語系:
繁體中文
English
說明(常見問題)
回圖書館首頁
手機版館藏查詢
登入
回首頁
到查詢結果
[ subject:"Computer engineering." ]
切換:
標籤
|
MARC模式
|
ISBD
Efficient Image Registration and Fus...
~
Ghantous, Milad M.
FindBook
Google Book
Amazon
博客來
Efficient Image Registration and Fusion Algorithms for Mobile Emergency Surveillance.
紀錄類型:
書目-電子資源 : Monograph/item
正題名/作者:
Efficient Image Registration and Fusion Algorithms for Mobile Emergency Surveillance./
作者:
Ghantous, Milad M.
面頁冊數:
111 p.
附註:
Source: Dissertation Abstracts International, Volume: 73-05, Section: B, page: 3121.
Contained By:
Dissertation Abstracts International73-05B.
標題:
Computer engineering. -
電子資源:
http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3496497
ISBN:
9781267161024
Efficient Image Registration and Fusion Algorithms for Mobile Emergency Surveillance.
Ghantous, Milad M.
Efficient Image Registration and Fusion Algorithms for Mobile Emergency Surveillance.
- 111 p.
Source: Dissertation Abstracts International, Volume: 73-05, Section: B, page: 3121.
Thesis (Ph.D.)--University of Louisiana at Lafayette, 2011.
Surveillance systems cover almost every aspect of our daily life. They are found in airports, schools, public transportation systems, industries, military bases, etc. However, security levels can be raised immediately due to specific events such as terrorism threats, attacks, or natural disasters. In that case, existing surveillance systems fall short of the task. Deploying new ones might be time and money consuming, if not impossible. Therefore, there is a growing need for mobile emergency surveillance units. The small, battery operated units with multimodal cameras on board, are easy to deploy and perform all the required processing on the sensor side rather than a central hub, thus eliminating the single point of failure, improving the system robustness, and decreasing communication costs by sending the results over the network rather than raw images or videos. However, image processing algorithms need to be tailored for mobile units due to the resource constrained nature of the latter. In this dissertation, we tackle the aforementioned challenges with an integrated Image Registration and Fusion system comprising three main components: an Image Registration algorithm, an Image Fusion algorithm, and a hardware implementation of a 2-D Discrete Wavelet Transform architecture, that are all low power and low memory designs.
ISBN: 9781267161024Subjects--Topical Terms:
621879
Computer engineering.
Efficient Image Registration and Fusion Algorithms for Mobile Emergency Surveillance.
LDR
:03322nmm a2200301 4500
001
2073114
005
20160914074005.5
008
170521s2011 ||||||||||||||||| ||eng d
020
$a
9781267161024
035
$a
(MiAaPQ)AAI3496497
035
$a
AAI3496497
040
$a
MiAaPQ
$c
MiAaPQ
100
1
$a
Ghantous, Milad M.
$3
3188341
245
1 0
$a
Efficient Image Registration and Fusion Algorithms for Mobile Emergency Surveillance.
300
$a
111 p.
500
$a
Source: Dissertation Abstracts International, Volume: 73-05, Section: B, page: 3121.
500
$a
Adviser: Magdy Bayoumi.
502
$a
Thesis (Ph.D.)--University of Louisiana at Lafayette, 2011.
520
$a
Surveillance systems cover almost every aspect of our daily life. They are found in airports, schools, public transportation systems, industries, military bases, etc. However, security levels can be raised immediately due to specific events such as terrorism threats, attacks, or natural disasters. In that case, existing surveillance systems fall short of the task. Deploying new ones might be time and money consuming, if not impossible. Therefore, there is a growing need for mobile emergency surveillance units. The small, battery operated units with multimodal cameras on board, are easy to deploy and perform all the required processing on the sensor side rather than a central hub, thus eliminating the single point of failure, improving the system robustness, and decreasing communication costs by sending the results over the network rather than raw images or videos. However, image processing algorithms need to be tailored for mobile units due to the resource constrained nature of the latter. In this dissertation, we tackle the aforementioned challenges with an integrated Image Registration and Fusion system comprising three main components: an Image Registration algorithm, an Image Fusion algorithm, and a hardware implementation of a 2-D Discrete Wavelet Transform architecture, that are all low power and low memory designs.
520
$a
The first subpart is an automatic Image Registration algorithm in which source and reference images are decomposed into several resolution levels. Registration starts at the coarsest resolution generating a rough parameter estimate that is refined later throughout the pyramid. The pyramidal scheme offers a major advantage over its counterparts due to the reduction in image sizes.
520
$a
The second subpart is a hybrid Image Fusion scheme that replaces the heavy segmentation process with a simple object extraction technique. Extracted objects undergo a region-based fusion while the background follows a pixel-based approach. The proposed algorithm not only exhibits higher performance in fusion quality, but also offers reduction in computations in comparison with state-of-the-art schemes.
520
$a
Finally, an area-efficient, parallel and pipelined architecture for 2-D DWT is proposed. A modified image scan coupled with "multiplier-free" multiplications allow for an efficient and fast implementation. The architecture is tested and implemented in 45 nm CMOS.
590
$a
School code: 1363.
650
4
$a
Computer engineering.
$3
621879
690
$a
0464
710
2
$a
University of Louisiana at Lafayette.
$3
1025962
773
0
$t
Dissertation Abstracts International
$g
73-05B.
790
$a
1363
791
$a
Ph.D.
792
$a
2011
793
$a
English
856
4 0
$u
http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3496497
筆 0 讀者評論
館藏地:
全部
電子資源
出版年:
卷號:
館藏
1 筆 • 頁數 1 •
1
條碼號
典藏地名稱
館藏流通類別
資料類型
索書號
使用類型
借閱狀態
預約狀態
備註欄
附件
W9305982
電子資源
11.線上閱覽_V
電子書
EB
一般使用(Normal)
在架
0
1 筆 • 頁數 1 •
1
多媒體
評論
新增評論
分享你的心得
Export
取書館
處理中
...
變更密碼
登入