• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to project page

2019 Fiscal Year Final Research Report

Construction of acoustic library and species classifier of Japanese bats for acoustic monitoring

Research Project

  • PDF
Project/Area Number 16K00568
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Research Field Environmental impact assessment
Research InstitutionThe University of Tokyo

Principal Investigator

Fukui Dai  東京大学, 大学院農学生命科学研究科(農学部), 助教 (60706670)

Co-Investigator(Kenkyū-buntansha) 松井 孝典  大阪大学, 工学研究科, 助教 (30423205)
Project Period (FY) 2016-04-01 – 2020-03-31
Keywordsコウモリ / エコーロケーション / 種判別 / 音声モニタリング / 音声データベース
Outline of Final Research Achievements

A bat echolocation call library was constructed by collecting 1,400 reference calls from 30 species. These call files were converted into spectrogram images. By using the Convolutional Neural Network (CNN), we aimed to develop a species classifier that is highly accurate and robust against call variability and noise. The mean correct answer rate was 98.1% in the 10-fold cross validation. As results of test acoustic monitoring conducted in Hokkaido, 169,240 bat calls were detected in June and 296,730 bats in July. The sample size is huge, applying the classifier developed above to these is still in progress.

Free Research Field

哺乳類生態学

Academic Significance and Societal Importance of the Research Achievements

本研究の成果から、これまで非効率な捕獲調査に頼らざるを得なかったり、音声モニタリングを行っても種識別が不可能であった地域でも、コウモリ類の種ごとの音声モニタリングが可能になる。これにより、コウモリ類の利用環境調査の効率が飛躍的に向上すると同時に、その情報量も格段と増える。例えば、風力発電施設建設に伴う環境影響評価など、調査の効率化と高精度化が求められているような領域にブレークスルーをもたらし、ひいては野生動物保全と人間活動の共生という、社会的課題の解決に大きな貢献をすることが期待できる。

URL: 

Published: 2021-02-19  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi