Open AccessResearch on Defect Detection of Electric Energy Metering Box Based on YOLOv5
Author(s) -
Yong Yu,
Yanchao Sun,
Chunxue Zhao,
Chong Qu
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/2087/1/012081
Subject(s) - metering mode , workload , computer science , energy (signal processing) , state (computer science) , set (abstract data type) , electric energy , work (physics) , power (physics) , engineering , automotive engineering , reliability engineering , real time computing , algorithm , mechanical engineering , mathematics , statistics , physics , quantum mechanics , programming language , operating system
The manual inspection for the damage state of the electric energy metering box consumes a lot of time, the workload is large, and the data storage is difficult. In order to solve these problems, this paper proposes an automatic detection method for the damage state of the electric energy metering box based on the YOLOv5 algorithm. The actual metering box pictures taken by the operation and maintenance inspectors are used as the training set, LabelImage is used to annotate the data set, and YOLOv5s model is used to train the data set. The experimental results show that the method proposed in this paper can accurately mark the position of the metering box lid and accurately predict its damage state. The average accuracy reaches 98%, which can meet the requirements for the detection accuracy of the power metering box damage state in the operation and maintenance inspection work.