Development of an avian monitoring system using species identification by bird songs
| Project/Area Number |
16K16222
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Environmental and ecological symbiosis
|
|---|
| Research Institution | Kyoto University |
|---|
Principal Investigator |
Fujita Motoko 京都大学, 東南アジア地域研究研究所, 連携研究員 (50456828)
|
|---|
| Research Collaborator |
OKUNO Hiroshi
ITOYAMA Katsutoshi
SUZUKI Reiji
MARUYAMA Akio
|
|---|
| Project Period (FY) |
2016-04-01 – 2019-03-31
|
| Project Status |
Completed (Fiscal Year 2018)
|
| Budget Amount *help |
¥3,380,000 (Direct Cost: ¥2,600,000、Indirect Cost: ¥780,000)
Fiscal Year 2018: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Fiscal Year 2017: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Fiscal Year 2016: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
|
|---|
| Keywords | 長期モニタリング / 音声解析 / さえずり / 生態学 / 生態系管理 / 鳥類学 / モニタリング |
|---|
| Outline of Final Research Achievements |
Automatic species identification system from bird songs consists of (1) detection of bird songs and (2) identification of bird songs. At the stage of (1), it turned out that with the 2-channel recordings, which was the target of this study in the beginning, detection rate remained quite low during analysis. The primary reason of this is due to difficulties in differentiation of songs of more than two individuals at the same time. Therefore, I decided to increase the recording channel up to 8 channels and to introduce Open Source Software for Robot Audition (HARK). Newly recorded samples of Fukui prefecture and Indonesian forests using 8 channels were analyzed, and it was clarified that the separation of individuals singing simultaneously was possible, which would lead to better identification results.
|
| Academic Significance and Societal Importance of the Research Achievements |
鳥類のモニタリングは環境や生物多様性の変化を知るうえで欠かせない調査であり、従来は目視での調査が行われてきたが、近年はデータの保存性や再現性の観点からさえずりの録音データを用いる方法が検討されている。しかし、モニタリング調査に録音データを用いる場合は、長期録音データを効率的に解析する技術が不可欠である。本研究では、①さえずりの検出、②さえずりの識別の2つの段階を機械的に行う手法を開発することを目的としているが、そのうち①において障害であった「同時にさえずっている個体の分離」が可能となった意義は大きく、高い精度で識別するための基礎的な知見を得た。
|
Report
(4 results)
Research Products
(8 results)
