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

Application of deep learning in product appearance defect detection

A technology of deep learning and defect detection, applied in the direction of optical test defects/defects, neural learning methods, measuring devices, etc. The waste of human resources, the effect of improving detection speed and improving work efficiency

Inactive Publication Date: 2020-06-26
浙江一木智能科技有限公司
View PDF3 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The flaw detection system based on deep learning in the above patents has the following deficiencies: low contrast between flawed and non-flawed areas, noise and similarity of subtle flaws, etc., which cannot meet the requirements of accuracy and real-time performance

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
  • Application of deep learning in product appearance defect detection
  • Application of deep learning in product appearance defect detection
  • Application of deep learning in product appearance defect detection

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] The application of deep learning in the detection of product appearance defects, such as Figure 1-3 As shown, it includes hardware platform, software detection platform, data labeling module, image partition module, data enhancement module, network training module, parallel detection module and defect segmentation module. The hardware platform includes precision loading control platform and image acquisition platform; data labeling The module is used to mark the type, position and pixel of the defect; the image partition module is used to divide the large image into small images; the data enhancement module is used to realize the expansion of the data set; the parallel detection module adopts multi-threading mode, which can choose to process the same The accuracy of each image is improved, and different images can be processed to achieve detection acceleration; the blemish segmentation module is used to obtain the specific shape of the blemish.

[0036] The precision l...

Embodiment 2

[0053] The application of deep learning in the detection of product appearance defects, such as Figure 1-3 As shown, it includes hardware platform, software detection platform, data labeling module, data enhancement module, parallel detection module and defect segmentation module. The hardware platform includes precision loading control platform and image acquisition platform; the data labeling module is used to mark the defect type, location and pixels; the image partition module is used to divide a large image into small images; the data enhancement module is used to realize the expansion of the data set; Different images can be processed to achieve detection acceleration; the blemish segmentation module is used to obtain the specific shape of the blemish.

[0054] The precision loading control platform includes a stage, which adopts a vacuum adsorption stage device, so that the detected flexible substrate is firmly attached to the stage; the motor and the driving device ar...

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 application of deep learning in product appearance defect detection, and relates to the technical field of defect detection. The objective of the invention is to solve problemsof accuracy and real-time performance of a detection system, a hardware platform, a software detection platform, a data labeling module, an image partitioning module, a data enhancement module, a network training module, a parallel detection module and a defect segmentation module are included; the hardware platform comprises a precise object carrying control platform and an image acquisition platform. According to the invention, segmentation is realized based on the detection technology; the types and positions of the flaws can be obtained; the specific shape can be obtained; a data enhancement operation is used to greatly reduce the cost required by data set making, deep learning parallel detection is adopted, the detection precision is guaranteed, the detection speed is improved, manual visual inspection is replaced, the working efficiency is improved, the waste of human resources caused by manual visual inspection is avoided, and the invention can automatically and quickly detectdefects, and marks the positions and types of the defects.

Description

technical field [0001] The invention relates to the technical field of defect detection, in particular to the application of deep learning in the detection of product appearance defects. Background technique [0002] Deep learning is to learn the internal laws and representation levels of sample data. The information obtained during the learning process is of great help to the interpretation of data such as text, images and sounds. Its ultimate goal is to enable machines to be as analytical as humans Learning ability, able to recognize data such as text, images and sounds. Deep learning is a complex machine learning algorithm. Multi-layer computing hierarchy, select the appropriate input layer and output layer, and establish the functional relationship from input to output through network learning and tuning. Although the functional relationship between input and output cannot be found 100%, it can be done as much as possible. Close to the real relationship, using a success...

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): G06T7/00G01N21/88G01N21/956G06N3/04G06N3/08G06T7/11
CPCG06T7/0006G06T7/11G01N21/956G01N21/8851G06N3/08G01N2021/95638G01N2021/8883G01N2021/888G01N2021/8854G06N3/045
Inventor 董海波
Owner 浙江一木智能科技有限公司
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