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

Railway wagon lower pull rod breaking fault detection method based on deep learning

A railway freight car and deep learning technology, applied in the field of image recognition, can solve the problems of low detection accuracy and detection efficiency, achieve safe and efficient operation, improve detection efficiency, and ensure detection accuracy

Inactive Publication Date: 2020-09-11
HARBIN KEJIA GENERAL MECHANICAL & ELECTRICAL CO LTD
View PDF11 Cites 14 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to solve the problem of low detection accuracy and low detection efficiency in the manual detection method for the detection of the failure of the lower rod, and propose a method for detecting the failure of the lower rod of railway wagons based on deep learning

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
  • Railway wagon lower pull rod breaking fault detection method based on deep learning
  • Railway wagon lower pull rod breaking fault detection method based on deep learning
  • Railway wagon lower pull rod breaking fault detection method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0020] Specific implementation mode 1. Combination figure 1 This embodiment will be described. A deep learning-based method for detecting the breakage of the lower rod of a railway freight car described in this embodiment, the method is specifically implemented through the following steps:

[0021] Step 1. Collect the image of the visual parts of the truck through the image acquisition equipment arranged at the detection station, and save the collected image in a designated location;

[0022] Step 2. After reading the collected image from the designated position, roughly locate the area of ​​the pull-down rod in the collected image, and intercept the sub-image of the location of the pull-down rod;

[0023] Step 3. Perform data amplification processing on the sub-image at the location of the intercepted pull-down rod, and then mark the image obtained by the data amplification processing to obtain a training data set;

[0024] Step 4, input the obtained training data set into ...

specific Embodiment approach 2

[0034] Embodiment 2. This embodiment differs from Embodiment 1 in that: after the sub-picture where the pull-down rod is located is cut out in step 2, the background color of the sub-picture where the pull-down rod is located is set to white.

[0035] In this embodiment, in order to eliminate the interference of the background around the pull-down rod on the detection of the broken position, after obtaining the sub-image of the location of the pull-down rod, the background color near the pull-down rod is set to white to reduce the interference information for identification.

specific Embodiment approach 3

[0036] Specific Embodiment 3. The difference between this embodiment and specific embodiment 2 is: the data amplification processing is performed on the intercepted sub-picture at the location of the pull-down bar, and the data amplification method adopted includes image rotation and image random cropping. , image horizontal flip, image vertical flip, image stretch and image zoom.

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 railway wagon lower pull rod breaking fault detection method based on deep learning, and belongs to the technical field of image recognition. According to the invention, theproblems of low detection precision and low detection efficiency existing in the detection of the pull-down rod breaking fault by adopting a manual detection mode are solved. According to the invention, a deep learning model is built on the basis of a convolutional neural network, automatic identification and detection are carried out on a pull-down rod breaking fault occurring in an image, if thepull-down rod breaking fault is detected in the image, position information of the breaking fault is generated and uploaded to a train inspection operation platform, and after manual rechecking, a faulted train is correspondingly disposed in time, so that safe and efficient operation of railway freight transportation is guaranteed. An automatic image recognition mode is used for replacing a puremanual vehicle inspection mode, and the detection efficiency can be effectively improved while the detection precision is guaranteed. The method can be applied to fracture fault detection of the lowerpull rod of the wagon.

Description

technical field [0001] The invention belongs to the technical field of image recognition, and in particular relates to a deep learning-based detection method for a broken down rod of a railway freight car. Background technique [0002] The lower rod is located at the brake beam station of the railway freight car, which is an important part to ensure the safe and stable operation of the freight car. If the lower rod breaks down and loses its original ability to ensure the safety of the freight car, it is easy to cause major accidents such as derailment and overturning of the vehicle. The purely manual detection method is likely to cause missed detection, and the detection accuracy is difficult to guarantee. At the same time, the detection efficiency of the manual detection method is low. Since the quality and efficiency of the inspection work are difficult to be effectively guaranteed, the safety hazards of railway freight car operation have been increased. Therefore, it 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): G06K9/62G06N3/04G06T7/00B61H13/02B61H13/20G01M17/08
CPCG06T7/0002B61H13/02B61H13/20G01M17/08G06N3/045G06F18/241G06F18/214
Inventor 于洋
Owner HARBIN KEJIA GENERAL MECHANICAL & ELECTRICAL CO LTD
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