Underground space pipeline abnormality detection method based on 3D convolutional neural network

A convolutional neural network, underground space technology, applied in the field of computer vision and underground pipelines, can solve the problem of not considering continuous frame motion information, etc.

Inactive Publication Date: 2019-07-19
CHINA UNIV OF PETROLEUM (EAST CHINA)
View PDF2 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

A simple way to apply CNN is to use CNN to identify each frame, but this method does not take into account the motion information between consecutive frames

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
  • Underground space pipeline abnormality detection method based on 3D convolutional neural network
  • Underground space pipeline abnormality detection method based on 3D convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0022] In order to make the purpose, technical solution and advantages of the present invention clearer, the implementation manners of the present invention will be further described in detail below.

[0023] Through computer vision and deep learning technology, especially the three-dimensional convolutional neural network structure that has become popular in the field of computer vision in recent years, the accuracy of computers in the field of image and video recognition has been greatly improved.

[0024] The described computer vision-based underground space pipeline anomaly detection method may comprise the following steps:

[0025] Step 1: Prepare two sample sets of underground space pipelines A and B required for training; the A sample set is a normal image, and the B sample set is an abnormal pipeline image. Image types should be diverse, and the amount of data for the two samples should be the same.

[0026] Step 2: Preprocess the sample set, modify it to a uniform si...

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 provides a computer vision-based underground space pipeline abnormality detection method. The method is characterized in that the sampled video data is marked, the video frames are extracted, a three-dimensional convolution kernel method (3D CNN) in deep learning is adopted to train a sample set needed by a convolutional neural network, the sample set is preprocessed, and the size ismodified to be 200*200 in batches. By designing a three-dimensional convolution kernel network structure, and training, aiming at the adopted video data, roughly selecting a defect frame, then sampling a video at every 15ms time interval, and inputting a sampling frame into a neural network, whether an abnormal condition exists or not can be detected.

Description

technical field [0001] The invention relates to the field of underground pipelines in underground spaces, in particular to an abnormal detection method for pipelines in underground spaces based on the field of computer vision. Background technique [0002] With the continuous acceleration of the urbanization process, most cities have experienced limited spatial development to varying degrees, specifically manifested in traffic congestion, serious shortage of public service infrastructure, and deterioration of the urban space environment. Urban Syndrome". The limited space seriously hinders the sustainable development of modern cities. The development and utilization of urban underground space is an effective way to solve the shortage of urban construction space and improve the comprehensive function of the city, and it is an effective measure to improve the land utilization rate and promote the sustainable development of the city. [0003] Although many large and medium-si...

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): G06K9/00G06N3/04
CPCG06V20/10G06V20/40G06N3/045
Inventor 庞善臣孟璠韩宁生姚加敏董立媛
Owner CHINA UNIV OF PETROLEUM (EAST CHINA)
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