
SIFT & SURF on GPUs for Feature Extraction
Analysis on WANG Dataset for CBIR
LAP Lambert Academic Publishing
Published on 17. February 2017
Book
Paperback/Softback
72 pages
978-3-330-03474-7 (ISBN)
Description
CBIR is required since large image databases are used both in scientific and medical domains and in the vast advertising/marketing domain.Since, images are a complex data to handle as they are composed of matrices and vectors of data and also due to multithread execution of algorithms, programmability and low cost, image processing becomes an appropriate field of achieving parallelism. GPU is a powerful graphics engine and a highly parallel programmable processor having better efficiency and high speed that overshadows CPU. A Content Based Image Retrieval system is implemented and tested on benchmark WANG database using suitable and efficient techniques so as to achieve efficient results. The feature extraction phase is parallelized on GPU processor to obtain the speed up. Sequential and parallel implementation of CBIR is done using OpenCV. SURF and SIFT algorithms when implemented on GPU show better performance for feature extraction and for retrieval of images. The execution time is noted and on comparison with its corresponding serial counterpart a considerable speed up has been achieved.
More details
Language
English
Dimensions
Height: 220 mm
Width: 150 mm
Thickness: 5 mm
Weight
125 gr
ISBN-13
978-3-330-03474-7 (9783330034747)
Schweitzer Classification
Persons
Authors are working as Assistant Professor at Department of CSE, Shaheed Bhagat Singh State Technical Campus, Ferozepur, India. Both the authors have published several research papers in reputed International/National Journals as well as Conferences. Several research Thesis for PG students are also supervised by the authors.