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

Method for detecting position and type of workpiece on subway vehicle side inspection image

A subway and workpiece technology, applied in the field of machine vision industrial inspection, can solve a large number of manual operations and other problems, achieve the effect of reducing labor intensity, reducing difficulty, and facilitating the analysis of workpiece status

Pending Publication Date: 2021-02-26
成都川哈工机器人及智能装备产业技术研究院有限公司 +1
View PDF0 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The invention is a method for detecting the position and type of the workpiece on the inspection image on the side of the subway car. The technical problem to be solved is to solve the problem that a large number of manual operations are required in the inspection process of the subway train. The method of detecting the position of different workpieces on the scanned pictures of the train side and identifying the type of the workpiece, the output results of which can be used for further machine learning and intelligent analysis, so as to realize the functions of intelligent subway inspection and automatic train body state detection

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
  • Method for detecting position and type of workpiece on subway vehicle side inspection image
  • Method for detecting position and type of workpiece on subway vehicle side inspection image
  • Method for detecting position and type of workpiece on subway vehicle side inspection image

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0030] The present invention is a method for detecting the position and type of the workpiece on the inspection image on the side of the subway car. The model training embodiment mainly describes that in order to obtain the workpiece detection model, the subway body image data is acquired, the workpiece target is marked in the image, and the YOLOv4 deep learning network is built. And training, as well as model output, the specific implementation process is as follows:

[0031] 1) Install two industrial cameras on both sides of the subway track to scan the body of the subway train, output the original scanned picture with a resolution of 65535×1808, and divide the original picture into sub-pictures of 1808×1808 size for storage.

[0032] 2) Use the Colabeler labeling tool to label each workpiece that appears on the picture. In this embodiment, it mainly includes different screws, upper brake pads, lower brake pads and iron wires. There are 10 workpiece target types in total. The...

Embodiment 2

[0036] Such as figure 1 As shown, the invention builds a subway train workpiece detection model based on the YOLOv4 target detection network architecture. YOLO is a target detection model based on deep learning. The present invention builds a new subway workpiece detection model based on YOLOv4. The detection algorithm flow is as follows figure 1 As shown, the steps of model training are as follows:

[0037]a) First, use the pre-trained weights of the CSPDarknet53 network to initialize the backbone network of YOLOv4. The weights are trained using MS COCO target detection data, which can detect 80 objects in the MS COCO data set. In order to adapt to the application of subway body workpiece detection, it is necessary The output classification of YOLOv4 network is revised to the kind number of workpiece (being 10 kinds in the present invention), thereby can satisfy the condition of transfer learning;

[0038] b) For a sub-picture to be trained, its resolution is 1808×1808, and ...

Embodiment 3

[0052] After the above-mentioned training method is trained in the subway car body workpiece detection data set, the present invention can obtain a trained workpiece detection model, which can run under the Darknet deep learning framework, and output a length of S × S × (B × 5+ C) prediction tensor, in order to detect the position of each workpiece and identify the type of the workpiece, the present invention introduces a non-maximum suppression algorithm (non maximum suppression, NMS) to find the target frame and confidence from the prediction tensor. The steps are as follows:

[0053] 1) Set a threshold and filter out all target candidate bounding boxes smaller than this threshold;

[0054] 2) Select the one with the largest target confidence of a certain workpiece candidate bounding box, which is box_best;

[0055] 3) Calculate the IOU (Intersection over Union) of box_best and other artifact candidate bounding boxes, that is, the intersection of the two boxes is more than ...

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 detecting the position and type of a workpiece on a subway train side inspection image. The method employs a software algorithm to replace manual operation, employs a subway train side workpiece detection and recognition model, can be used for the development of an intelligent subway inspection system, and replaces a manual observation detection mode. Accordingto the method, the YOLOv4 is used as a basis to construct the subway train side workpiece positioning and recognition model, the YOLOv4 has a very good effect in target detection application, and thecapability is migrated to subway inspection application, so that the accuracy of train side workpiece positioning can be effectively improved; and powerful support is provided for subsequent defect analysis, by using the detection model provided by the invention, a workpiece can be quickly and accurately positioned from a subway vehicle body scanning picture, a sub-picture area can be segmented,and the output picture only comprises a workpiece with a known model, so that the subsequent analysis on the state of the workpiece is greatly facilitated, and the difficulty of subsequent analysis isreduced.

Description

technical field [0001] The invention relates to the field of machine vision industrial detection, in particular to a method for detecting the position and type of a workpiece on a subway car side inspection image. Background technique [0002] In my country, with the expansion of city scale and the continuous increase of urban population, the traffic pressure in the city is increasing. Therefore, many cities are vigorously building subway systems. my country is currently in a period of rapid development of rail transit, and the opening and operation of a large number of subway lines has put forward higher and higher requirements for the management and maintenance of rail transit equipment. Among them, the daily inspection of the train to ensure that the workpieces on the train are in a normal state is an important guarantee to ensure the safety of subway traffic. In order to ensure that the workpieces on the subway train are in a normal state, it is necessary to frequently 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
IPC IPC(8): G06K9/00G06K9/46G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V20/20G06V10/40G06N3/045G06F18/214
Inventor 张朝龙周治宇周帆王健王贵东朱均梁海清朱冬
Owner 成都川哈工机器人及智能装备产业技术研究院有限公司
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