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

A Crack Detection Method for Asphalt Pavement Image

A technology of asphalt pavement and detection method, which is applied in the field of deep learning, can solve problems such as the inability to locate cracks and fail to consider the global information of the picture well, and achieve the effect of strong generalization ability

Active Publication Date: 2019-12-10
SOUTHWEST JIAOTONG UNIV
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Because the detection is based on sub-blocks, these methods cannot take into account the global information of the picture well, and cannot locate cracks on the original image.

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
  • A Crack Detection Method for Asphalt Pavement Image
  • A Crack Detection Method for Asphalt Pavement Image
  • A Crack Detection Method for Asphalt Pavement Image

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0035] Step 1. Using a rectangular frame with a line width of 1 pixel, mark the image with the obtained pavement cracks, and draw the precise position of the crack in the image, that is, Ground Truth;

[0036] Step 2. Train the Crack-Faster-RCNN model

[0037] Step 2.1: Take 230 road surface crack images marked with Ground Truth as input, and pass them into the deep residual neural network ResNet101 trained with the ImageNet dataset. Extract image features through 100 convolution operations, 100 ReLU function activations, and 2 pooling operations, and output the feature map Feature map;

[0038] Step 2.2: Input the feature map Feature map to the region proposal network RPN. The RPN network first performs one convolution and one pooling operation on the Feature map, and then uses the anchor strategy to generate 9 on each pixel of the feature map. Anchor, where anchors are the size: {128 pixels*128 pixels, 256 pixels*256 pixels, 512 pixels*512 pixels}, anchors aspect ratio: {0....

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 crack detection method for an asphalt pavement image. The method comprises the following steps of inputting marked training data into a Cache-Fast-RCNN model; secondly, training a deep residual convolutional neural network ResNet101 to extract image features, classifying the cracks and backgrounds by a regional suggestion network RPN, classifying the cracks by a classification network Classifier, and carrying out frame position regression; and finally, storing the trained Cache-Fast-RCNN model parameters. The trained Cache-Fast-RCNN model is used for detecting a new pavement crack image, parameters of the model are loaded firstly, and features of the image are extracted; then crack detection is carried out on the sample; and finally the position information of thecrack is marked. By constructing the deep neural network framework for pavement crack detection, the asphalt pavement crack detection method which can be applied to an actual scene is realized.

Description

technical field [0001] The invention relates to the field of deep learning, in particular to the detection of road surface cracks in images using a deep learning method. Background technique [0002] China's highway traffic is in a period of rapid development. While building, rebuilding and expanding highways, especially high-grade highways, the importance and urgency of road surface maintenance and management are becoming increasingly prominent. Asphalt materials have been widely used as highway construction materials in my country because of their advantages such as fast construction, convenient maintenance, and strong adaptability. For asphalt pavement, cracks are the most common and important form of disease manifestation. Pavement cracks will largely induce roadbed or base-level diseases. For example, rainwater enters the base layer along cracks and causes base-level instability, thereby aggravating road surface diseases. At the same time, if the cracks are not repaire...

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 Patents(China)
IPC IPC(8): G06T7/00
Inventor 彭博苟聪李天瑞唐堂
Owner SOUTHWEST JIAOTONG UNIV
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