國立虎尾科技大學 |

Graph-Based Models and Transforms for Signal/data Processing With Applications to Video Coding.

紀錄類型:	書目-語言資料,手稿 : Monograph/item
正題名/作者:	Graph-Based Models and Transforms for Signal/data Processing With Applications to Video Coding./
作者:	Egilmez, Hilmi Enes.
面頁冊數:	1 online resource (135 pages)
附註:	Source: Dissertation Abstracts International, Volume: 79-06(E), Section: B.
Contained By:	Dissertation Abstracts International79-06B(E).
標題:	Electrical engineering. -
電子資源:	click for full text (PQDT)

Graph-Based Models and Transforms for Signal/data Processing With Applications to Video Coding.
Egilmez, Hilmi Enes.

Graph-Based Models and Transforms for Signal/data Processing With Applications to Video Coding. - 1 online resource (135 pages)

Source: Dissertation Abstracts International, Volume: 79-06(E), Section: B.

Thesis (Ph.D.)--University of Southern California, 2017.

Includes bibliographical references

Graphs are fundamental mathematical structures used in various fields to represent data, signals and processes. Particularly in signal processing, machine learning and statistics, structured modeling of signals and data by means of graphs is essential for a broad number of applications. In this thesis, we develop novel techniques to build graph-based models and transforms for signal/data processing, where the models and transforms of interest are defined based on graph Laplacian matrices. For graph-based modeling, various graph Laplacian estimation problems are studied. Firstly, we consider estimation of three types of graph Laplacian matrices from data and develop efficient algorithms by incorporating associated Laplacian and structural constraints. Then, we propose a graph signal processing framework to learn graph-based models from classes of filtered signals, defined based on functions of graph Laplacians. The proposed approach can also be applied to learn diffusion (heat) kernels, which are popular in various fields for modeling diffusion processes. Additionally, we study the problem of multigraph combining, which is estimating a single optimized graph from multiple graphs, and develop an algorithm. Finally, we propose graph-based transforms for video coding and develop two different techniques, based on graph learning and image edge adaptation, to design orthogonal transforms capturing the statistical characteristics of video signals. Theoretical justifications and comprehensive experimental results for the proposed methods are presented.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2018

Mode of access: World Wide Web

Subjects--Topical Terms:

596380
Electrical engineering.
Index Terms--Genre/Form:

554714
Electronic books.

Graph-Based Models and Transforms for Signal/data Processing With Applications to Video Coding.
LDR:02781ntm a2200313Ki 4500 001 919664
005 20181129115240.5
006 m o u
007 cr mn||||a|a||
008 190606s2017 xx obm 000 0 eng d
035 $a (MiAaPQ)AAI10820433
035 $a (MiAaPQ)US_Calif_427945
035 $a AAI10820433
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Egilmez, Hilmi Enes. $3 1194287
245 1 0 $a Graph-Based Models and Transforms for Signal/data Processing With Applications to Video Coding.
264 0 $c 2017
300 $a 1 online resource (135 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: Dissertation Abstracts International, Volume: 79-06(E), Section: B.
500 $a Adviser: Antonio Ortega.
502 $a Thesis (Ph.D.)--University of Southern California, 2017.
504 $a Includes bibliographical references
520 $a Graphs are fundamental mathematical structures used in various fields to represent data, signals and processes. Particularly in signal processing, machine learning and statistics, structured modeling of signals and data by means of graphs is essential for a broad number of applications. In this thesis, we develop novel techniques to build graph-based models and transforms for signal/data processing, where the models and transforms of interest are defined based on graph Laplacian matrices. For graph-based modeling, various graph Laplacian estimation problems are studied. Firstly, we consider estimation of three types of graph Laplacian matrices from data and develop efficient algorithms by incorporating associated Laplacian and structural constraints. Then, we propose a graph signal processing framework to learn graph-based models from classes of filtered signals, defined based on functions of graph Laplacians. The proposed approach can also be applied to learn diffusion (heat) kernels, which are popular in various fields for modeling diffusion processes. Additionally, we study the problem of multigraph combining, which is estimating a single optimized graph from multiple graphs, and develop an algorithm. Finally, we propose graph-based transforms for video coding and develop two different techniques, based on graph learning and image edge adaptation, to design orthogonal transforms capturing the statistical characteristics of video signals. Theoretical justifications and comprehensive experimental results for the proposed methods are presented.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2018
538 $a Mode of access: World Wide Web
650 4 $a Electrical engineering. $3 596380
655 7 $a Electronic books. $2 local $3 554714
690 $a 0544
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a University of Southern California. $3 1182588
773 0 $t Dissertation Abstracts International $g 79-06B(E).
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10820433 $z click for full text (PQDT)