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-0902114-095721 Statistics This thesis had been viewed 1472 times. Download 219 times. Author Jhe-Wei Gao Author's Email Address No Public. Department Computer Science and Enginerring Year 2013 Semester 2 Degree Master Type of Document Master's Thesis Language zh-TW.Big5 Chinese Page Count 42 Title Improving SIFT Matching Efficiency Using Hashing and Descriptor Binarization Techniques Keyword SIFT descriptor binarization hashing hashing descriptor binarization SIFT Abstract Scale-invariant feature transform (SIFT) is an algorithm in computer vision. Although it can achieve high accuracy in image matching, the speed of image matching is slow. The thesis presents a method that uses hashing and descriptor binarization to improve SIFT matching efficiency. Our method applies SIFT descriptor binarization to reduce the cost of image matching. It decreases the computational complexity with only a little loss of matching accuracy. Also, our method utilizes hashing to decrease the quantity of the matching pairs substantially and hence reduce the matching time. The experimental result demonstrates that, with only a small decrease in accuracy, the matching speed of our method is about 2500 times faster than that of SIFT linear matching. Moreover, our hashing method can be applied to other methods that adopt SIFT descriptor binarization. Advisor Committee Shang-Lin Hsieh - advisor
Chen-Chiung Hsieh - co-chair
Liang-Teh Lee - co-chair
Files Date of Defense 2014-07-30 Date of Submission 2014-09-02