Article defect detection system and method based on machine vision

Abstract

The invention discloses an article defect detection system and method based on machine vision. The system comprises a feeding device, a conveying device, an image obtaining device and a control and image processing module. The feeding device and an inlet of the conveying device are in butt joint, the feeding device and the conveying device are installed on a rack, the feeding device is used for conveying an article to be detected to the conveying device, and the conveying device is used for conveying the article to be detected to a detection station. The action of the feeding device and the action of the conveying device are controlled by the control and image processing module. The image obtaining device is arranged on the rack and used for obtaining an image of the article to be detected at the detection station and transmitting the image into the control and image processing module for image processing so as to judge whether the corresponding article to be detected has a defect or not. The system is compact in structure, abundant in function, capable of detecting various defects of articles, ingenious in concept and capable of being easily implemented on a high-speed detection line.

Description

technical field [0001] The invention relates to a machine vision-based object defect detection system and method. Background technique [0002] In an industrial production line, it is often necessary to carry out image-based detection of items, which are transparent or opaque cylindrical or cylindrical objects, such as infusion bottles, ampoules, mineral water bottles, car lamp glass bulbs, etc. [0003] When testing, it is first necessary to design the feeding device to transfer the items to the transmission mechanism in an orderly manner; [0004] Then, the conveying mechanism needs to send the item to the detection position. At the detection position, since the item needs to be inspected in all directions, it is necessary to control the spin of the item. On this basis, the images of the end and side of the item are obtained. Finally, the detection is based on the image. [0005] Common appearance defects of articles are air lines, dirty spots, end surface cracks and oth...

AI Technical Summary

Problems solved by technology

[0005] Common appearance defects of items are air lines, dirty spots, end surface cracks and other appearance defects, and there are also defects whose size does not meet the requirements

[0006] In the prior art, there is no such an overall solution, therefore, it is necessary to design a system and method for defect detection of items based on machine vision

Images