An identification method for the failure of the handle of the corner plug door of the railway freight car

A technology for knuckle cocks and railway wagons, applied to railway car body parts, railway signals, neural learning methods, etc., can solve the problem of low accuracy of knuckle cock handles, achieve accurate pixel information of key points, and improve Accuracy, the effect of improving accuracy

Active Publication Date: 2022-04-19
HARBIN KEJIA GENERAL MECHANICAL & ELECTRICAL CO LTD
View PDF7 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 solve the problem of low accuracy of existing methods for detecting out-of-position faults of knuckle cock handles, and propose a method for identifying out-of-position faults of knuckle cock handles of railway wagons

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
  • An identification method for the failure of the handle of the corner plug door of the railway freight car
  • An identification method for the failure of the handle of the corner plug door of the railway freight car
  • An identification method for the failure of the handle of the corner plug door of the railway freight car

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0057] Specific implementation mode 1. Combination figure 1 and figure 2 This embodiment will be described. In this embodiment, a method for identifying a failure of a railway wagon knuckle cock handle not in the correct position, the method is specifically implemented through the following steps:

[0058] Step 1, obtaining the side image of the railway freight car, and intercepting the image of the inter-hook difference station area from the obtained image;

[0059] Step 2. Input the image of the inter-hook difference station area intercepted in step 1 into the YOLOX target detection network, and use the YOLOX target detection network to predict the pixel coordinate information of the bounding box of the knuckle plug part in the inter-hook difference station area image, and then Using the predicted pixel coordinate information, cut out the image of the knuckle plug part area from the input inter-hook difference station area image;

[0060] Step 3, extracting the corner fe...

specific Embodiment approach 2

[0066] Specific implementation mode two: the difference between this implementation mode and specific implementation mode one is: the specific process of step three is:

[0067] Step 31. Based on the Gaussian scale space theory, use parameters of different scales to construct the Gaussian filter kernel (the specific scale can be set manually according to the needs), and then build a multi-scale based on the Gaussian filter kernel and the cropped corner plug part area image space image;

[0068] Before performing step 31, it is also possible to perform contrast enhancement on the cropped corner plug part area image, and then establish a multi-scale spatial image based on the contrast-enhanced image;

[0069] Step 32. For the spatial image of each scale, use the Harris corner detection algorithm to find the corners of the image, and record the set of corners of the spatial images of each scale as the initial candidate corner set P;

[0070] Step 33, for the spatial image of eac...

specific Embodiment approach 3

[0077] Specific embodiment three: the difference between this embodiment and specific embodiment two is: the corner point positioning binary image is generated based on the corner points in the final corner point set F, which is specifically:

[0078] For any pixel in the obtained corner cock part area image, if the Manhattan distance between the pixel and at least one corner point in the final corner point set F is less than or equal to the threshold b (the value of the threshold b is artificially set according to the actual image size set), then set the pixel value of the pixel point to 255, otherwise, set the pixel value of the pixel point to 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 method for identifying faults of incorrect position of a door handle of a folded-angle plug of a railway freight car, belonging to the field of recognition of the incorrect position of a door handle of a folded-angle plug. The invention solves the problem that the existing method has low accuracy in detecting the malfunction of the door handle of the folded-angle plug. The technical scheme adopted by the present invention is as follows: step 1, cut out the image of the interlocking difference station area from the acquired image of the side of the railway freight car; step 2, use the YOLOX network to predict the pixel coordinate information of the bounding box of the corner plug door component, from the mutual Crop out the image of the corner plug door area from the image of the hook difference station area; step 3, extract the corner feature of the image of the corner plug door component area, and generate a corner positioning binary image; Step 4, combine the area image of the corner plug door component with the corner points The binary image is located for channel merging, and the key point detection network is used to detect the coordinates of the key points in the merged image; step 5, fault identification is performed according to the detection results of the key point coordinates. The invention can be applied to the fault detection of the incorrect position of the handle of the hinged plug door.

Description

technical field [0001] The invention belongs to the technical field of out-of-position identification of a knuckle cock handle of a railway freight car, and in particular relates to an identification method for an out-of-position fault of a knuckle cock handle of a railway freight car. Background technique [0002] The out-of-position fault of the knuckle cock handle of railway freight cars is a fault that seriously endangers the traffic safety. In the fault detection of the out-of-position knuckle cock door handle, manual inspection of images is a common method for fault detection. However, due to the fact that the inspectors are prone to fatigue and omissions during the work process, resulting in missed inspections and wrong inspections, which affect driving safety. The method of automatic image recognition can improve the detection efficiency and stability. In recent years, deep learning and artificial intelligence have continued to develop, and the technology has continu...

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): G06V10/44G06V10/70G06N3/02B61L15/00
CPCB61L15/0081G06N3/08G06N3/045
Inventor 杨宇
Owner HARBIN KEJIA GENERAL MECHANICAL & ELECTRICAL CO LTD
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