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

Thermal imaging defect detection method based on convolution auto-encoder image amplification

A convolutional self-encoding and defect detection technology, applied in the field of thermal imaging defect detection based on convolutional self-encoder image amplification, can solve problems such as limitations, improve quality, improve accuracy and reliability, and reduce image noise Effect

Pending Publication Date: 2021-05-25
ZHEJIANG UNIV OF TECH
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, a basic problem, that is, thermal image data contains noise, has not been concerned in most of the modeling methods mentioned above, which actually limits the performance of most thermal imaging data analysis models from the perspective of data quality

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
  • Thermal imaging defect detection method based on convolution auto-encoder image amplification
  • Thermal imaging defect detection method based on convolution auto-encoder image amplification
  • Thermal imaging defect detection method based on convolution auto-encoder image amplification

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0055] The present invention will be further described below in conjunction with the accompanying drawings.

[0056] refer to Figure 1 to Figure 5 , based on the manifold learning thermal imaging defect detection method of autoencoder image enhancement, the method comprises the following steps:

[0057] Step 1. Obtain the thermal image and thermal image data of the carbon fiber reinforced polymer composite material.

[0058] Resin transfer molding is used to manufacture rectangular carbon fiber reinforced polymer samples with a size of 18cm×18cm and thickness, and three polytetrafluoroethylene strips of trapezoidal, circular, and diamond shapes are inserted at different depths inside to simulate defects, and the size of each defect area about 3cm 2 . The IRT system is established, and the working flash excites the thermal pulse to heat the sample. After the thermal front touches the surface of the experimental sample and passes through the sample, an infrared thermal camer...

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 thermal imaging defect detection method based on convolution auto-encoder image amplification. The method comprises the following steps: 1, obtaining a thermal image and thermal image data of a carbon fiber reinforced polymer composite material; 2, selecting and processing the thermal image; 3, establishing a convolutional auto-encoder image enhancement model, and carrying out noise reduction on the thermal image; 4, preprocessing the thermal image data; 5, establishing a Laplau feature mapping thermal imaging model based on the experimental thermal image and the convolutional auto-encoder reconstructed thermal image, extracting defect information in the thermal image and visualizing the defect information to realize qualitative monitoring of the defect; and 6, evaluating the established model by adopting a separation degree index, and verifying the effect of the image enhancement strategy in defect detection modeling.

Description

technical field [0001] The invention belongs to the field of composite material defect detection, and in particular relates to a thermal imaging defect detection method based on convolutional self-encoder image amplification. Background technique [0002] Invisible defects such as voids, inclusions, and fiber breaks in composite materials seriously affect the quality of composite materials, and poor-quality composite products will cause economic losses, and even endanger life safety due to product failure. Therefore, defect assessment of composite materials is a necessary work. Non-destructive testing technology has been widely used in many occasions because it does not destroy the material when inspecting the discontinuous properties of the material. Among many non-destructive testing technologies, active thermal imaging (IRT) technology is widely used in the detection of material subsurface defects due to its low cost and easy implementation. The principle of IRT is base...

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): G06T7/00G06T5/00G06N3/04G06N3/08
CPCG06T7/0004G06N3/08G06T2207/10048G06T2207/20081G06T2207/20084G06T2207/30108G06N3/048G06T5/70
Inventor 刘毅刘凯新娄维尧杨建国
Owner ZHEJIANG UNIV OF 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