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

Concrete defect detection method and device based on full convolutional neural network, and storage medium

A convolutional neural network and defect detection technology, applied in the field of concrete defect detection method, device and storage medium based on full convolutional neural network, can solve problems such as low detection efficiency, achieve high recognition efficiency, high detection accuracy, reduce The effect of calculating the cost

Inactive Publication Date: 2020-09-01
STATE GRID HUNAN ELECTRIC POWER +2
View PDF5 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The invention provides a concrete defect detection method, device and storage medium based on a fully convolutional neural network to solve the problem of low detection efficiency mainly relying on manual detection at present

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
  • Concrete defect detection method and device based on full convolutional neural network, and storage medium
  • Concrete defect detection method and device based on full convolutional neural network, and storage medium
  • Concrete defect detection method and device based on full convolutional neural network, and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0050] like figure 1 As shown, this embodiment provides a concrete defect detection method based on a fully convolutional neural network, including:

[0051] S01: acquiring a concrete surface image;

[0052]S03: Input the acquired concrete surface image into the pre-trained concrete defect semantic segmentation model, and output the concrete defect semantic segmentation map classified pixel by pixel to realize concrete defect recognition; wherein, the concrete defect semantic segmentation model is based on the Annotated concrete surface image data is obtained after training a fully convolutional neural network.

[0053] The above scheme obtains the concrete defect semantic segmentation model by pre-training the fully convolutional neural network based on the marked concrete surface image data, and then directly inputs the acquired concrete surface image into the pre-trained concrete defect when performing concrete detection. In the semantic segmentation model, the semantic s...

Embodiment 2

[0063] Such as figure 2 As shown, this embodiment provides a concrete defect detection method based on a fully convolutional neural network, including:

[0064] S01: acquiring a concrete surface image;

[0065] S02: Perform size scaling preprocessing on the acquired concrete surface image, such as uniformly scaling the resolution to 480*480, through size scaling preprocessing, the high-resolution image can be scaled to a preset size image, thereby reducing Computational cost, reducing the requirements for the processor, which is conducive to reducing the cost of implementation;

[0066] S03: Input the preprocessed concrete surface image into the pre-trained concrete defect semantic segmentation model, output the concrete defect semantic segmentation map classified pixel by pixel, and realize concrete defect recognition; wherein, the concrete defect semantic segmentation model is based on The labeled concrete surface image data is obtained after training a fully convolutiona...

Embodiment 3

[0075] Such as image 3 As shown, this embodiment provides a concrete defect detection method based on a fully convolutional neural network, including:

[0076] S01: acquiring a concrete surface image;

[0077] S02: Use the sliding window to process the acquired concrete surface image to generate several partial images of the concrete surface; through the preprocessing of the sliding window, such as selecting a sliding window with a resolution of 480*480, the high-resolution image can be divided into Several partial images, and then process the partial images, thereby reducing the calculation cost, reducing the requirements for the processor, and helping to reduce the cost of implementation;

[0078] S03: Input the preprocessed concrete surface image into the pre-trained concrete defect semantic segmentation model, output the concrete defect semantic segmentation map classified pixel by pixel, and realize concrete defect recognition; wherein, the concrete defect semantic segm...

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 concrete defect detection method and device based on a full convolutional neural network and a storage medium. The method comprises the steps of acquiring a concrete surfaceimage; and inputting the acquired concrete surface image into a pre-trained concrete defect semantic segmentation model, and outputting a concrete defect semantic segmentation graph classified pixel by pixel to realize concrete defect identification. The concrete defect semantic segmentation model is obtained by training a full convolutional neural network according to marked concrete surface image data. According to the scheme, an effective means is provided for pixel-level automatic identification and detection of concrete defect images, the identification efficiency is high, and the problems that manual inspection consumes time and labor, and defect details and the development process are difficult to control are solved.

Description

technical field [0001] The invention relates to the field of concrete safety monitoring, in particular to a concrete defect detection method, device and storage medium based on a fully convolutional neural network. Background technique [0002] In many cases, it is necessary to monitor and detect the safety status of concrete. Among them, patrol inspection is an indispensable part of concrete safety monitoring. Through patrol inspection, cracks, leakage, calcium analysis, exposed ribs, damage, etc. can be found in time to affect the safety of concrete structures. Defects. However, patrol inspections currently mainly rely on manual work, which inevitably has problems such as time-consuming and labor-intensive inspections, and difficulties in controlling defect details and development processes. [0003] Image semantic segmentation technology can automatically obtain the category information of each pixel in the image, which provides an effective technical means for automatic...

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): G06T7/00G06K9/34G06K9/62G06N3/04G06N3/08
CPCG06T7/0004G06N3/08G06T2207/30132G06T2207/20081G06T2207/20084G06V10/267G06N3/045G06F18/24G06F18/214
Inventor 张军刘海峰朱聪徐波彭敏陆新洁
Owner STATE GRID HUNAN ELECTRIC POWER
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