METHOD FOR PROMPTLY DETECTING SURFACE DEFECTS OF CROWN CAPS BASED ON IMAGE PROCESSING Russian patent published in 2022 - IPC G01N21/95 G06V10/40 

FIELD: detecting faults.

SUBSTANCE: invention relates to a method for promptly detecting surface defects of crown caps based on image processing. Method includes stage 1: obtaining an image of the inner surface of the crown cap using an image capturing system and processing the resulting image by the method for Gaussian filter; stage 2: isolating the central section in the desired image by thresholding and analysing the characteristics of the region with the associated components, and selecting the circular contour of the edge pixels of the central contour of the section by selecting the least squares to determine the central position of the inner surface of the crown cap; stage 3: presetting the information about the physical radius of different sections for different types of crown caps according to the coordinates of the central position of the crown cap obtained at stage 2, and performing segmentation of the section of the inner surface of the crown cap in combination with the parameters of pixel ratio of the image; stage 4: isolating an image in the desired section of the inner surface based on the result of image segmentation at stage 3, setting a fixed threshold brightness value to complete image segmentation, and removing the noise section by analysing the characteristics of the region with the connectivity components in order to detect dark spots on the inner surface of the crown cap; stage 5: isolating an image in the inner and outer annular sections based on the result of image segmentation at stage 3, completing the image segmentation by the local threshold method, and removing the noise section by analysing the characteristics of the region with the connectivity components in order to detect bulging in the inner and outer annular sections on the inner surface of the crown cap; stage 6: obtaining the edge contour of the outer annular section of the obtained image by fixed thresholding, performing circular selection of pixels of the outer annular contour by the least squares method, and comparing the differences between the outer annular contour section and the selection section in order to detect the absence of rubber; stage 7: isolating an image of the flash section based on on the result of image segmentation at stage 3 and detecting the flash line and the soldering by fixed thresholding and by analysing the characteristics of the region; stage 8: isolating the skirt teeth connectivity component based on the result of image segmentation at stage 3, calculating the number of connectivity components, calculating the inner angle between the connectivity component and the centre of the crown cap, and determining whether skirt teeth defects are present based on the difference between the inner corners of the adjacent skirt teeth; stage 9: obtaining an image on the outer surface of the standard crown cap, processing the resulting images by the method for Gaussian filter, and creating an autonomous characteristic template according to the image contour characteristics; stage 10: establishing the template match indicator, and detecting and positioning the image, which undergoes the detection procedure by comparing the characteristics with the template after the Gaussian filter processing; and stage 11: performing an affine transformation on the target image based on the result of the template match at stage 10, performing thresholding and comparing the sections in the template and target images, respectively, and detecting outer surface defects by analysing the characteristics of the region with the connectivity components.

EFFECT: higher accuracy of detecting defects of crown caps.

10 cl, 9 dwg