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

Turnout and non-turnout rail fastener positioning method based on deep learning

A technology of deep learning and positioning methods, applied in neural learning methods, computer components, optical devices, etc., can solve the problems of complex interference of fastener types, difficulty in obtaining good results, etc., and achieve high-precision and stable positioning detection effect, suppression of complex features, effect of raising attention

Active Publication Date: 2021-10-15
SOUTHEAST UNIV
View PDF3 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In fact, there are many types of fasteners in the turnout section and the interference is complex. It is difficult for the usual target detection method to achieve good results for the fasteners in the turnout section.

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
  • Turnout and non-turnout rail fastener positioning method based on deep learning
  • Turnout and non-turnout rail fastener positioning method based on deep learning
  • Turnout and non-turnout rail fastener positioning method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0039] The image used in this embodiment is the image data sample collected by the high-definition camera carried by the inspection vehicle to shoot the track, wherein the lens needs to be kept parallel to the plane where the track is located when shooting. The images detected in this example are from the non-turnout road section and the turnout road section scene, such as figure 1 It is a typical non-turnout section (a) and a turnout section image (b).

[0040] This embodiment takes figure 1 Take the images of non-turnout and turnout road sections shown in as an example. These images are collected by the camera on the inspection vehicle. The image data set contains a total of 1200 images. The process of the fastener positioning method is as follows Figure 5 shown.

[0041] After the image collection is completed, the collected image data sets are manually marked. Among them, one kind of fastener to be detected is marked on the non-turnout road section, and three kinds of f...

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 relates to a turnout and non-turnout rail fastener positioning method based on deep learning, and the method comprises the following steps: obtaining a rail image through an inspection vehicle with a high-definition camera, and carrying out the manual marking of an image data set; building an improved Faster R-CNN deep learning model, and optimizing the model by multi-scale feature fusion, attention module addition and the like; according to the marked actual size of the to-be-detected target, modifying a predefined anchor frame in the model; training the improved Faster R-CNN model, and selecting an optimal model; and inputting a to-be-detected image into the trained model to obtain an accurate and stable fastener positioning result. The invention can be suitable for images of turnout road section and non-turnout road section scenes, and can accurately position and detect various types of fasteners at the same time.

Description

technical field [0001] The invention belongs to the field of track defect detection, in particular to a method for locating track fasteners of turnouts and non-turnouts based on deep learning. Background technique [0002] In my country's railway track fasteners, rail fasteners and bolts are important components. Their main function is to fix the rails on the sleepers, so that the rails can maintain a stable distance, and at the same time avoid horizontal or vertical during operation. move. Therefore, the normal state of the fasteners plays a very important role in the safety of the track. When the fasteners are abnormal, such as breakage, displacement, torsion, loosening, loss, etc., it will cause huge safety hazards to the normal operation of the railway, and even cause serious problems. Therefore, it is very necessary to detect the fasteners in time. [0003] The traditional detection method of railway track fasteners in our country is mainly carried out through manual i...

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): G06T7/00G06T7/73G01B11/00G06K9/62G06N3/08
CPCG06T7/0004G06T7/73G06N3/08G01B11/00G06T2207/20081G06T2207/20084G06F18/214G06F18/2415G06F18/253
Inventor 路小波彭鑫冯魏运
Owner SOUTHEAST 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