We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Sliced integral histogram: an efficient histogram computing algorithm and its FPGA implementation.
- Authors
Yang, Yang; Liu, Yun-Xia; Dong, Qi-Fan
- Abstract
Integral histogram provides efficient histogram computation for all possible target regions, and is widely applied in many computer vision tasks. In this paper, to address the intensive computation and frequent memory accessing bottleneck in real-time applications, a sliced integral histogram algorithm is proposed for efficient integral histogram computation. We explore how maximum parallel computation and storage reduction are simultaneously achieved. Hardware implementation architecture on Field-programmable gate array (FPGA) platform is presented. We also suggest criterion for the optimal number of slices, which allows the most appropriate architecture to be selected. Comparing with the state-of-the-art methods, experimental results on Cyclone platform demonstrate the validity of the proposed algorithm in terms of computation speed, storage capacity and power consumption. Meanwhile, the proposed algorithm can be extended to other histogram based feature descriptors and implemented on any parallel processing platforms.
- Subjects
HISTOGRAMS; ALGORITHMS; FIELD programmable gate arrays; ENERGY consumption; COMPUTER storage capacity
- Publication
Multimedia Tools & Applications, 2017, Vol 76, Issue 12, p14327
- ISSN
1380-7501
- Publication type
Article
- DOI
10.1007/s11042-016-3816-1