A method for online detection and sorting of t-shirts based on machine vision

Abstract

The present invention relates to a T-shirt online detection and sorting method based on machine vision. The T-shirt online detection and sorting method comprises: 1, establishing a T-shirt standard template database, wherein data is obtained from a manufacturer; 2, determining the color, the size, the model and the logo feature of a T-shirt to be detected, finding the data of the standard T-shirttemplate from the T-shirt standard template database, comparing to the to-be-detected T-shirt data collected by a CCD industrial camera, and checking out the inconsistent T-shirt; and 3, matching theinconsistent T-shirt checked in the step 2 and other standard T-Shirt templates, and carrying out detailed classification. According to the present invention, the detection and classification processis not affected by the multiple detection indexes, the complex detection scene environment, the subjective personal factor and other detection factors; the detection and sorting method does not require the too much human detection, has the low labor cost, can avoid the detection miss and the error detection due to the worker attention decreasing caused by the long-time work; and the direct contactbetween the worker and the T-shirt is not required, such that the secondary pollution on the T-shirt cannot be generated; and the accuracy of the detection is high.

Description

technical field [0001] The invention relates to the technical field of machine vision detection, in particular to an online detection and sorting method for T-shirts based on machine vision. Background technique [0002] T-shirts are one of people's favorite clothes in spring and summer, especially in the scorching sun and hot summer. T-shirts are gradually replacing men's vests or shirts with their natural, comfortable, chic and solemn advantages. The pattern of undershirt plus a short-sleeved shirt or Hong Kong shirt appears in social occasions and becomes a seasonal clothing that people are willing to wear. At present, it has become a fashionable dress loved by men, women and children all over the world. It is said that the annual sales volume in the world has reached billions of pieces, and jeans constitute the most popular and most worn clothing in the world. [0003] As T-shirts are flexible products, the quality inspection before leaving the factory is still mainly ...

AI Technical Summary

Problems solved by technology

[0003] As T-shirts are flexible products, the quality inspection before leaving the factory is still mainly manual, and the inspection work faces many difficulties.

For example: there are many and complex detection indicators; the detection site environment is complex; manual evaluation standards are different; long-term work reduces manual attention, and there are missed detections and false detections, etc.

Moreover, the manual contact detection method is easy to cause secondary pollution to the T-shirt, resulting in a decrease in production efficiency, which does not meet the requirements of modern intelligent production

Images