Announcement for Downloading full text filePlease respect the Copyright Act.
All digital full text dissertation and theses from this website are authorized the copyright owners. These copyrighted full-text dissertation and theses can be only used for academic, research and non-commercial purposes. Users of this website can search, read, and print for personal usage. In respect of the Copyright Act of the Republic of China, please do not reproduce, distribute, change, or edit the content of these dissertations and theses without any permission. Please do not create any work based upon a pre-existing work by reproduction, Adaptation, Distribution or other means.
URN etd-0828107-160413 Statistics This thesis had been viewed 3740 times. Download 1418 times. Author Yu-yi Chen Author's Email Address No Public. Department Computer Science and Enginerring Year 2006 Semester 2 Degree Master Type of Document Master's Thesis Language English Page Count 69 Title Integrating SAD and NCC Methods to Design a Super-Resolution Model for Multi-Object Movement Keyword Super-resolution Motion estimation Interpolation Image enlargement Bock matching Bock matching Image enlargement Interpolation Motion estimation Super-resolution Abstract Due to hardware limitation, the resolution of the digital camera cannot be greatly increased. Image enlargement is mostly carried out by software and it becomes an important research topic in image processing. Traditionally, image enlargement is magnified by a single image. Because the information is only from one image, the quality of the image is greatly constrained.
Super-resolution enlargement algorithm is proposed that uses multiple frames as additional information to estimate the high-resolution image. If we have enough observed sequential low-resolution images with subpixel shifts, the reconstruction of high-resolution image can be computed. Two major problems in super-resolution enlargement are the motion estimation and the super-resolution algorithm. The super-resolution algorithm takes the observed low-resolution images with the estimated motion as the inputs and produces a high-resolution image. The method utilizes both the spatial and temporal information of an image sequence to create a higher resolution image.
In this thesis, a complete system based on super-resolution enlargement algorithm is proposed for image sequence enlargement of the multi-object movement. We use multiple blocks matching to solve image sequence enlargement of the multi-object movement. Then, we take both the sum of absolute difference (SAD) and the normalized cross-correlation (NCC) as the criteria in the block matching. Both methods can be retrieved by each other and the accuracy of the block matching can be enhanced. And we do nothing for the non-moving regions in the image sequence. Therefore, the executing time is reduced.
In this study, the proposed method achieves better magnification quality than the traditional one. Besides, both the visual and quantitative improvements especially in the high frequency are significant.
Advisor Committee Yo-Ping Huang - advisor
Chen-Chiung Hsieh - co-chair
Chiou-Shann Fuh - co-chair
Files Date of Defense 2007-06-18 Date of Submission 2007-08-28