Cold-rolled copper strip surface defect recognition model training method, cold-rolled copper strip surface defect recognition method and system

A technology for identifying models and training methods, applied in neural learning methods, character and pattern recognition, biological neural network models, etc., can solve problems such as poor application effect and few defect categories, and achieve fast recognition speed, good application effect, The effect of high recognition accuracy

Pending Publication Date: 2021-11-02
YANSHAN UNIV
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Up to now, the research on copper strip surface defect detection mainly adopts the traditional machine vision method, which is easily disturbed by on-site environmental factors such as light, fog and vibration, and there are few types of defects that can be identified, and the actual application effect is not good.

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
  • Cold-rolled copper strip surface defect recognition model training method, cold-rolled copper strip surface defect recognition method and system
  • Cold-rolled copper strip surface defect recognition model training method, cold-rolled copper strip surface defect recognition method and system
  • Cold-rolled copper strip surface defect recognition model training method, cold-rolled copper strip surface defect recognition method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] Compared with traditional machine vision methods, deep learning methods have better nonlinear learning perception ability and generalization anti-interference ability, and can well overcome the shortcomings of traditional methods. Therefore, this embodiment adopts a deep learning method to establish a recognition model, such as figure 1 As shown, the present embodiment is used to provide a cold-rolled copper strip surface defect recognition model training method, the training method comprising:

[0036] S1: Establish a surface defect data set of cold-rolled copper strip; the surface defect data set includes a plurality of training images corresponding to various surface defect categories;

[0037] Before S1, the training method of this embodiment also includes: according to actual needs, selecting lines, black spots, bumps, edge cracks, holes, insect spots, peeling and dirt as the surface defect categories included in the surface defect data set . Specifically, there ...

Embodiment 2

[0088] This embodiment is used to provide a method for identifying defects on the surface of a cold-rolled copper strip, such as Figure 9 As shown, the identification method includes:

[0089] T1: Obtain the image to be recognized corresponding to the cold-rolled copper strip;

[0090] Specifically, the acquisition process is that at the end of the cold-rolled copper strip cleaning line and before the coiling unit, multiple sets of high-speed cameras are used to continuously shoot the surface of the copper strip to obtain images to be recognized.

[0091] T2: Using the image to be recognized as input, use the recognition model trained in Embodiment 1 to perform real-time recognition on the image to be recognized to obtain the surface defect category corresponding to the image to be recognized.

[0092] At the same time, the captured original image information, recognition result information, and strip property information are displayed and stored in the computer in real time...

Embodiment 3

[0096] This embodiment is used to provide a system for identifying defects on the surface of a cold-rolled copper strip, such as Figure 11 As shown, the identification system includes:

[0097] The acquisition module M1 is used to acquire the image to be identified corresponding to the cold-rolled copper strip;

[0098] The acquisition process is that at the end of the cold-rolled copper strip cleaning line and before the coiling unit, multiple sets of high-speed cameras are used to continuously photograph the surface of the copper strip.

[0099] The identification module M2 is configured to use the image to be identified as an input, and use the identification model trained in Embodiment 1 to identify the image to be identified in real time to obtain the surface defect category corresponding to the image to be identified.

[0100] At the same time, the captured original image information, recognition result information, and strip property information are displayed and stor...

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 relates to a cold-rolled copper strip surface defect recognition model training method, which comprises the following steps: establishing a surface defect data set of a cold-rolled copper strip, the surface defect data set comprising a plurality of training images corresponding to a plurality of surface defect categories; constructing an initial recognition model, wherein the initial recognition model is a deep convolutional neural network model; and finally, training the initial recognition model by using the surface defect data set to obtain a recognition model, so that the established recognition model can recognize various defect categories, and the actual application effect is good. The invention further provides a cold-rolled copper strip surface defect recognition method and system, the recognition model obtained through training of the training method is used for recognizing the to-be-recognized image, the surface defect category corresponding to the to-be-recognized image is obtained, the recognition precision is high, the recognition speed is high, and the surface defect category can be detected online in real time.

Description

technical field [0001] The invention relates to the technical field of detection of strip surface quality, in particular to a training method, recognition method and system for surface defect recognition models of cold-rolled copper strips. Background technique [0002] Cold-rolled copper strip is a typical high-end product in the field of non-ferrous metals, widely used in new energy vehicles, aerospace and precision electronic equipment and other fields. Surface quality is one of the important quality indicators of cold-rolled copper strip. Surface defects not only seriously affect product appearance and yield, but may also have adverse effects on downstream production processes. Therefore, it is of great significance to improve product quality to realize accurate and rapid classification and identification of copper strip surface defects. [0003] At present, in industrial production, a large number of manual visual inspection methods are still used to detect copper stri...

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
IPC IPC(8): G06K9/62G06N3/04G06N3/08G06T7/00G01N21/01G01N21/88G01N21/95
CPCG06T7/0004G06N3/08G01N21/8851G01N21/95G01N21/01G01N2021/8887G01N2021/0112G06T2207/20081G06T2207/20084G06T2207/30136G06N3/045G06F18/214G06F18/241
Inventor 王东城徐扬欢刘计尊段伯伟杨实禹于华鑫刘宏民
Owner YANSHAN UNIV
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