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

Common invader object detection and identification method of power transmission corridor based on deep learning

A target detection and recognition method technology, applied in character and pattern recognition, instruments, computer parts and other directions, can solve problems such as economic losses, hidden dangers of transmission line safety, restricted external environment, etc., to achieve high accuracy and robustness , Good accuracy and stability, good detection and recognition effect

Inactive Publication Date: 2017-05-31
CHENGDU TOPPLUSVISION TECH CO LTD
View PDF5 Cites 53 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] With the development of science and technology and economy, the construction of the national grid is also developing rapidly, and the number of transmission lines erected is increasing. However, the transmission lines have been exposed to the wild for a long time, and the terrain of the line corridors is complex and changeable, which is easy to be invaded by foreign objects. This is the transmission line. Bring huge potential safety hazards, and in severe cases may lead to major transmission equipment accidents, resulting in huge economic losses and even casualties
[0003] For the monitoring of transmission corridors, the traditional method is to use helicopters to conduct line inspections, but this method is costly and is easily restricted by weather, environmental and other conditions
Afterwards, line inspection robots appeared, but this method is not only complicated in system design, but also limited by the external environment
Recently, video surveillance systems have been researched and applied a lot, but for transmission corridor scenarios, most of the existing surveillance products are limited to the detection of one or several types of intruders, and cannot provide comprehensive protection for transmission lines. Can not meet the actual application needs

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
  • Common invader object detection and identification method of power transmission corridor based on deep learning
  • Common invader object detection and identification method of power transmission corridor based on deep learning
  • Common invader object detection and identification method of power transmission corridor based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment 1

[0027] A method for detecting and recognizing common intrusive objects in power transmission corridors based on deep learning. The specific method is: in the training phase, using the deep learning method to learn the pictures of foreign objects intrusion acquired by the video capture device, and obtain the required network through learning Model; in the use phase, the actual monitored screen is transferred to the network model, and finally the detection and identification of intrusive objects is completed.

specific Embodiment 2

[0029] On the basis of specific embodiment 1, such as figure 1 As shown, the specific method steps of the training phase are:

[0030] S11. Extracting sub-images containing intrusions from the original image of the power transmission corridor from the video images collected by the camera in real time; performing scaling processing on the extracted sub-images, and using a uniform size to form a training data set;

[0031] S12. Calibrate the detection frame and object category information of the intrusive object in the training data set;

[0032] S13. Input the calibrated data into the designed convolutional neural network (CNN), and forward it to obtain the detection frame information output by the model and the category information of the sample;

[0033] S14. Calculate the regression loss function value of the detection frame information output result and the actual detection frame location information, and the classification loss function value of the sample category information and ...

specific Embodiment 3

[0035] On the basis of specific embodiment 1, the specific method steps of the training phase further include: S15. Re-add the error result of the division result in the training process to the training set, as a typical negative sample to replace the randomly generated negative sample, once again Perform model training. This ensures that the number of positive samples and negative samples during training will not differ too much, and at the same time further improves the accuracy of the classifier and regressor.

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 provides a common invader object detection and identification method of a power transmission corridor based on deep learning. The method comprises that in a training phase, the deep learning method is used to learn pictures, collected by a video collector, of invasion of foreign matters, and a needed network module is obtained by learning; and in the using phase, pictures obtained by practical monitoring is transmitted to the network module to detect and identity invaders. Thus, different types of invaders can be detected and identified, the accuracy and robustness are high, a relatively high processing speed is ensured, and reliable safety guarantee is provided for power transmission lines.

Description

Technical field [0001] The present invention relates to a method for detecting and identifying common intrusive objects in power transmission corridors based on deep learning, in particular to a method for detecting and identifying common intrusive objects in power transmission corridors in the field of machine vision. Background technique [0002] With the development of technology and economy, the construction of the national grid is also developing rapidly, and the number of transmission lines erected is increasing. However, the transmission lines have been exposed to the wild for a long time, and the terrain of the line corridors is complex and changeable, and it is easy to be invaded by foreign objects. Bring huge safety hazards, and in severe cases may lead to major power transmission equipment accidents, causing huge economic losses and even casualties. [0003] For the monitoring of power transmission corridors, the traditional way is to use helicopters to conduct line insp...

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 Applications(China)
IPC IPC(8): G06K9/00G06K9/62
CPCG06V20/40G06V20/52G06V2201/07G06F18/24G06F18/25G06F18/214
Inventor 李轩周剑韩明燕陈志超徐一丹
Owner CHENGDU TOPPLUSVISION TECH 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
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