Textile fabric defect detection model and training method and application thereof

A training method and point detection technology, applied in the field of deep learning and computer vision, can solve problems such as low accuracy, poor real-time performance, and lack of versatility, and achieve the effects of improving accuracy, fast detection speed, and improving accuracy

Active Publication Date: 2018-09-11
HUAZHONG UNIV OF SCI & TECH
View PDF4 Cites 16 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In view of the above defects or improvement needs of the prior art, the present invention provides a textile defect detection model and its training method and application, thereby solving the problems of low accuracy, poor real-time performance and no general technical issues

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Textile fabric defect detection model and training method and application thereof
  • Textile fabric defect detection model and training method and application thereof
  • Textile fabric defect detection model and training method and application thereof

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0062] The convolutional neural network used in the present invention can accept the input of any pixel image in principle, but considering that an image with too large pixels will cause small defects to be distorted after resize, and an image with too small pixels will cause some features to be extracted. , the reference pixel given by the official is 416×416. In embodiment 1 of the present invention, a pixel size of 1216×1020 is adopted in consideration of the large scope of one-time identification of cloth and the adequacy of feature extraction.

[0063] An application of a textile defect detection model, including:

[0064](1) Use an industrial camera to collect and select images containing six types of defects including warp breaks, weft breaks, holes, foreign matter, oil stains, and creases as sample textile defect images. The sample textile defect images are in jpg format with a resolution of 96dpi The three-channel color image with a pixel size of 1216×1020; create two...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a textile fabric defect detection model and a training method and application thereof. The training method includes the steps that sample textile fabric defect images are acquired, a data set is established, and the textile fabric defect detection model is established based on YOLOv2; dimensional clustering is used before the model is trained, when the model is trained, coordinate prediction, loss value calculation and counterpropagation are directly carried out, and current network weight parameters are obtained; network weight parameters of the textile fabric defect detection model are updated through the current network weight parameters, then multiple times of network weight calculation and update are carried out through a training set, optimal network weight parameters are obtained, and accordingly the trained textile fabric defect detection model is obtained; then textile fabric images are acquired in real time, the trained textile fabric defect detectionmodel is used for detection, and defect detection results of the textile fabric images are obtained. The textile fabric defect detection model has the advantages of being high in defect accuracy, highin real-time performance and high in universality.

Description

technical field [0001] The invention belongs to the technical field of deep learning and computer vision, and more specifically relates to a textile defect detection model and its training method and application. Background technique [0002] In the production and development of the world's textile industry, the quality inspection of textile fabrics has always been a very important link. However, in the traditional quality inspection of textile fabrics, due to the lack of good automatic inspection tools, most of the solutions are still judged by artificial vision. Work is prone to fatigue, and accuracy is difficult to guarantee. With the rapid increase in the production volume and production speed of textile fabrics, the artificial vision method is increasingly unsuitable for the needs of the modern textile industry. It is urgent to find an automatic, accurate and fast method for quality or defect detection. At present, the detection methods for textile defects in China in...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06F17/50G01N21/88G06N3/04G06N3/08
CPCG06N3/084G01N21/88G01N2201/1296G06F30/00G06N3/045
Inventor 孙志刚禹万泓江湧王卓肖力
Owner HUAZHONG UNIV OF SCI & TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap