Machine vision-based intelligent detection method for surface defect of bottle cap

Abstract

The invention discloses a machine vision-based intelligent detection method for the surface defect of a bottle cap. The method comprises the following steps of: (1) acquiring the image information of a qualified bottle cap; (2) performing range positioning and characteristic parameter extraction on the surface image of the bottle cap; (3) respectively storing all the acquired parameters of the surface image of the bottle cap in different arrays; (4) judging whether the acquisition times of the bottle cap reaches the preset acquisition times; (5) counting the average value, the upper limit value and the lower limit value of all the parameters, and taking the difference of the upper limit value and the lower limit value as an error allowance range; (6) acquiring the image information of the bottle cap to be detected and exacting characteristic parameters; and (7) judging whether errors corresponding to the parameters exist in the error allowance range, and eliminating the unqualified bottle cap with the defect through identifying and other steps. In the method, the intelligently-extracted characteristic parameters of the surface image of the bottle cap can serve as a detection standard, the quick image detection is realized and the detection efficiency is high. The characteristic parameters can be attached to the conventional production line of the bottle cap so as to realize the on-line detection of the production and the detection at the same time.

Description

technical field [0001] The invention relates to an intelligent detection method for surface defects of bottle caps based on machine vision. Background technique [0002] In the industrial automation production line, especially in the production of metal bottle caps, due to multiple processes such as trademark printing, cutting and stamping, inner pad injection, rubber pad forming, etc., the surface of the bottle cap is prone to defects such as printing defects and scratches. These defects mainly include: ink flying, ghosting, color cast, offset printing, missing printing, black spots, pollution, scratches, etc. In order to sort out unqualified products and improve the pass rate of products, it is necessary to carry out careful inspection of products. The traditional detection is completed by professionals using artificial visual inspection, and its disadvantages are: 1. Slow speed and low efficiency; 2. Unstable detection quality. Under repeated labor, manual inspection is...

AI Technical Summary

Problems solved by technology

The traditional detection is completed by professionals using artificial visual inspection, and its disadvantages are: 1. Slow speed and low efficiency; 2. Unstable detection quality

However, most of the outer surface images of bottle caps do not have rotational symmetry, which brings difficulties for fast image matching detection. Generally, the calculation of asymmetric image matching with random rotation is more complicated and time-consuming, and it is difficult to meet the requirements of fast image matching. Especially in the bottle cap production line, at least 2,500 pieces must be detected per minute. How to quickly and accurately detect the outer surface of bottle caps has been a problem that has plagued production technicians for many years.

Images