Research On Automatic Identification System Of Casting Defects Based On Image Processing

X-ray nondestructive testing technology is widely used in the defect detection of parts.But the defects recognition of the X-ray detection image,most still rely on manual completion,resulting in low detection efficiency and poor detection accuracy.In recent years,with the development of image processing,pattern recognition and computer technology,automatic detection of defects in industrial parts by X-ray can be achieved through the technology of computer vision and machine learning.It is also a hot topic of current research,and has been successfully applied cases.In order to meet the demand of precision casting defects detection for electric vehicle,the paper designs and develops an automatic recognition system for casting defects based on X-ray detection.This paper mainly studies the following aspects: In recent years,the development of automatic defect detection technology based on image processing and pattern recognition has greatly improved the intelligent level of on line inspection of automobile castings,which meets the requirements of digital and automatic production.In order to solve the problem of automatic defect recognition of the existing automobile castings inspection system,The main research works are as follows:Firstly,based on the domestic and foreign research status of casting X-ray detection,Using the existing X-ray detection platform,the X-ray automatic detection system is proposed.According to the requirement of detection system,the hardware and software of the automatic detection system are designed.Secondly,in order to solve the problem of defect detection of thin wall castings,the defect feature and feature extraction are studied.The method of automatic defect recognition based on image registration and subtraction is presented.Thirdly,for the inherent noise and random noise generated by the imaging process,contrasting the denoising effect of several filtering algorithms and the median filter is used as the denoising algorithm.Fourthly,studied an automatic defect recognition algorithm based on imageregistration and subtraction.The modified Hough transform is used to detect circles and the center as the feature point of image registration.Completed registration with the template image.The defects was extracted by subtraction method,and the defects information was marked on the casting by morphological reconstruction.The corresponding defect detection software is developed.Finally,building the GUI interface of detection system which detect defect automatically based on MATLAB,and compiler software.The text results show that the detecting system can correctly identify the defects in the casting image,achieving defect automatic recognition,and the GUI interface has better practicability.