"5G + AI Grey Fabric Quality Inspection" Helps To Upgrade The Wisdom Of Textile Industry

Recently, Fujian Mobile and Fujian HANGGANG Knitting Co., Ltd. jointly set up a research team to apply 5g technology, artificial intelligence and big data to all aspects of cloth production and inspection, and carried out the application research and development of AI quality inspection of "non thread cut lace grey fabric". In the process of grey fabric production and inspection, the surface defects were detected timely, quickly and accurately, and the defect location was automatically recorded, And assist manual inspection and processing of defect information, so as to promote the improvement of production process, save production costs, and improve the appearance quality of products.

In recent years, with the rapid development of digital image technology, automatic detection mode based on image processing and recognition has been continuously optimized and updated. There are many kinds of grey fabric defects, the frequency of occurrence is unstable and the difference of some defects is small, which leads to the recognition effect of some defects is not as expected, so it needs continuous optimization. How to improve the efficiency of grey fabric defect detection process, solve the problems of high false detection rate and high missing detection rate, and improve the production process of grey fabric has become one of the important factors to improve the quality of grey cloth.

To this end, Fujian Mobile uses 5g industrial camera and other front-end equipment to quickly collect high-definition visual information of different dimensions of grey fabric to be inspected, and transmits it to the quality inspection platform at high speed through 5g private network. It also uses "Ai grey fabric defect recognition algorithm", and applies a variety of machine vision detection technologies, supporting grid division and FPN multi-scale detection, In particular, the small-scale target is optimized to accurately identify the small defects under the background of complex patterns. It can provide on-line high-speed and accurate fabric surface defect detection for 15 common defects in non thread cutting lace grey fabric, and support automatic inspection and intervention shutdown according to defect level information and manual processing configuration rules, so as to realize the transformation from "one person watching over one cloth inspection machine" to "one person multi machine", so as to improve the quality inspection efficiency and reduce the work threshold of cloth inspectors, Save cost and improve product quality for enterprises.

As an important application of AI in the industrial field, "5g + AI grey fabric quality inspection" conforms to the development trend of automation and intelligence in the textile industry, and better meets the needs of modern large-scale production in terms of the speed, accuracy and consistency of detection results of gray fabric anomaly detection, and has a strong promotion effect in the same industry. In the next step, Fujian Mobile will continue to carry out cooperation and innovation with Fujian HANGGANG Knitting Co., Ltd., continuously enrich the "5g + smart textile" application scenarios, and improve the informatization level of the textile industry.

(source: People's post and Telecommunications News)