MARC 主機 00000nam  2200373   4500 
001    AAI3306502 
005    20081202141740.5 
008    081202s2008    ||||||||||||||||| ||eng d 
020    9780549537762 
035    (UMI)AAI3306502 
040    UMI|cUMI 
100 1  Adi, C. Kuntoro 
245 10 Hidden Markov Model based animal acoustic censusing: 
       Learning from speech processing technology 
300    180 p 
500    Source: Dissertation Abstracts International, Volume: 69-
       03, Section: B, page: 1722 
500    Adviser: Michael T. Johnson 
502    Thesis (Ph.D.)--Marquette University, 2008 
520    Individually distinct acoustic features have been observed
       in a wide range of vocally active animal species and have 
       been used to study animals for decades. Only a few studies,
       however, have attempted to examine the use of acoustic 
       identification of individuals to assess population, either
       for evaluating the population structure, population 
       abundance and density, or for assessing animal seasonal 
       distribution and trends 
520    This dissertation presents an improved method to 
       acoustically assess animal population. The integrated 
       framework combines the advantages of supervised 
       classification (repertoire recognition and individual 
       animal identification), unsupervised classification 
       (repertoire clustering and individual clustering) and the 
       mark-recapture approach of abundance estimation, either 
       for population structure assessment or population 
       abundance estimate. The underlying algorithm is based on 
       clustering of Hidden Markov Models (HMMs), commonly used 
       in the signal processing and automatic speech recognition 
       community for speaker identification, also referred to as 
520    A comparative study of wild and captive beluga, 
       Delphinapterus leucas, repertoires shows the reliability 
       of the approach to assess the acoustic characteristics 
       (similarity, dissimilarity) of the established social 
       groups. The results demonstrate the feasibility of the 
       method to assess, to track, and to monitor the beluga 
       whale population for potential conservation use 
520    For the censusing task, the method is able to estimate 
       animal population using three possible scenarios. Scenario
       1, assuming availability of training data from a specific 
       species with call-type labels and speaker labels, the 
       method estimates total population. Scenario 2, with 
       availability of training data with only call-type labels 
       but no individual identities, the proposed method is able 
       to perform local population estimation. Scenario 3 with 
       availability of a few call-type examples, but no full 
       training set on individual identities, the method is able 
       to perform local population estimation 
520    The experiments performed over the Norwegian ortolan 
       bunting,  Emberiza hortulana, data set show the 
       feasibility and effectiveness of the method in estimating 
       ortolan bunting population abundance 
590    School code: 0116 
590    DDC 
650  4 Engineering, Electronics and Electrical 
650  4 Computer Science 
690    0544 
690    0984 
710 2  Marquette University 
773 0  |tDissertation Abstracts International|g69-03B 
856 40 |u