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

Improved AlexNet-based burning arc identification method

A recognition method and arc burning technology, applied in the field of image recognition, can solve the problems of many parameters and low training efficiency, and achieve the effect of high recognition accuracy, deep network structure and reduced number of parameters

Active Publication Date: 2018-07-27
CHENGDU NAT RAILWAYS ELECTRICAL EQUIP
View PDF5 Cites 16 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] AlexNet is a classic that was proposed in 2012 and achieved the best results in ImageNet that year, but the traditional AlexNet model has too many parameters and the training efficiency is low. This invention proposes an improved AlexNet model and applies it to Identification of catenary arcing

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
  • Improved AlexNet-based burning arc identification method
  • Improved AlexNet-based burning arc identification method
  • Improved AlexNet-based burning arc identification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0029] In order to have a clearer understanding of the technical features, purposes and effects of the present invention, the specific implementation manners of the present invention will now be described with reference to the accompanying drawings.

[0030] Such as figure 1 As shown, an arcing identification method based on improved AlexNet, this method is applied to arcing identification but not limited to arcing identification, including the following steps:

[0031] S1. Establish a convolutional neural network and obtain a training model;

[0032] S2. Acquiring images;

[0033] S3. Input the collected images into the training model for arc recognition;

[0034] Further, step S1 includes:

[0035] S11. Establish and initialize a convolutional neural network;

[0036] S12. Extract the image from the server storing the image data and perform preprocessing, including clipping, compression, de-averaging, and normalization;

[0037] S13. Add a label to each image, the image...

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 an improved AlexNet-based burning arc identification method. The method comprises the following steps of S1, building a convolutional neural network and obtaining a training model; S2, obtaining an image; and S3, inputting the obtained image to the training model for performing image identification, wherein the convolutional neural network is improved AlexNet, the structureof the improved AlexNet sequentially comprises an input layer, multiple convolutional layers connected in sequence, a full connection layer and an output layer, the convolutional layers connected with the input layer adopt MXM convolutional kernel architectures, and the test of the convolutional layers adopt 1XM and MX1 convolutional kernel architectures. According to the method, MXM convolutional kernels in original AlexNet are replaced with 1XM and MX1 convolutional kernels connected in series, so that a parameter quantity is greatly reduced; a nonlinear layer is added, so that the networkstructure is deeper; and the model is simplified, so that the training time is greatly shortened, the training efficiency is improved, and the identification accuracy is higher than that of the original AlexNet.

Description

technical field [0001] The invention relates to the field of image recognition, in particular to an arc recognition method based on improved AlexNet. Background technique [0002] The arcing of the railway catenary is due to the dancing of the catenary, the bouncing of the pantograph, and the voltage exceeding the tolerance of the air, which makes the air ionized and becomes a conductor and then generates an arc; arcing will cause the operation of the electric locomotive to be unstable, so that the electric locomotive will Intermittent, causing abnormal deceleration and acceleration during the train operation, increasing the discomfort during the journey, through the arcing alarm, the faulty contact wire or pantograph can be repaired in time, reducing the risk of railway power supply safety accidents occurrence; the arcing alarm needs to be confirmed by manual interpretation after it is transmitted back to the data terminal. , effectively identifying the arc can greatly red...

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): G06K9/00G06N3/04G06N3/08
CPCG06N3/084G06V20/10G06N3/045
Inventor 范国海张娜何洪伟何进
Owner CHENGDU NAT RAILWAYS ELECTRICAL EQUIP
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