Reverse System Of Two-dimensional Precision Parts Based On CCD Vision

With the deepening of the plan to become an industrial power,in industrial manufacturing,the sensitivity and accuracy of many instruments or equipment are required to be extremely high,such as: car engines,rocket bearings and high-speed centrifuges.In order to meet the needs of industrial production and social development,many parts have extremely high requirements on accuracy.The reverse technology is a technology for obtaining CAD models based on physical objects,which is of vital importance for the production and development of precision parts.If the system is to reproduce parts with high precision,high precision and high accuracy,it is necessary to construct a series of powerful data processing methods,such as filtering,denoising,recognition and so on.Existing technologies are difficult to meet actual industrial requirements in terms of stability,and accuracy.For the above-mentioned problems,this paper studies the reverse system of precision parts based on two-dimensional vision,briefly describes the research background,research significance and current status of precision measurement of the reverse system,and focuses on the relevant theories of the precision parts reverse system.The system designed in this article includes hardware and software.The precision parts reverse system has extremely high requirements on data image processing methods.In order to successfully output CAD models,the following methods are introduced and proposed:(1)Calibration.Introduce Zhang's camera calibration method,the purpose of calibration is to obtain the corresponding relationship between pixel coordinates and world coordinate system,and obtain distortion parameters to eliminate camera distortion.(2)An improved filtering algorithm based on homomorphic filtering.In this part,this paper proposes an improved algorithm for filtering based on information entropy.This algorithm plays an important role in practical industrial applications in pursuit of low cost and high precision.This algorithm combines information entropy theory with homomorphic filtering and has a very significant effect on vignetting.Experiments show that high-precision images can be obtained under overexposure.Choosing bilateral filtering method to remove noise is beneficial to the next step of image edge detection.(3)Image edge detect.Base on the experimental the Canny algorithm is used to extract the edge of the part,and the sub-pixel edge is extracted based on the Zernike method.Freeman chain code is used for contour approximation.Using contour approximation method based on genetic algorithm to graphic element segmentation.(4)Highly repeatable graphic element recognition based on the three-point fixed angle method.In the process of image acquisition,due to the dust and dirt in the air,the results of consecutively captured images in the same environment are also different,and this problem of high repeatable graphic element recognition is mostly ignored by scholars,so here is the traditional the primitive recognition algorithm is improved,and a highly repeatable graphic element recognition algorithm based on the three-point angle method is proposed.The algorithm selects pixels for the extracted edges,and then fits the least square method after removing the error points.The algorithm is divided into two steps: the first step is to coarsely screen the pixel points: the least square method is used to roughly fit the contour points after the primitive segmentation.In the second step,sampling is performed at equal intervals,noise is removed at three points,and finally the least square method is used to fit the contour directly to obtain the final result.Use the above methods,and complete the visual system design,image algorithm design,experimental platform construction,and take common industrial parts as the research object,design a set of high-precision inverse system of two-dimensional visual precision parts based on machine vision,so as to obtain parts Vector illustration.Design the corresponding experiment and use VS2015 and Halcon software to process the image.The experimental results show that,compared with the original image without processing,it can be clearly seen that the system has a very good effect on the inverse result of precision parts.It further verifies the correctness of the system and can accurately and efficiently output DXF format files,laying a solid foundation for subsequent design.