Research On Design And Control Of Robotic Arm For Petrochemical Inspection Operations

With the continuous development of the automotive industry,people’s requirements for car interiors are getting higher and higher,but the button pattern in the car is easily damaged,so polymer chrome plated materials are used for printing.In order to ensure the printing quality of automotive key patterns,it is particularly important to detect defects in the keys.Most of the traditional key defect detection adopts manual detection,but this method is labor-intensive,human subjective factors are easy to affect the detection results,and the efficiency is very low.With the increasing maturity of digital image processing technology and machine vision theory,machine vision instead of manual defect detection has become an unstoppable trend,machine vision technology is one of the important means to achieve industrial automation,this paper for the quality detection of automotive keys,proposed a solution based on machine vision technology.This paper first briefly describes the performance and development of machine vision and image processing technology at home and abroad,and then mainly introduces the image enhancement and filtering in terms of image preprocessing,because in the complex environment of industry,may produce a variety of noise,better pattern in the defect detection will have better effect,in the mean filtering and median filtering on the basis of improved filtering has a very good denoising effect.Secondly,the research and improvement of template matching algorithm,before template matching,it is necessary to perform image segmentation and extract edges of the template,which is convenient for the research of image edge oriented and image shape oriented algorithm in the process of subsequent template matching positioning,and the pyramid layering strategy and random sampling consensus algorithm are used to optimize in the process of template matching.Locate to the template area by means of template matching and affine transformation.The defect part can be accurately detected by using the defect detection method of the difference model after positioning and registration,and the color model and chromatic aberration evaluation method are introduced at the end of this thesis.Through the analysis and verification and test results,the developed machine vision-based fault detection system has excellent robustness and has passed multiple field tests,while providing technical guidance for subsequent vehicle test workshops.It also provides viable ideas for testing other food packaging,electrical appliances and other electronic products.