東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Image motion estimation: New sensor ...

Chen, Wei-ge.

Linked to FindBook

Google Book

Amazon

博客來

Image motion estimation: New sensor and algorithms.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Image motion estimation: New sensor and algorithms./
Author:	Chen, Wei-ge.
Description:	160 p.
Notes:	Source: Dissertation Abstracts International, Volume: 56-09, Section: B, page: 5054.
Contained By:	Dissertation Abstracts International56-09B.
Subject:	Engineering, Electronics and Electrical. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=9600466

Image motion estimation: New sensor and algorithms.
Chen, Wei-ge.

Image motion estimation: New sensor and algorithms. - 160 p.

Source: Dissertation Abstracts International, Volume: 56-09, Section: B, page: 5054.

Thesis (Ph.D.)--University of Virginia, 1995.

Information about image motion is important to many tasks in image understanding and video coding. Over the years, many techniques have been proposed to estimate image motion, but a satisfactory solution is yet to be found. This thesis presents several approaches addressing the limitations that the existing techniques are subject to.Subjects--Topical Terms:

626636
Engineering, Electronics and Electrical.

Image motion estimation: New sensor and algorithms.
LDR:02321nmm 2200289 4500 001 1840154
005 20050714101548.5
008 130614s1995 eng d
035 $a (UnM)AAI9600466
035 $a AAI9600466
040 $a UnM $c UnM
100 1 $a Chen, Wei-ge. $3 1928502
245 1 0 $a Image motion estimation: New sensor and algorithms.
300 $a 160 p.
500 $a Source: Dissertation Abstracts International, Volume: 56-09, Section: B, page: 5054.
502 $a Thesis (Ph.D.)--University of Virginia, 1995.
520 $a Information about image motion is important to many tasks in image understanding and video coding. Over the years, many techniques have been proposed to estimate image motion, but a satisfactory solution is yet to be found. This thesis presents several approaches addressing the limitations that the existing techniques are subject to.
520 $a In the second chapter, a novel visual cue for motion perception, namely motion smear, is exploited and "motion-from-smear" algorithms are developed. A sensor system is designed and implemented to accommodate the "motion-from-smear" algorithms. Experimental results using the sensor are presented.
520 $a In the third and fourth chapters, the focus is on another facet of motion estimation, namely the spatio-temporal approach, which traditionally assumes time-invariance and space-invariance of image motion within a spatio-temporal window. Two generalizations on the assumptions are considered: the polynomial time-variation and the spatial discontinuity of motion. Under certain conditions, it is shown that time-varying and discontinuous motion estimation can be formulated as a harmonic retrieval problem. Experimental verification of these formulations is presented.
520 $a The last topic of this dissertation concerns image sequences acquired by moving cameras, which contain image motion due to camera as well as object (independent) motion. A new algorithm for detecting independent motion is developed, whose main advantages are its simplicity and computational efficiency.
590 $a School code: 0246.
650 4 $a Engineering, Electronics and Electrical. $3 626636
650 4 $a Computer Science. $3 626642
650 4 $a Psychology, General. $3 1018034
690 $a 0544
690 $a 0984
690 $a 0621
710 2 0 $a University of Virginia. $3 645578
773 0 $t Dissertation Abstracts International $g 56-09B.
790 $a 0246
791 $a Ph.D.
792 $a 1995
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=9600466