poor,so the least squares method is used to fit the extracted block contour.Finally,the feasibility and significance of the active vision method are measured by comparing the measured results of the active visual method with the standard values of the measured gears and the measured results of the general machine vision method.