國立虎尾科技大學 |

Shift Invariant Support Vector Machine for Image Classification in Automatic Target Recognition Systems.

紀錄類型:	書目-語言資料,手稿 : Monograph/item
正題名/作者:	Shift Invariant Support Vector Machine for Image Classification in Automatic Target Recognition Systems./
作者:	Omoregbee, Ehimwenma.
面頁冊數:	1 online resource (182 pages)
附註:	Source: Masters Abstracts International, Volume: 84-11.
Contained By:	Masters Abstracts International84-11.
標題:	Computer science. -
電子資源:	click for full text (PQDT)
ISBN:	9798379572990

Shift Invariant Support Vector Machine for Image Classification in Automatic Target Recognition Systems.
Omoregbee, Ehimwenma.

Shift Invariant Support Vector Machine for Image Classification in Automatic Target Recognition Systems. - 1 online resource (182 pages)

Source: Masters Abstracts International, Volume: 84-11.

Thesis (M.S.)--Tuskegee University, 2023.

Includes bibliographical references

Detection and classification of Synthetic Aperture Radar (SAR) targets are especially important in automatic target recognition (ATR) applications. ATR includes detecting, classifying, and identifying targets within a scene. There are several approaches to ATR, and expectedly, each class of algorithms has strengths and also, exhibits some weaknesses. In this thesis, the focus is on developing a novel algorithm that combines the concepts of distance classifier correlation filters(DCCF) and Support vector machines (SVMs) to achieve the best possible shift-invariant classification tailored to real-time scenarios. The DCCF framework will be used as the kernel of the SVM algorithm; that is, we use DCCF to develop a new kernel function to make the non-linearly separable input data separable. We will demonstrate that the proposed kernel satisfies Mercer's condition, a theoretical requirement for viable SVM kernels.Although not shift-invariant, SVM algorithms are well-known for classification and tend to generalize well for targets not contained in the training set. DCCF increases the distance of separation between classes while making each class more compact by minimizing the intra-class distance and maximizing the inter-class distance. The proposed algorithm will rely on the strengths of DCCF and SVM algorithms. Performance results are assessed by comparing the proposed algorithm to state-of-the-art shift invariant ATR algorithms such as Maximum margin correlation filters (MMCF), Unconstrained Minimum average correlation energy (UMACE), and Optimum tradeoff synthetic discriminant function (OTSDF).

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2024

Mode of access: World Wide Web

ISBN: 9798379572990Subjects--Topical Terms:

573171
Computer science.
Subjects--Index Terms:

Synthetic aperture radarIndex Terms--Genre/Form:

554714
Electronic books.

Shift Invariant Support Vector Machine for Image Classification in Automatic Target Recognition Systems.
LDR:03023ntm a22003977 4500 001 1143580
005 20240517104604.5
006 m o d
007 cr mn ---uuuuu
008 250605s2023 xx obm 000 0 eng d
020 $a 9798379572990
035 $a (MiAaPQ)AAI30423432
035 $a AAI30423432
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Omoregbee, Ehimwenma. $3 1468306
245 1 0 $a Shift Invariant Support Vector Machine for Image Classification in Automatic Target Recognition Systems.
264 0 $c 2023
300 $a 1 online resource (182 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Masters Abstracts International, Volume: 84-11.
500 $a Advisor: Ndoye, Mandoye;Khan, Jesmin.
502 $a Thesis (M.S.)--Tuskegee University, 2023.
504 $a Includes bibliographical references
520 $a Detection and classification of Synthetic Aperture Radar (SAR) targets are especially important in automatic target recognition (ATR) applications. ATR includes detecting, classifying, and identifying targets within a scene. There are several approaches to ATR, and expectedly, each class of algorithms has strengths and also, exhibits some weaknesses. In this thesis, the focus is on developing a novel algorithm that combines the concepts of distance classifier correlation filters(DCCF) and Support vector machines (SVMs) to achieve the best possible shift-invariant classification tailored to real-time scenarios. The DCCF framework will be used as the kernel of the SVM algorithm; that is, we use DCCF to develop a new kernel function to make the non-linearly separable input data separable. We will demonstrate that the proposed kernel satisfies Mercer's condition, a theoretical requirement for viable SVM kernels.Although not shift-invariant, SVM algorithms are well-known for classification and tend to generalize well for targets not contained in the training set. DCCF increases the distance of separation between classes while making each class more compact by minimizing the intra-class distance and maximizing the inter-class distance. The proposed algorithm will rely on the strengths of DCCF and SVM algorithms. Performance results are assessed by comparing the proposed algorithm to state-of-the-art shift invariant ATR algorithms such as Maximum margin correlation filters (MMCF), Unconstrained Minimum average correlation energy (UMACE), and Optimum tradeoff synthetic discriminant function (OTSDF).
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2024
538 $a Mode of access: World Wide Web
650 4 $a Computer science. $3 573171
650 4 $a Remote sensing. $3 557272
650 4 $a Electrical engineering. $3 596380
653 $a Synthetic aperture radar
653 $a SVM algorithms
653 $a Support vector machines
653 $a Automatic target recognition
653 $a Input data
655 7 $a Electronic books. $2 local $3 554714
690 $a 0544
690 $a 0984
690 $a 0799
710 2 $a Tuskegee University. $b College of Engineering. $3 1468307
710 2 $a ProQuest Information and Learning Co. $3 1178819
773 0 $t Masters Abstracts International $g 84-11.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30423432 $z click for full text (PQDT)