자료유형 | 학위논문 |
---|---|
서명/저자사항 | Information Theoretic Classification of Marine Animal Imagery. |
개인저자 | Cao, Zheng. |
단체저자명 | University of Florida. Electrical and Computer Engineering. |
발행사항 | [S.l.]: University of Florida., 2017. |
발행사항 | Ann Arbor: ProQuest Dissertations & Theses, 2017. |
형태사항 | 108 p. |
기본자료 저록 | Dissertation Abstracts International 79-12B(E). Dissertation Abstract International |
ISBN | 9780438165571 |
학위논문주기 | Thesis (Ph.D.)--University of Florida, 2017. |
일반주기 |
Source: Dissertation Abstracts International, Volume: 79-12(E), Section: B.
|
요약 | To help analyze marine animals behavior, seasonal distribution and abundance, digital imagery can be acquired by Lidar or optical camera. The Unobtrusive Multistatic Serial Lidar Imager (UMSLI) system is designed to collect and classify Lidar im |
요약 | For the purpose of classifying optical images, convolutional neural network (CNN) features are extracted and are tested on two real-world marine animal datasets, yielding better classification results than existing approaches that use hand-desig |
요약 | For both cases of dissimilarity matrices derived from different shape analysis methods (shape context, internal distance shape context, etc.) and features (shape, color, texture, etc.), multi-view learning is critical in integrating more than on |
일반주제명 | Electrical engineering. |
언어 | 영어 |
바로가기 |
: 이 자료의 원문은 한국교육학술정보원에서 제공합니다. |