Review of Application Research of Video Image Intelligent Recognition Technology in Power Transmission and Distribution Systems

The digital construction of the power grid has generated massive data. Video image intelligent recognition technology has attracted extensive attention for its powerful ability to extract image data value in application scenarios such as equipment and environmental video surveillance and unmanned in...

Full description

Saved in:
Bibliographic Details
Main Authors: Junhuang ZHOU, Tingcheng HUANG, Xiaoyu XIE, Wenjun FAN, Tingting YI, Yongjun ZHANG
Format: Article
Language:zho
Published: State Grid Energy Research Institute 2021-01-01
Series:Zhongguo dianli
Subjects:
Online Access:https://www.electricpower.com.cn/CN/10.11930/j.issn.1004-9649.201912202
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The digital construction of the power grid has generated massive data. Video image intelligent recognition technology has attracted extensive attention for its powerful ability to extract image data value in application scenarios such as equipment and environmental video surveillance and unmanned inspection in power transmission and transformation systems, etc. In response, this paper first introduces the basic concept and research framework of video image intelligent recognition technology, and summarizes the common image recognition technologies that are commonly used in power transmission and transformation systems. Then from the two perspectives: intelligent inspection for power grid equipment and environmental safety monitoring, and intelligent patrol for personal safety monitoring, the four application research scenarios of video image intelligent recognition technology in power transmission and transformation systems are reviewed respectively. On this basis, three challenges faced by video image intelligent recognition in the application of power transmission and transformation systems are discussed, possible solutions and research ideas are put forward, and suggestions are given.
ISSN:1004-9649