Mirror surface imaging bottle cap surface labeling detecting system and method based on machine vision

Abstract

The invention discloses a mirror surface imaging bottle cap surface labeling detecting system and method based on machine vision. The system comprises a conveying belt, an action triggering system, a mirror surface imaging system, a rejection control system and a rejection system, wherein the action triggering system comprises a light triggering sensor and an ultrasonic distance measuring sensor; the mirror surface imaging system comprises an area-array camera, a camera support frame and a plane mirror; the rejection control system comprises an upper computer and a programmable logic controller; the rejection system comprises a spray nozzle, a rejection box and a falling prevention cover plate. The detecting system and the detecting method have the advantages that the structure is simple; the mechanical cost is low; the bottle cap labeling detecting speed is high; the system detecting efficiency is high; meanwhile, the bottle cap slipping phenomenon does not exist; the bottle cap labeling detecting accuracy is high.

Description

technical field [0001] The invention relates to the field of machine vision and industrial product quality inspection, in particular to a machine vision-based mirror imaging bottle cap surface labeling detection system and method. Background technique [0002] Before the liquor bottle cap leaves the factory, it is necessary to use an automatic labeling machine to paste the trademark on the bottle cap cylinder. Since the automatic labeling machine may miss the label when it is working normally, it is necessary to carry out labeling inspection on the bottle cap. If manual inspection is used, it will consume a lot of human resources and the detection efficiency is not high. Therefore, a type of automatic detection device is needed to detect whether the bottle cap is labeled or not quickly on-line. [0003] At present, the application of machine vision technology in the field of automatic inspection is becoming more and more extensive. The existing method of labeling detection ...

AI Technical Summary

Problems solved by technology

[0005] (1) Use the friction wheel rotation technology combined with the line array camera to detect the cylindrical surface information of the bottle cap. Although the image accuracy is guaranteed, due to the time-consuming friction wheel rotation and the time-consuming start-stop of the friction wheel with the conveyor belt, the system detection speed can only reach a maximum 120 pcs / min

In order to reduce production costs, it is necessary to improve the detection efficiency and detect more than 300 per minute. Obviously, this method is not applicable

[0006] (2) The bottle cap may slip when it rotates with the friction wheel. When it slips, the collected image information of the bottle cap will be incomplete, which will seriously affect the labeling detection of the bottle cap. It is likely that the labeled bottle cap will be regarded as missing bottle cap knocked out

[0007] (3) There may be additional vibration that interferes with the stable operation of the friction wheel

Images