access icon free Reconfigurable hardware architecture for faster descriptor extraction in SURF

Speeded up robust features (SURFs) are considered to be the most efficient feature extraction algorithm and it has been implemented in powerful hardware for real-time operation due to its characteristics of data-intensive computation of high complexity. Especially, the computational load of the descriptor extraction procedure is very significant and the overall performance of SURF can be improved by speeding up the descriptor extraction step with increasing parallel hardware accelerators. However, simply increasing the hardware accelerators is burdensome because of causing significant area and power consumption. Therefore, a reconfigurable hardware architecture is proposed that enables achieving the maximum performance of the descriptor extraction step with making the best use of the existing accelerators without any additional ones. Experimental results show that the proposed architecture improves the performance of the descriptor extraction step by 24.77–47.45% with negligible area overheads when compared with the existing hardware implementations of the SURF algorithm.

Inspec keywords: reconfigurable architectures; feature extraction; real-time systems

Other keywords: descriptor extraction procedure; real-time operation; parallel hardware accelerators; reconfigurable hardware architecture; computational load; feature extraction algorithm; SURF algorithm; speeded up robust features; data-intensive computation

Subjects: Computer architecture

References

    1. 1)
      • 4. Intel: ‘Get free download & trial’. Available at http://software.intel.com/intel-vtune-amplifier-xe, accessed May 2016.
    2. 2)
    3. 3)
      • 6. Embedded Systems Lab., Sookmyung Women's Univ.: ‘Benchmarks’. Available at http://esl.sookmyung.ac.kr/surf_benchmarks.html, accessed April 2017.
    4. 4)
    5. 5)
      • 2. Bouris, D., Nikitakis, A., Papaefstathiou, I.: ‘Fast and efficient FPGA-based feature detection employing the SURF algorithm’. Int. Symp. Field-Programmable Custom Computing Machines, Charlotte, NC, USA, May 2010, pp. 310, doi:10.1109/FCCM.2010.11.
    6. 6)
      • 5. OpenCV: ‘2.4.9’. Available at http://opencv.org/releases.html, accessed May 2016.
http://iet.metastore.ingenta.com/content/journals/10.1049/el.2017.3133
Loading

Related content

content/journals/10.1049/el.2017.3133
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading