東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Automatic target detection in hypers...

Islam, Muhammad Faysal.

FindBook

Google Book

Amazon

博客來

Automatic target detection in hyperspectral imagery using one-dimensional MACH and EMACH filters.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Automatic target detection in hyperspectral imagery using one-dimensional MACH and EMACH filters./
作者:	Islam, Muhammad Faysal.
面頁冊數:	70 p.
附註:	Source: Masters Abstracts International, Volume: 45-02, page: 0992.
Contained By:	Masters Abstracts International45-02.
標題:	Engineering, Electronics and Electrical. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=1439107
ISBN:	9780542953088

Automatic target detection in hyperspectral imagery using one-dimensional MACH and EMACH filters.
Islam, Muhammad Faysal.

Automatic target detection in hyperspectral imagery using one-dimensional MACH and EMACH filters. - 70 p.

Source: Masters Abstracts International, Volume: 45-02, page: 0992.

Thesis (M.S.E.E.)--University of South Alabama, 2007.

Accurate detection of targets in hyperspectral imagery is a challenging task because the targets usually occupy only a few pixels or even sub-pixels. Recognition and classification of objects in hyperspectral imagery are performed based on their spectral signatures rather than color or shape. The presence of absorption, sensor artifacts and background noise in hyperspectral data can make the detection process difficult as the spectral signatures may vary significantly. To alleviate the aforementioned limitations, in this thesis, two two-step correlation training filter-based algorithms are proposed for target detection. In the first step, a filter is trained using various spectral signatures of a desired object, and in the second step, the filter is applied to a hyperspectral data cube using a modified classifier to detect the desired objects.

ISBN: 9780542953088Subjects--Topical Terms:

626636
Engineering, Electronics and Electrical.

Automatic target detection in hyperspectral imagery using one-dimensional MACH and EMACH filters.
LDR:02233nmm 2200277 4500 001 1834631
005 20071127114958.5
008 130610s2007 eng d
020 $a 9780542953088
035 $a (UMI)AAI1439107
035 $a AAI1439107
040 $a UMI $c UMI
100 1 $a Islam, Muhammad Faysal. $3 1923271
245 1 0 $a Automatic target detection in hyperspectral imagery using one-dimensional MACH and EMACH filters.
300 $a 70 p.
500 $a Source: Masters Abstracts International, Volume: 45-02, page: 0992.
500 $a Adviser: Mohammad S. Alam.
502 $a Thesis (M.S.E.E.)--University of South Alabama, 2007.
520 $a Accurate detection of targets in hyperspectral imagery is a challenging task because the targets usually occupy only a few pixels or even sub-pixels. Recognition and classification of objects in hyperspectral imagery are performed based on their spectral signatures rather than color or shape. The presence of absorption, sensor artifacts and background noise in hyperspectral data can make the detection process difficult as the spectral signatures may vary significantly. To alleviate the aforementioned limitations, in this thesis, two two-step correlation training filter-based algorithms are proposed for target detection. In the first step, a filter is trained using various spectral signatures of a desired object, and in the second step, the filter is applied to a hyperspectral data cube using a modified classifier to detect the desired objects.
520 $a Maximum average correlation height (MACH) and extended MACH (EMACH) filters are widely used to detect targets for two-dimensional pattern recognition applications. In this thesis, one-dimensional MACH and EMACH filters are developed using spectral information for training purposes. Then, these filters are applied to detect the desired target(s). Detailed simulation programs using the MATLAB software package are developed to investigate the performance of the proposed algorithms.
590 $a School code: 0491.
650 4 $a Engineering, Electronics and Electrical. $3 626636
690 $a 0544
710 2 0 $a University of South Alabama. $3 1017878
773 0 $t Masters Abstracts International $g 45-02.
790 1 0 $a Alam, Mohammad S., $e advisor
790 $a 0491
791 $a M.S.E.E.
792 $a 2007
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=1439107