基于图像识别的电力设备运行状态自动化监测
Automatic Monitoring of Power Equipment Operation Status Based on Image Recognition
刘智兴 1代强强1
作者信息
- 1. 国网安徽省电力有限公司蒙城县供电公司,安徽 亳州 233500
- 折叠
摘要
由于电力设备种类繁多、结构复杂且运行环境各异,设备运行状态的实时监测率较低,为此,提出基于图像识别的电力设备运行状态自动化监测研究.采用高性能摄像头实时捕获电力设备运行图像,通过高斯滤波和灰度转换优化图像质量,利用直方图均衡化技术改善电力设备运行图像质量,并通过尺度不变特征变换算法提取图像中的关键点特征,使用模板匹配算法,对比实时捕获的电力设备图像特征与标准数据库中的图像特征,实现电力设备的自动监测和精准识别异常.结果表明,设计方法能够快速准确地捕捉到电力设备异常情况,显著提高实时监测率.
Abstract
Due to the wide variety,complex structure,and diverse operating environments of power equipment,the real-time monitoring rate of equipment operation status is relatively low.Therefore,a research on automatic monitoring of power equipment operation status based on image recognition is proposed.Using high-performance cameras to capture real-time images of power equipment operation,optimizing image quality through Gaussian filtering and grayscale conversion,improving image quality through histogram equalization technology,and extracting key point features from images through scale invariant feature transformation algorithm.Using template matching algorithm,comparing real-time captured image features of power equipment with image features in standard databases,achieving automatic monitoring and accurate identification of anomalies in power equipment.The results show that the design method can quickly and accurately capture abnormal situations of power equipment,significantly improving the real-time monitoring rate.
关键词
图像识别/电力设备/运行状态/特征提取/自动化监测Key words
image recognition/power equipment/operating status/feature extraction/automated monitoring引用本文复制引用
出版年
2024
NETL
NSTL
万方数据