Comparison of deep learing algorithms for indoor monitoring using bioelectric potential of living plants

Hidetaka Nambo, Imam Tahyudin, Takeo Nakano, Tetsuya Yamada

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

Abstract

This study aims to develop a monitoring system for an indoor space. We are investigating to use the bioelectric potential of living plants as a human sensor system in an indoor environment. The system utilizes a change of the bioelectric potential to estimate a resident's location in a room. To build an estimation model, a lot of the bioelectric potential data are collected and processed by a machine learning method. We have studied to build the estimation model using a convolutional neural network. However, recently, there are many applications that utilize Long-Short Term Memory method for a time sequential data, and they obtained a good result successfully. Therefore, in this study we applied LSTM for the bioelectric potential data and investigate the availability of CNN and LSTM to estimate the location with the bioelectric potential. As the result of classification experiments with the model trained with collected bioelectric data, we obtained that CNN is better than LSTM for this problem. However, we need to improve the accuracy by adjusting parameters in future.

Original languageEnglish
Title of host publicationProceedings - 2018 3rd International Conference on Information Technology, Information Systems and Electrical Engineering, ICITISEE 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages110-113
Number of pages4
ISBN (Electronic)9781538670828
DOIs
Publication statusPublished - Jul 2 2018
Externally publishedYes
Event3rd International Conference on Information Technology, Information Systems and Electrical Engineering, ICITISEE 2018 - Yogyakarta, Indonesia
Duration: Nov 13 2018Nov 14 2018

Publication series

NameProceedings - 2018 3rd International Conference on Information Technology, Information Systems and Electrical Engineering, ICITISEE 2018

Conference

Conference3rd International Conference on Information Technology, Information Systems and Electrical Engineering, ICITISEE 2018
Country/TerritoryIndonesia
CityYogyakarta
Period11/13/1811/14/18

All Science Journal Classification (ASJC) codes

  • Information Systems and Management
  • Information Systems
  • Electrical and Electronic Engineering
  • Health Informatics
  • Instrumentation
  • Artificial Intelligence
  • Computer Science Applications

Fingerprint

Dive into the research topics of 'Comparison of deep learing algorithms for indoor monitoring using bioelectric potential of living plants'. Together they form a unique fingerprint.

Cite this