Identification of Electrical Equipment Based on Faster LSTM-CNN Network

Xiaoping Xiong, Shuang Xu, Wenliang Wu, Deran Tu, Jie Zhang, Zhi Wei

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Power equipment inspection is one of the most important tasks to guarantee safe and stable operation of power grids. Although traditional power equipment detection methods are simple, their performances are not stable under complex outdoor environments. In this paper, we integrated the LSTM structure into the Faster R-CNN network, and designed a Faster LSTM-CNN network. We collected both normal samples and special samples, and used a variety of identification neural network models to conduct various experiments. The experimental results show that, compared with other methods such as Faster R-CNN and R-FCN, the proposed Faster LSTM-CNN network has better recognition performance for both normal samples and special samples.

Original languageEnglish (US)
Title of host publication2020 IEEE International Conference on Networking, Sensing and Control, ICNSC 2020
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728168531
DOIs
StatePublished - Oct 30 2020
Event2020 IEEE International Conference on Networking, Sensing and Control, ICNSC 2020 - Nanjing, China
Duration: Oct 30 2020Nov 2 2020

Publication series

Name2020 IEEE International Conference on Networking, Sensing and Control, ICNSC 2020

Conference

Conference2020 IEEE International Conference on Networking, Sensing and Control, ICNSC 2020
CountryChina
CityNanjing
Period10/30/2011/2/20

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Computer Networks and Communications
  • Statistics, Probability and Uncertainty
  • Control and Optimization
  • Sensory Systems

Keywords

  • equipment recognition
  • Faster R-CNN
  • LSTM

Fingerprint Dive into the research topics of 'Identification of Electrical Equipment Based on Faster LSTM-CNN Network'. Together they form a unique fingerprint.

Cite this