Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Steel plate surface defect detection method based on deep learning

A deep learning and defect detection technology, applied in neural learning methods, image data processing, instruments, etc., can solve problems such as difficult to meet production needs, slow speed, poor performance, etc., to achieve the effect of improving accuracy and precision

Inactive Publication Date: 2020-08-11
HARBIN UNIV OF SCI & TECH
View PDF7 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The PSO algorithm optimizes the steel plate surface detection technology of the BP network, which is not easy to fall into the local optimum and is more applicable, but the accuracy rate is still not high
[0004] In recent years, with the rapid development of high-performance computing equipment such as GPU, breakthroughs have been made in deep learning technology. At present, deep learning has achieved remarkable research and application results in many fields, but in steel plate detection, although there are some attempts , but the performance in some evaluation indicators is poor, the speed is slow, and the support for Windows platform is poor, so it is difficult to meet the production needs

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
  • Steel plate surface defect detection method based on deep learning
  • Steel plate surface defect detection method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0017] Specific implementation mode one: as Figure 1 ~ Figure 2 As shown, the present invention discloses a steel plate surface defect detection method based on deep learning, and the method includes the following steps:

[0018] S1: Obtain the original image of the steel plate sample;

[0019] S2: determine the key area of ​​the original image;

[0020] S3: Recognize and process the image of S2 through the image recognition module based on deep learning;

[0021] S4: perform statistical processing on the output result of S3 according to the counting module;

[0022] S5: If the result of S4 is a good product, then the steel plate sample is a good product, otherwise, repeat S1 to S4 to recheck the suspected defective product.

specific Embodiment approach 2

[0023] Embodiment 2: This embodiment is a further description of Embodiment 1. The original image of the steel plate sample in S1 is captured by an industrial camera.

specific Embodiment approach 3

[0024] Specific implementation mode three: this implementation mode is a further description of specific implementation mode one. Determining the key region of the original image described in S2 includes the following steps:

[0025] S301: Perform a cutting operation on the original image to obtain multiple slices;

[0026] S302: Input each slice into the image recognition module based on deep learning, perform forward propagation, and obtain the probability that each slice belongs to the key region;

[0027] S303: Determine the slice with the highest probability as the key region of the original image.

[0028] For ease of description, the above three steps are combined for description.

[0029] After the target picture (original image) is determined, a cutting operation is performed on the target picture to obtain multiple slices corresponding to the target picture.

[0030] Specifically, the target image can be cut through a preset sliding window to obtain multiple slices...

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 steel plate surface defect detection method based on deep learning and belongs to the technical field of steel plate detection. The method comprises the steps of: acquiring an original image of the steel plate sample, determining the key area of the original image, performing identification processing on the image of the S2 through an image identification module based ondeep learning, performing statistical processing on an output result of the S3 according to a counting module, if the result of the S4 is a good product, determining that the steel plate sample is a good product, and otherwise, repeating the S1 to S4 to recheck suspected defective products. The steel plate surface defects are automatically detected based on deep learning, and the accuracy and precision of steel plate surface multi-type defect detection are greatly improved.

Description

technical field [0001] The invention relates to a method for detecting steel plate surface defects based on deep learning, which belongs to the technical field of steel plate detection. Background technique [0002] Defects on the surface of an object have a direct impact on the quality of the object, and also affect the user experience. Especially for some objects with high precision requirements and special usage scenarios, the presence or absence of surface defects directly determines whether the object can enter the market. [0003] The iron and steel industry plays an irreplaceable role in my country's economic development, and has invested a lot of human and material resources. In the process of competing with other steel companies in the world, the detection technology of steel plate surface defects is still a major bottleneck. The defect detection of traditional manufacturers only relies on human eye recognition, which has relatively large limitations. With the devel...

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): G06T7/00G06N3/08G06K9/46
CPCG06T7/0002G06N3/08G06V10/462
Inventor 董静薇王欣
Owner HARBIN UNIV OF SCI & TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com