Analyzing Vocal Behavior for Sperm Whales Based on Hilbert-Huang Transformation-3

View Statistics Email Alert RSS Feed

Information

Details

Project title

Code/計畫編號

MOST104-2221-E022-014

Translated Name/計畫中文名

一種抹香鯨希爾伯特黃時頻特徵解析技術-3

Project Coordinator/計畫主持人

Chin-Feng Lin

Funding Organization/主管機關

National Science and Technology Council

Co-Investigator(s)/共同執行人

張順雄
文展權

Department/Unit

National Kaohsiung University of Science and Technology, NKUST

Website

https://www.grb.gov.tw/search/planDetail?id=11575265

Year

2015

Start date/計畫起

01-08-2015

Expected Completion/計畫迄

01-07-2016

Bugetid/研究經費

577千元

ResearchField/研究領域

電子電機工程

我們解析四個 10 毫秒抹香鯨點擊聲樣本，cI1(t)、cI2(t)、cI3(t) 和cI4(t)，定義為抹香鯨點擊聲樣本I類型；四個 5 毫秒抹香鯨點擊聲樣本，cII1(t)、cII2(t)、cII3(t) 和cII4(t)，定義為抹香鯨點擊聲樣本II類型。經驗模態拆解，將抹香鯨點擊聲(I)拆解成7個本質模態函數和一個殘餘函數；將抹香鯨點擊聲(II)拆解成6個本質模態函數和一個殘餘函數，應用本質模態函數和殘餘函數可進行高解析度抹香鯨點擊聲(I)(II)時間頻率聯合特徵分析。我們分析抹香鯨點擊聲(I)(II)本質模態函數、殘餘函數和瞬時頻率特徵。四個抹香鯨點擊聲(I)之平均本質模態函數1邊際頻率能量分佈主要都集中在 8-17 KHz 和 28-33 KHz，其平均能量參考總能量比例分別為42.23%和10.69%。四個抹香鯨點擊聲(I)之平均本質模態函數2邊際頻率能量分佈主要都集中在 4-8 KHz 和 11-15 KHz，其平均能量參考總能量比例分別為3.53%和2.59%。四個抹香鯨點擊聲(I)之平均本質模態函數3邊際頻率能量分佈主要都集中在 1-3 KHz 和 6-8 KHz，其平均能量參考總能量比例分別為1.25%和1.78%。四個抹香鯨點擊聲(II)之平均本質模態函數1邊際頻率能量分佈主要都集中在 8-17 KHz 、23-25 KHz 和 34-36 KHz，其平均能量參考總能量比例分別為51.12%、3.50% 和3.80%。四個抹香鯨點擊聲(II)之平均本質模態函數2邊際頻率能量分佈主要都集中在 2-7 KHz和 38-39 KHz，其平均能量參考總能量比例分別為10.79%和1.76%。四個抹香鯨點擊聲(II)之平均本質模態函數RF邊際頻率能量分佈主要都集中在 0-1 KHz，其平均能量參考總能量比例分別為7.82%。In this project, four click samples of a sperm whale, defined as click I, with a duration of 10 ms each were denoted as cI1(t), cI2(t), cI3(t), and cI4(t). Four click samples of a sperm whale, defined as click II, with a duration of 10 ms each were denoted as cII1(t), cII2(t), cII3(t), and cII4(t). The click-I samples were decomposed into seven intrinsic mode functions (IMFs) and one residue function (RF) using the empirical mode decomposition method. The click-II samples were decomposed into six intrinsic mode functions (IMFs) and one residue function (RF) using the empirical mode decomposition method. The time-frequency features with high-resolution analysis can be extracted by applying the several IMFs and one RF. The characteristics of click I and click II for IMFs, RF, and instantaneous frequencies (IFs) were analyzed. The average energy ratios of marginal frequency 1 (MF1) to the referred total energy of the four sperm whale click-I samples, in the 8-17 KHz, and 28-33 KHz bands, were 42.23% and 10.69%, respectively. The average energy ratios of MF2 to the referred total energy of the four sperm whale click-I samples, in the 4-8 KHz, and 11-15 KHz bands, were 3.53% and 2.59%, respectively. The average energy ratios of MF3 to the referred total energy of the four sperm whale click-I samples, in the 1-3 KHz, and 6-8 KHz bands, were 1.25% and 1.78%, respectively. The average energy ratios of MF1 to the referred total energy of the four sperm whale click-II samples, in the 8-17 KHz, 23-25 KHz, and 34-36 KHz bands, were 51.12%, 3.50%, and 3.82%, respectively. The average energy ratios of MF2 to the referred total energy of the four sperm whale click-II samples, in the 2-7 KHz, and 38-39 KHz bands, were 10.79%, and 1.76%, respectively. The average energy ratios of MF3 to the referred total energy of the four sperm whale click-II samples, in the 0-1 KHz, was 7.82%.

Keyword(s)

抹香録
點擊聲
希爾伯特黃轉換
B夺間頻率能量分布
貝氏嚷録
Sperm whale
clicks
Hilb^rt-Huang transform
time-frequency-ernrgy distribution
Bairds Beaked whales

DSpace CRIS

Analyzing Vocal Behavior for Sperm Whales Based on Hilbert-Huang Transformation-3

Details

Description