Bridge crack detection method based on deep learning framework

A technology of deep learning and detection methods, applied in the field of image processing, can solve the problems of many noise points, high missed detection rate, complex parameter debugging, etc., to eliminate the influence of noise, alleviate the problem of missed detection, and strengthen the generalization ability.

Pending Publication Date: 2021-10-15
NORTHWESTERN POLYTECHNICAL UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to propose a bridge crack detection method based on deep learning to solve the shortcomings of the traditional crack detection method based on edge detection, such as complex parameter debugging, many noise points, and high missed detection rate

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
  • Bridge crack detection method based on deep learning framework
  • Bridge crack detection method based on deep learning framework
  • Bridge crack detection method based on deep learning framework

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0056] The program of the algorithm of this embodiment is based on the deep learning open source framework keras 2.3.1 and tensorflow-gpu2.2.0, and is developed in Python language; the hardware environment of the experiment is Intel i7 processor, NVIDIA GeForce RTX2060 graphics card.

[0057] In this embodiment, a total of 189 images with original bridge cracks were collected, and the original bridge crack images were marked with cracks to obtain a large labeled image. Synchronously cut the large label image and the original bridge crack image. After cutting, 9788 images in the training set and 6208 images in the verification set were obtained through screening, classification and amplification, both of which were 224*224 in size. Note that the image data must first be classified into the training set and the verification set before being amplified, otherwise it will cause data leakage and affect the training effect. The fracture data set is obtained after the amplification, ...

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 bridge crack detection method based on a deep learning framework, and the method comprises the following steps: 1, obtaining a bridge image, and selecting an image containing a crack as an original bridge crack image; 2, preprocessing the original bridge crack image to obtain an image data set; 3, inputting the image data set into a segmentation model for training; and 4, inputting a to-be-detected image into the segmentation model to complete feature crack extraction. The defects that a traditional crack detection method based on edge detection is complex in parameter debugging, multiple in noise points and high in omission ratio are overcome.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a bridge crack detection method based on a deep learning framework. Background technique [0002] In recent years, due to the rapid growth of traffic flow, it has caused great pressure on the operation safety of bridges. Due to the long construction time, poor design performance, harsh natural environment and other reasons, bridge collapse incidents have occurred frequently in recent years, causing great losses. Existing data show that the vast majority of concrete bridge damage is related to bridge cracks. Therefore, the detection of bridge cracks is very important for the maintenance of concrete bridges. For a long time, the crack detection of bridges mostly adopts local non-destructive detection methods and overall detection methods mainly based on ultrasonic and infrared detection. The image-based detection method is mainly based on manual detection. This method is...

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): G06T7/00G06T7/136G06T5/00G06K9/62G06N3/04G06N3/08
CPCG06T7/0004G06T7/136G06N3/08G06T2207/10004G06T2207/20021G06T2207/20081G06T2207/20084G06T2207/30108G06N3/047G06N3/045G06F18/214G06T5/70
Inventor 张夷斋姬文鹏黄攀峰闫雨晨李鹏辉杨奇磊章勇威
Owner NORTHWESTERN POLYTECHNICAL UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap