Unlock instant, AI-driven research and patent intelligence for your innovation.

A classification method for industrial product defect images based on knowledge graph

An industrial product and knowledge map technology, applied in the field of computer vision, can solve the problems of unbalanced and small differences in categories of defects, achieve the effect of improving robustness, good versatility and universality, and reducing the dependence of defect samples

Active Publication Date: 2022-03-11
ZHEJIANG UNIV
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] However, unlike the image classification in natural scenes, the defect classification rules of industrial images are often based on the production experience of the factory, not just based on the characteristics of the image itself.
In addition, in many cases, the difference between different types of defects is small, and the defects of different types of defects are often unbalanced, which brings great challenges to the defect classification of deep learning in industrial images.

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
  • A classification method for industrial product defect images based on knowledge graph
  • A classification method for industrial product defect images based on knowledge graph
  • A classification method for industrial product defect images based on knowledge graph

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0033] The purpose and effects of the present invention will become clearer by describing the present invention in detail according to the accompanying drawings and preferred embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

[0034] Such as figure 1 As shown, the knowledge graph-based industrial product defect image classification method of the present invention comprises the following steps:

[0035] S1: Create defect gallery X, label library Y corresponding to each defect map, and additional attribute vector V corresponding to defect category;

[0036] S2: Using the image enhancement technology to enhance the defect map to obtain the enhanced defect library X′.

[0037] S3: Construct defect image feature extraction network, and then continuously obtain image batches x from the defect gallery X′ enhanced by S2 B and its corresponding label y B . Batch defe...

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 method for classifying industrial product defect images based on knowledge graphs. The method can classify industrial product defects by combining empirical knowledge in industrial production and image features of the defects themselves, and solves the problem that the previous convolutional neural network only relies on the image itself. The features of the defect classify the defects. This method can greatly improve the accuracy of deep learning in the classification of industrial product defects while reducing the dependence on defect samples.

Description

technical field [0001] The invention belongs to the field of computer vision, and in particular relates to a method for classifying industrial product defect images based on knowledge graphs. Background technique [0002] In industrial production, in order to optimize the production process and improve production quality, it is of great significance to correctly classify the defects in the production process of products. Usually, the classification of industrial product defects can only be effectively classified by professionally trained workers. This method is inefficient and the accuracy is often difficult to be effectively guaranteed. On the one hand, enterprises have to pay high labor costs for this, on the other hand, the process of image classification is extremely boring, and the job rotation rate is high. With the introduction of Industry 4.0 and the development of artificial intelligence technology, image classification technology represented by deep learning has b...

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 Patents(China)
IPC IPC(8): G06T7/00G06V10/764G06V10/82G06K9/62G06N3/04
CPCG06T7/0004G06N3/04G06F18/241
Inventor 余永强楼利璇刘小为
Owner ZHEJIANG UNIV