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

An electric power equipment detection algorithm based on a convolution neural network

A technology of convolutional neural network and power equipment, applied in the direction of biological neural network model, neural learning method, neural architecture, etc., can solve the problems of difficult network convergence and low detection accuracy, and achieve excellent detection performance and real-time accurate detection effect

Inactive Publication Date: 2019-03-08
SUN YAT SEN UNIV
View PDF3 Cites 22 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, if the pre-trained model is not used for fine-tuning, the network is difficult to converge, and even if it converges, the detection accuracy is not high

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
  • An electric power equipment detection algorithm based on a convolution neural network
  • An electric power equipment detection algorithm based on a convolution neural network
  • An electric power equipment detection algorithm based on a convolution neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0054] The present invention is further described below.

[0055] Implementation process and examples of the present invention are as follows:

[0056] (1) First, 5,000 images were acquired through the visible light camera and the thermal imaging camera, totaling 10,000 images. Taking visible light images as an example, divide 5000 visible light images into 4500 training sets and 500 test sets, and there is no intersection between the training set and the test set. The division of thermal imaging images is the same as that of visible light images. Then use the labelImg tool to label the image to obtain the labels about the power equipment in the image. The label of the electrical equipment contains the coordinates (x1, y1) of the upper left corner of the relevant electrical equipment in the image, the coordinates (x2, y2) of the lower right corner and the type c of the electrical equipment. In particular, there may be multiple different types of electrical equipment in an i...

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 relates to an electric power equipment detection algorithm based on a convolution neural network, comprising the following steps of (1) a labelImg tool is used for labeling the visible light image and infrared image of the electric power equipment, and obtaining the label about the electric power equipment in the image; (2) the images and device-level tags in the training set are inputted to the detection model of electric power equipment for training, and update the model parameters by using the gradient descent algorithm of driving quantity to back propagate; (3) in the testingphase, the prediction box is reduced by soft non-maximum suppression, and the prediction box whose confidence level is higher than the threshold value is outputted finally; (4) compared with the current power equipment detection methods, the algorithm can converge and obtain excellent detection performance without fine-tuning on the pre-training model, has excellent robustness for small power equipment, can effectively reduce the phenomenon of frame error and frame leakage, and has high detection accuracy and detection speed to achieve real-time processing.

Description

technical field [0001] The invention relates to the field of image target detection, namely a convolutional neural network-based power equipment detection algorithm. Background technique [0002] At present, there are a large number of power equipment in our country. They are all independent individuals and are combined into a huge power system through power lines. The health status of power equipment directly affects the normal operation of the power system and is related to our daily life. For electrical equipment, if we can quickly and accurately locate their locations and identify the corresponding categories, it will be of great help to us in analyzing their internal temperature changes and calculating their external rust and damage. It is beneficial for us to evaluate the health of the power equipment, avoid the impact of repairing the power equipment after it is damaged, and reduce the workload of construction personnel. However, the health analysis of the power sys...

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): G06K9/00G06K9/62G06N3/04G06N3/08
CPCG06N3/084G06V20/10G06N3/045G06F18/214
Inventor 陈楚城戴宪华
Owner SUN YAT SEN UNIV
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