Bobbin residual yarn detection method based on machine vision

Abstract

The invention provides a residual yarn detection method for bobbins based on machine vision. The residual yarn detection method specifically comprises three parts of image detection area acquisition,image gradient processing and maximum communication domain statistics. Firstly, a to-be-detected target bobbin image needs to be obtained. Then, gradient processing needs to be carried out on the yarnjudgment image, two times of threshold rejection function processing are matched, and the intersection edge of the yarn and the yarn tube is accurately found. And finally, performing connected domaincorrosion expansion processing on the processed three-channel gradient map to restore the yarn, counting the maximum connected domain area of each channel, and comparing the maximum connected domainarea with a set threshold to determine whether the yarn is contained. According to a large number of experimental measurements, by adopting the detection method provided by the invention, the detection precision can reach single-loop residual yarn detection, the detection method is applied to the textile industry, the detection efficiency and the detection precision are greatly improved, and the misjudgment rate is reduced.

Description

technical field [0001] The invention relates to the technical field of machine vision detection, in particular to a method for detecting residual yarn in the textile field. Background technique [0002] In the field of industrial textiles, it is generally necessary to select whether the yarn bobbins contain yarn or not, especially when the yarn bobbins contain very little residual sand during the sorting process, it is necessary to separate the bobbins with remaining yarns Selected from the management machine, sent to the return device. The traditional manual detection method has been unable to meet the current technological requirements, which largely limits the development and progress of the textile manufacturing industry. It comes from the low efficiency, high error rate and high labor cost of traditional manual detection methods; on the other hand, the physiological limit of human glasses also makes it impossible for humans to achieve the accuracy and durability of com...

AI Technical Summary

Problems solved by technology

At present, in the research on the residual yarn detection method of the yarn bobbin in the domestic textile field, the yarn is identified according to the gray value of the yarn, and the color contrast method is used to determine whether the yarn bobbin contains yarn, although these methods It can identify the yarn bobbin with a relatively large amount of yarn, but the recognition and judgment effect of the yarn bobbin with a very small amount of residual yarn is not ideal, and the accuracy is not ideal, and it is affected by the shooting environment light, the yarn Problems such as noise in the reflective area of ​​the tube lead to high initial investment costs and unsatisfactory post-production effects

Images