Please use this identifier to cite or link to this item:
Title: Automatic image matching and applications
Authors: Singhal, Aseem.
Keywords: DRNTU::Engineering::Computer science and engineering::Computing methodologies::Computer graphics
DRNTU::Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision
DRNTU::Engineering::Computer science and engineering::Computing methodologies::Pattern recognition
Issue Date: 2009
Abstract: While image matching, that is, finding correspondence between two images is a trivial manual job, it scales new heights when done automatically by a system. This is largely due to different kind of changes that might have taken place between the two images. In this paper, we present an automatic algorithm for efficiently matching two images. We primarily focus on the case in which the source image and the target image are allowed to translate with respect to one another and then briefly consider extensions to handle the more general case of rigid motion including scaling, intensification and rotation. The project basically revolves around Shi and Tomasi feature point detection, Kanade-Lucas- Tomasi feature tracking and a new algorithm called “Multiple Image Triangulation” developed and implemented by the author. These methods are quite tolerant of small position errors. Moreover, the author shows that the method extends naturally to morph one image to another. The proposed algorithm was successfully tested on different image sets. Although, the functional requirements of the project were met, there are still a few improvements and research that may be extended to the algorithm including edge correspondence instead of just feature point correspondence.
Rights: Nanyang Technological University
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:SCSE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
  Restricted Access
4.6 MBAdobe PDFView/Open

Page view(s) 50

Updated on Nov 30, 2020

Download(s) 50

Updated on Nov 30, 2020

Google ScholarTM


Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.