Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

High-speed railway contact network image recognition method combining YOLOv3 and SENet

A technology for image recognition and high-speed railways, applied in character and pattern recognition, scene recognition, neural learning methods, etc., can solve problems such as accumulation

Inactive Publication Date: 2020-08-25
ZHEJIANG UNIV +1
View PDF4 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although the absence of a spare mother is temporarily insufficient to cause insulator connection failure, if it is not discovered and supplemented in time, the risk of inducing connection failure will gradually accumulate

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
  • High-speed railway contact network image recognition method combining YOLOv3 and SENet
  • High-speed railway contact network image recognition method combining YOLOv3 and SENet
  • High-speed railway contact network image recognition method combining YOLOv3 and SENet

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0058] The present invention will be described in further detail below in conjunction with the accompanying drawings and specific embodiments.

[0059] The embodiment and implementation process thereof implemented according to the complete method of the content of the present invention are as follows:

[0060] 1) Capture and collect catenary sample images containing U-shaped hoops through the on-board camera (such as Figure 5 shown), the U-shaped hoop contained in the catenary sample image is divided into two types: the missing spare mother and the non-missing spare mother (such as Image 6 shown);

[0061] There are a total of 1081 experimental pictures, of which 865 are used for training, 339 of which are in the absence of standby mothers, 526 in the absence of standby mothers, and the remaining 216 pictures are used as test set pictures. There are 88 pictures in the missing state, and 128 pictures in the non-missing state of the standby mother. The size of the original p...

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 high-speed railway contact network image recognition method combining YOLOv3 and SENet. The method comprises: detecting and positioning a U-shaped hoop in an insulator area in the high-speed railway overhead line system image through a YOLO v3 model, and then deducting the positioned U-shaped hoop from the original image and zooming the U-shaped hoop; and carrying out classification and identification on the scaled U-shaped hoop image through an SENet model to obtain a U-shaped hoop standby mother missing image, and finally carrying out defect positioning on the U-shaped hoop image by adopting class activation mapping. According to the method, the problem of high resource occupancy rate of small object identification on the high-resolution image is solved, and theidentification result based on the Xinjiang high-speed railway overhead contact system U-shaped hoop standby bus missing scene shows that the identification method is high in accuracy and can be applied to high-speed railway overhead contact system U-shaped hoop standby bus missing identification.

Description

technical field [0001] The invention relates to a method for identifying the state of a catenary U-shaped hoop, in particular to a high-speed railway catenary image recognition method combined with YOLO v3 and SENet. Background technique [0002] High-speed railway catenary is a special form of transmission line erected along the high-speed railway line to supply power to EMUs. The current that high-speed trains rely on is transmitted through the catenary on the upper end of the locomotive. It is composed of contact suspension, supporting device, positioning device, pillar and foundation. Since the catenary has no backup and needs to withstand the dynamic flow of the pantograph of the high-speed train, the condition of its equipment directly affects the driving. [0003] As a part of the high-speed railway catenary, the U-shaped hoop in the insulator area plays a decisive role in the stable operation of the catenary. Although the absence of a spare mother is temporarily in...

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): G06K9/62G06K9/00G06N3/08
CPCG06N3/08G06V20/13G06F18/24G06F18/214
Inventor 齐冬莲陈强闫云凤刘俊宏郑伊季瑞松
Owner ZHEJIANG UNIV
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
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