國立高雄大學圖資館 |

語系: 繁體中文

說明(常見問題)

圖資館首頁

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

New technologies for video surveilla...

Lin, Weiyao.

New technologies for video surveillance systems.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	New technologies for video surveillance systems.
作者:	Lin, Weiyao.
面頁冊數:	114 p.
附註:	Source: Dissertation Abstracts International, Volume: 71-04, Section: B, page: 2598.
附註:	Advisers: Ming-Ting Sun; Radha Poovendran.
Contained By:	Dissertation Abstracts International71-04B.
標題:	Engineering, Electronics and Electrical.
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3406086
ISBN:	9781109725520

New technologies for video surveillance systems.
Lin, Weiyao.

New technologies for video surveillance systems. - 114 p.

Source: Dissertation Abstracts International, Volume: 71-04, Section: B, page: 2598.

Thesis (Ph.D.)--University of Washington, 2010.

Video surveillance is of increasing importance to many applications. In this dissertation, we investigate several key problems and develop new technologies for video surveillance systems. Two basic problems are investigated: automatic recognition of human activities and optimization of video codec systems. Firstly, automatic activity recognition plays a key role in video surveillance. We propose a new human activity recognition algorithm that can recognize basic human activities with good accuracy. The problems of recognition of activities lacking training data and flexibility for recognition algorithms are also addressed in the proposed algorithm. Furthermore, we propose a group activity recognition algorithm for recognizing activities and interactions of groups of people. The proposed algorithm can deal with the problems of recognition with varying numbers of group members as well as recognition of hierarchical interactions among people. Secondly, many video surveillance systems require the video codec to deliver the video in real time with low power in order for battery-based sensors to work as long as possible. This makes the video codec optimization very important. We propose a fast sub-pixel Motion Estimation algorithm which can greatly reduce the complexity for one of the most time-consuming parts in the video encoder. Furthermore, we propose a new Computation Scalable Motion Estimation algorithm and a Computation Control strategy which can perform Motion Estimation adaptively and allocate the computation to different video parts under different computation or power budgets to achieve high coding performance and reduced complexity. Experimental results show the effectiveness of our proposed algorithms.

ISBN: 9781109725520Subjects--Topical Terms:

226981
Engineering, Electronics and Electrical.

New technologies for video surveillance systems.
LDR:02638nmm 2200277 4500 001 280804
005 20110119094958.5
008 110301s2010 ||||||||||||||||| ||eng d
020 $a 9781109725520
035 $a (UMI)AAI3406086
035 $a AAI3406086
040 $a UMI $c UMI
100 1 $a Lin, Weiyao. $3 492923
245 1 0 $a New technologies for video surveillance systems.
300 $a 114 p.
500 $a Source: Dissertation Abstracts International, Volume: 71-04, Section: B, page: 2598.
500 $a Advisers: Ming-Ting Sun; Radha Poovendran.
502 $a Thesis (Ph.D.)--University of Washington, 2010.
520 $a Video surveillance is of increasing importance to many applications. In this dissertation, we investigate several key problems and develop new technologies for video surveillance systems. Two basic problems are investigated: automatic recognition of human activities and optimization of video codec systems. Firstly, automatic activity recognition plays a key role in video surveillance. We propose a new human activity recognition algorithm that can recognize basic human activities with good accuracy. The problems of recognition of activities lacking training data and flexibility for recognition algorithms are also addressed in the proposed algorithm. Furthermore, we propose a group activity recognition algorithm for recognizing activities and interactions of groups of people. The proposed algorithm can deal with the problems of recognition with varying numbers of group members as well as recognition of hierarchical interactions among people. Secondly, many video surveillance systems require the video codec to deliver the video in real time with low power in order for battery-based sensors to work as long as possible. This makes the video codec optimization very important. We propose a fast sub-pixel Motion Estimation algorithm which can greatly reduce the complexity for one of the most time-consuming parts in the video encoder. Furthermore, we propose a new Computation Scalable Motion Estimation algorithm and a Computation Control strategy which can perform Motion Estimation adaptively and allocate the computation to different video parts under different computation or power budgets to achieve high coding performance and reduced complexity. Experimental results show the effectiveness of our proposed algorithms.
590 $a School code: 0250.
650 4 $a Engineering, Electronics and Electrical. $3 226981
690 $a 0544
710 2 $a University of Washington. $3 492771
773 0 $t Dissertation Abstracts International $g 71-04B.
790 1 0 $a Sun, Ming-Ting, $e advisor
790 1 0 $a Poovendran, Radha, $e advisor
790 $a 0250
791 $a Ph.D.
792 $a 2010
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3406086