Research On The Defect Method Of Detection Circular Ceramic Metal Coating

In order to meet the requirements for high-precision,high-efficiency,and non-destructive surface defect detection,machine vision inspection methods based on digital image processing are increasingly applied in various fields of industrial production.Aiming at the problems of surface defects detection of ring-shaped devices,this paper takes ring-shaped metal-coated ceramics as the research object,and proposes a set of ring-shaped ceramic metal-coating defect detection algorithms based on digital image processing technology,and applies them to the online visual inspection system.in.The main contents of the algorithm proposed in this paper are as follows:1)Image pre-processing algorithm design:the ring-shaped metal-coated ceramic initial grayscale image obtained by the camera contains three concentric circles,namely the bare ceramic inner ring,the metal-coated inner ring and the outer ring.The interference of defect detection mainly comes from the background of the ceramic inner ring,the background of the inner and outer rings of the metal coating,and the edge noise of the metal coating.Therefore,before detection,the image must be pre-processed to reduce the impact of many noises and gradually separate multiple backgrounds.In this paper,median filtering is used to complete the noise reduction of the image;Otsu method is used to find the global optimal threshold value in the image to complete the binarization;the method based on Canny edge detection+Hough transform is used to detect the concentric circle parameters,and the multiple parameters are separated by analyzing and calculating the multiples.Background to achieve the purpose of extracting the metal coating area.2)Ring image defect detection algorithm design:In order to simplify the calculation,the smallest circumscribed rectangle of the metal-coated ring is cropped,the area is projected radially,the ring image is converted into a rectangular shape,and the threshold segmentation method is used to filter In addition to metal coating edge noise.Find the connected domain in the area of the metal coating where the noise has been filtered,calculate the area and perimeter of the connected domain,determine whether it is defective,and locate it.In order to verify the defect detection method proposed in this paper,and to meet the needs of enterprises for defect detection of annular ceramic metal coatings.A set of on-line visual inspection system for circular ceramic metal coating defects is designed.The system includes mechanical part,control part,image acquisition part and application software.The mechanical part completes the tasks of feeding,conveying,turning over,and rejection;the control part is responsible for driving the three-way stepping motor to rotate within a certain speed range,and accepting the sorting signal,sending out the rejection signal,and rejecting the defective products;the image acquisition part is completed The acquisition of the front and back of the original image;the application software is responsible for processing and identifying the original image,displaying the processing results,camera working status,number of processed samples and other related information,and sending the sorting instructions to the control section.The experimental results show that the method of Canny edge detection+Hough transform is convenient to detect multiple concentric circles and obtain the center position,radius and other parameters.The various parts work together accurately and the detection system runs stably.Experimental data shows that the algorithm proposed in this paper has a detection accuracy rate of 98%for samples with a defect area greater than 0.5 mm × 0.5 mm;the part turnover rate is 99%.