Textile crease detection method based on deep learning

A technology of deep learning and textiles, applied in neural learning methods, image data processing, instruments, etc., can solve the problems of manual design parameters such as heavy workload, easy to be affected by uneven illumination, and low system security, so as to reduce manual adjustment. parameters, improve convenience, and improve detection efficiency

Inactive Publication Date: 2020-08-25
郑州蓝智枫智能科技有限公司
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

It is also unable to overcome the influence of uneven illumination on feature extraction
In addition, in the existing crease detection calculation, due to the lack of corresponding encryption algorithm design, the system security is low
[0004] In summary, the existing textile crease detection technology has problems such as low detection accuracy, heavy workload of manual design parameters, easy to be affected by uneven illumination, and low safety.

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 crease detection method based on deep learning
  • Textile crease detection method based on deep learning
  • Textile crease detection method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0051] A method for detecting creases in textiles based on deep learning, the method comprising:

[0052] (1) Extract the textile creases from the original image of the textile.

[0053] When the camera used is an RGB camera, the image needs to be grayscaled. This can be undone if using a grayscale camera. Assuming that the image S is a grayscale image, the corresponding pixel should have S(i,j)=max(R(i,j),G(i,j),B(i,j)), where R(i ,j), G(i,j), and B(i,j) are the data of the R, G, and B channels of the original textile image respectively, and the original grayscale image S(i,j) of the textile is obtained.

[0054] (1a) Perform histogram equalization processing on the original grayscale image of the textile.

[0055] In order to perform self-quotient processing on the image, it is necessary to filter the spatial domain of the grayscale image to enhance the contrast with the original image, so that the self-quotient effect is better. First, histogram equalization is performed ...

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 crease detection method based on deep learning, and the method comprises the steps: firstly collecting a plurality of original gray images of a textile, detecting a textile crease binary image corresponding to the gray images of the textile, and obtaining a textile crease sample set; then, carrying out sample enhancement on the textile original grayscale image sample set and the corresponding textile crease sample set to obtain an amplified sample set; training a textile crease detection semantic segmentation deep convolutional neural network by using the amplified sample set; and performing textile crease detection by using the trained semantic segmentation deep convolutional neural network in combination with a block chain technology. By means of the textile crease detection device and method, in textile crease detection, the detection precision can be improved, manual parameter adjustment is reduced, the illumination influence is overcome, and the safety in the data processing process can be improved.

Description

technical field [0001] The invention relates to the technical fields of artificial intelligence, computer vision and block chain, in particular to a method for detecting creases in textiles based on deep learning. Background technique [0002] Crease detection is a difficult point in textile defect detection. For textile materials, there are many factors for the formation of creases. For example, the cloth opener did not fully unfold the cloth in the early processing, or the cloth was not completely spread when the rotary drum dryer was used, and there were accumulations and wrinkles. condition, creases appear after the machine is started. The existence of creases affects the practicability and appearance of textiles. If there are creases, it means that there are problems with the previous machine or process, and timely adjustments are required. At the same time, the treatment of creased fabrics is a very important part of the textile factory inspection. [0003] Textile ...

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): G06T7/00G06T7/136G06T5/40G06N3/04G06N3/08
CPCG06T7/0004G06T7/136G06T5/40G06N3/08G06T2207/20032G06T2207/20081G06T2207/30124G06N3/045
Inventor 李玉枫蔡路张俊鹏张勇赵雨航
Owner 郑州蓝智枫智能科技有限公司
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