Effectiveness of dorsal ratio as computed by computer assisted feature extraction system /
Notches in the dorsal fin are used by marine sity Microfilm Inc. biologists to identify individual animals. Based on the two prominent notches, a feature called dorsal ratio is defined and widely used in the identification of possible candidates. Manual extraction of this feature is subjective and r...
| Main Author: | |
|---|---|
| Format: | Thesis Book |
| Language: | English |
| Published: |
[Place of publication not identified] :
[publisher not identified] ;
1998.
|
| Subjects: | |
| Online Access: | http://proxy.library.tamu.edu/login?url=http://proquest.umi.com/pqdweb?did=733059541&sid=1&Fmt=2&clientId=2945&RQT=309&VName=PQD |
| Summary: | Notches in the dorsal fin are used by marine sity Microfilm Inc. biologists to identify individual animals. Based on the two prominent notches, a feature called dorsal ratio is defined and widely used in the identification of possible candidates. Manual extraction of this feature is subjective and requires a fair amount of experience. This dissertation presents a computer assisted system to overcome the difficulties associated with the manual approach. This system is used to study the effectiveness of dorsal ratio. The first component of the system consists of active contour modeling where the trailing edge of a fin is obtained. This is followed by a curvature module to find the characteristic tin points: tip and two most prominent notches. Considering the sensitivity of curvature to small fluctuations and noise, curvature smoothing is performed at various smoothing scales. Wavelet coefficients are utilized to select an appropriate smoothing scale. Dorsal ratio is then computed from the curvature function at the appropriate smoothing scale. By examining a total of 274 digitized images of dolphins, it is concluded that even though dorsal ratio is an effective identification feature, it is not adequate and the use of additional features is warranted. |
|---|---|
| Item Description: | Vita. "Major Subject: Electrical Engineering". |
| Physical Description: | x, 100 leaves : illustrations ; 28 cm. |
| Bibliography: | Includes bibliographical references (leaves 78-83). |