8 December 2015 An optimized structure on FPGA of key point description in SIFT algorithm
Author Affiliations +
Proceedings Volume 9875, Eighth International Conference on Machine Vision (ICMV 2015); 987519 (2015) https://doi.org/10.1117/12.2228710
Event: Eighth International Conference on Machine Vision, 2015, Barcelona, Spain
Abstract
SIFT algorithm is one of the most significant and effective algorithms to describe the features of image in the field of image matching. To implement SIFT algorithm to hardware environment is apparently considerable and difficult. In this paper, we mainly discuss the realization of Key Point Description in SIFT algorithm, along with Matching process. In Key Point Description, we have proposed a new method of generating histograms, to avoid the rotation of adjacent regions and insure the rotational invariance. In Matching, we replace conventional Euclidean distance with Hamming distance. The results of the experiments fully prove that the structure we propose is real-time, accurate, and efficient. Future work is still needed to improve its performance in harsher conditions.
© (2015) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Chenyu Xu, Chenyu Xu, Jinlong Peng, Jinlong Peng, En Zhu, En Zhu, Yuxin Zou, Yuxin Zou, } "An optimized structure on FPGA of key point description in SIFT algorithm", Proc. SPIE 9875, Eighth International Conference on Machine Vision (ICMV 2015), 987519 (8 December 2015); doi: 10.1117/12.2228710; https://doi.org/10.1117/12.2228710
PROCEEDINGS
5 PAGES


SHARE
RELATED CONTENT

RPython high-level synthesis
Proceedings of SPIE (September 27 2016)
Segmentation-free morphological character recognition
Proceedings of SPIE (March 22 1994)
Hardware-based image processing library for Virtex FPGA
Proceedings of SPIE (October 05 2000)
FPGA implementation of fuzzy morphological filters
Proceedings of SPIE (May 17 2004)
Real-time FPGA architectures for computer vision
Proceedings of SPIE (March 20 2000)

Back to Top