Moving object extraction by watershed algorithm considering energy minimization
Search this article
Abstract
金沢大学大学院自然科学研究科情報システム
MPEG-4, which is a video coding standard, supports object-based functionalities for high efficiency coding. MPEG-7, a multimedia content description interface, handles the object data in, for example, retrieval and/or editing systems. Therefore, extraction of semantic video objects is an indispensable tool that benefits these newly developed schemes. In the present paper, we propose a technique that extracts the shape of moving objects by combining snakes and watershed algorithm. The proposed method comprises two steps. In the first step, snakes extract contours of moving objects as a result of the minimization of an energy function. In the second step, the conditional watershed algorithm extracts contours from a topographical surface including a new function term. This function term is introduced to improve the estimated contours considering boundaries of moving objects obtained by snakes. The efficiency of the proposed approach in moving object extraction is demonstrated through computer simulations. © Springer-Verlag Berlin Heidelberg 2007.
Journal
-
- Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
-
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) LNIP 4678 711-719, 2007-01-01
Springer Verlag (Germany)
- Tweet
Details 詳細情報について
-
- CRID
- 1050282810913207680
-
- NII Article ID
- 120000811799
-
- NII Book ID
- AA0071599X
-
- ISSN
- 03029743
-
- Web Site
- http://hdl.handle.net/2297/9045
-
- Text Lang
- en
-
- Article Type
- journal article
-
- Data Source
-
- IRDB
- CiNii Articles