Check patentability & draft patents in minutes with Patsnap Eureka AI!

AOI defect classification method and device based on reinforcement learning

A defect classification and reinforcement learning technology, applied in neural learning methods, instruments, biological neural network models, etc., can solve the problem of low classification accuracy, and achieve the effect of reducing human and material resources and improving accuracy

Inactive Publication Date: 2019-07-19
WUHAN JINGLI ELECTRONICS TECH +1
View PDF10 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] In view of this, the present invention provides a method and device for classifying AOI defects based on reinforcement learning to solve or at least partially solve the technical problem of low classification accuracy in the methods in the prior art

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
  • AOI defect classification method and device based on reinforcement learning
  • AOI defect classification method and device based on reinforcement learning
  • AOI defect classification method and device based on reinforcement learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0055] This embodiment provides an AOI defect classification method based on reinforcement learning, please refer to figure 1 , the method includes:

[0056] Step S1: collecting panel inspection images.

[0057] Specifically, the panel inspection images may be collected through existing tools or equipment.

[0058] Step S2: Preprocessing the collected panel inspection images.

[0059] Specifically, the purpose of preprocessing is to better process images for subsequent classification and detection.

[0060] In one embodiment, preprocessing the collected panel inspection images includes:

[0061] Perform grayscale processing on the collected panel image;

[0062] Then, the gray-scaled panel defect detection image is cropped into a sub-image with a preset pixel size.

[0063] Specifically, the preset pixel size can be set according to actual conditions, and the number of cropped sub-images can also be adjusted.

[0064] Step S3: Construct an AOI data set based on the prepr...

Embodiment 2

[0089] This embodiment provides an AOI defect classification device based on reinforcement learning. Please refer to Figure 4 , the device consists of:

[0090] An image acquisition module 201, configured to acquire panel detection images;

[0091] A preprocessing module 202, configured to preprocess the collected panel detection images;

[0092] The AOI data set construction module 203 is used to construct an AOI data set based on the preprocessed panel detection image, wherein the AOI data set includes a training set, a verification set and a test set;

[0093] The expansion strategy generation module 204 is used to select the basic data amplification operation and operation range according to the characteristics of the AOI data set, and generate the expansion strategy in the selected basic data amplification operation and operation range through the reinforcement learning algorithm controller ;

[0094] The expansion strategy application module 205 is used to apply the ...

Embodiment 3

[0108] See Figure 5 , based on the same inventive concept, the present application also provides a computer-readable storage medium 300, on which a computer program 311 is stored. When the program is executed, the method as described in the first embodiment is implemented.

[0109] Since the computer-readable storage medium introduced in the third embodiment of the present invention is the computer equipment used to implement the AOI defect classification method based on reinforcement learning in the first embodiment of the present invention, based on the method introduced in the first embodiment of the present invention, the field Those who belong to it can understand the specific structure and deformation of the computer-readable storage medium, so details will not be repeated here. All computer-readable storage media used in the method in Embodiment 1 of the present invention fall within the scope of protection intended by the present invention.

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 an AOI defect classification method and device based on reinforcement learning. The method comprises the steps of firstly collecting a panel detection image; preprocessing theacquired panel detection image; then constructing an AOI data set based on the preprocessed panel detection image; selecting the basic data amplification operation and the operation amplitude for thecharacteristics of the AOI data set, and generating an amplification strategy in the selected basic data amplification operation and operation amplitude through a reinforcement learning algorithm controller; applying the generated amplification strategy to a training set and a verification set to obtain an amplified training set and an amplified verification set; and finally, carrying out AOI defect classification on the amplified training set and the verification set on a predefined convolutional neural network. According to the invention, the manpower and material resources for collecting alarge number of defect samples are reduced, and the accuracy of the AOI defect classification algorithm is greatly improved.

Description

technical field [0001] The invention relates to the technical field of automatic defect detection of panels, in particular to an AOI defect classification method and device based on reinforcement learning. Background technique [0002] With the popularity of mobile phones and consumer electronics and their rapid replacement, there is a huge demand for the production of LCD screens and OLED screens in industrial production lines. From the whole process of the final molding of the screen, due to raw materials, production processes, accidents, etc., there are often various defects (such as fragments, bubbles, scratches, missing corners, indentations, etc.) on the screen, and these defective products will be Affect its performance or reduce user experience, so it is not allowed to enter the market. Although in the past ten years, AOI (Automatic Optic Inspection technology) has made great progress, but the current AOI technology is mainly based on traditional computer vision alg...

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): G06K9/62G06K9/66G06N3/04G06N3/08
CPCG06N3/08G06V30/194G06N3/045G06F18/214
Inventor 陈春煦张胜森郑增强
Owner WUHAN JINGLI ELECTRONICS TECH
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com