A fault image recognition method for railway wagon floor damage

A technology for image recognition and railway wagons, applied in image enhancement, image analysis, image data processing, etc., can solve problems such as fatigue, omission, and low detection rate of inspection personnel, and achieve improved recognition recall rate, easy training, The effect of improving detection efficiency

Active Publication Date: 2020-11-27
HARBIN KEJIA GENERAL MECHANICAL & ELECTRICAL CO LTD
View PDF9 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 detection rate due to the fact that inspection personnel are prone to fatigue and omissions during the work process by manually inspecting images in the prior art, and propose a broken floor of railway freight cars Fault Image Recognition Method

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 fault image recognition method for railway wagon floor damage
  • A fault image recognition method for railway wagon floor damage
  • A fault image recognition method for railway wagon floor damage

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0040] Specific Embodiment 1: This embodiment will be described in detail with reference to the figure. A method for image recognition of damaged fault images on the floor of railway wagons described in this embodiment includes the following steps:

[0041] Step 1: Obtain high-definition line scan images of passing trucks;

[0042] Step 2: Cut out the area of ​​the part to be recognized from the image according to prior knowledge, and establish a sample data set;

[0043] Step 3: Perform data amplification on the sample data set;

[0044] Step 4: Mark the images in the dataset;

[0045] Step 5: Generate a data set from the original image and labeled data, and train the model;

[0046] Step 6: Use the SEGNET-UNET network to segment the image, and mark each segmented part;

[0047] Step 7: For the floor segmentation result, divide the image into multiple fault areas according to the contour information. For each fault area, judge whether there is a floor damage fault accordin...

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

A fault image recognition method for damaged floors of railway wagons, which relates to the technical field of freight train detection. In the prior art, manual inspection of images is used for fault detection. Since the inspectors are prone to fatigue and omissions during the work process, the detection In order to solve the problem of low detection rate, automatic image recognition is used instead of manual detection to improve detection efficiency and accuracy. Apply the deep learning algorithm to the automatic identification of floor damage faults to improve the stability and accuracy of the overall algorithm. In order to reduce the impact of rain on the recognition rate, in addition to the normal area and the damaged area, the beam body area on the floor and foreign objects such as weeds are marked separately to improve the recognition accuracy. Combining the U‑NET model with the SEGNET model for fault identification. Compared with U‑NET, SEGNET‑UNET has fewer parameters and is easier to train. Compared with SEGNET, SEGNET‑UNET imitates U‑NET and adds jump connections, pays more attention to details than SEGNET, and can better extract boundary information.

Description

technical field [0001] The invention relates to the technical field of freight train detection, in particular to a fault image recognition method for floor damage of railway freight cars. Background technique [0002] The fault of truck floor damage is a common fault that endangers driving safety. It is characterized by a wide range of identification, complex background and changeable fault forms. At present, the dynamic vehicle inspection operation is still carried out manually by looking at pictures one by one. The main problems are as follows: due to the influence of personnel quality and sense of responsibility, errors and omissions occur from time to time, and the quality of the operation is difficult to guarantee; a large number of dynamic vehicle inspection personnel are required , low efficiency and huge labor costs. However, image processing and deep learning methods are used for automatic identification of floor damage faults, and manual confirmation of the alarm ...

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): G06T7/00G06T7/11
CPCG06T7/0004G06T7/11G06T2207/20081G06T2207/20084G06T2207/30164G06T2207/30204
Inventor 高恩颖
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