Journal of IKEEE (전기전자학회논문지)

Institute of Korean Electrical and Electronics Engineers (한국전기전자학회)
권호 표지
DOI QR코드

DOI QR Code

Design and Implementation of Feature Detector for Object Tracking

객체 추적을 위한 특징점 검출기의 설계 및 구현

  • Lee, Du-hyeon (School of Electronics and Information Engineering, Korea Aerospace University) ;
  • Kim, Hyeon (School of Electronics and Information Engineering, Korea Aerospace University) ;
  • Cho, Jae-chan (School of Electronics and Information Engineering, Korea Aerospace University) ;
  • Jung, Yun-ho (School of Electronics and Information Engineering, Korea Aerospace University)
  • Received : 2019.03.10
  • Accepted : 2019.03.20
  • Published : 2019.03.31
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, we propose a low-complexity feature detection algorithm for object tracking and present hardware architecture design and implementation results for real-time processing. The existing Shi-Tomasi algorithm shows good performance in object tracking applications, but has a high computational complexity. Therefore, we propose an efficient feature detection algorithm, which can reduce the operational complexity with the similar performance to Shi-Tomasi algorithm, and present its real-time implementation results. The proposed feature detector was implemented with 1,307 logic slices, 5 DSP 48s and 86.91Kbits memory with FPGA. In addition, it can support the real-time processing of 54fps at an operating frequency of 114MHz for $1920{\times}1080FHD$ images.

본 논문에서는 객체 추적을 위한 간소화된 특징점 검출 알고리즘을 제안하고, 이의 실시간 처리를 위한 하드웨어 구조 설계 및 구현 결과를 제시한다. 기존 Shi-Tomasi 알고리즘은 객체 추적 응용에서 우수한 성능을 보이지만, 연산 복잡도가 큰 문제가 존재한다. 따라서, 기존 알고리즘에 비해 연산 복잡도를 간소화시키면서 유사한 성능 지원이 가능한 효율적인 특징점 검출 알고리즘을 제안하고, 하드웨어 설계 및 구현 결과를 제시한다. 제안된 특징점 검출기는 FPGA 기반 구현 결과, 1,307개의 logic slices, 5개의 DSP 48s, 86.91Kbit의 메모리로 구현 가능함을 확인하였으며, 114MHz의 동작 주파수로 $1920{\times}1080FHD$급 영상에 대해 54fps의 실시간 처리가 가능하다.

Keywords

JGGJB@_2019_v23n1_207_f0001.png 이미지

Fig. 1. Example of feature candidate selection according to eigenvalue calculation. 그림 1. 고윳값 계산에 따른 특징점 후보 선정 예시

JGGJB@_2019_v23n1_207_f0002.png 이미지

Fig. 3. Example of feature tracking results. 그림 3. 특징점 추적 결과 예시

JGGJB@_2019_v23n1_207_f0003.png 이미지

Fig. 2. Feature detection examples: (a) Harris, (b) Shi-Tomasi, (c) Proposed algorithm. 그림 2. 특징점 검출 결과 예시 (a) Harris, (b) Shi-Tomasi, (c) 제안된 알고리즘

JGGJB@_2019_v23n1_207_f0004.png 이미지

Fig. 4. Block diagram for the proposed feature detector. 그림 4. 제안된 특징점 검출기의 블록도

JGGJB@_2019_v23n1_207_f0005.png 이미지

Fig. 5. Block diagram for the Gaussian filter. 그림 5. 가우시안 필터의 블록도

JGGJB@_2019_v23n1_207_f0006.png 이미지

Fig. 6. Example of line buffer data flow. 그림 6. 라인 버퍼의 데이터 플로우 예시

JGGJB@_2019_v23n1_207_f0007.png 이미지

Fig. 7. Block diagram for the region checking unit. 그림 7. 영역 체크 유닛의 블록도

Table 1. Implementation results of the proposed feature detector with Virtex-7 FPGA. 표 1. Virtex-7 FPGA 기반 특징점 검출기의 구현 결과

JGGJB@_2019_v23n1_207_t0001.png 이미지

Table 2. Processing-rate comparisons. 표 2. 처리 속도 비교

JGGJB@_2019_v23n1_207_t0002.png 이미지

References

  1. G. Velez, O. Otaegui, "Embedding vision-based advanced driver assistance systems: a survey," IET Intelligent Transport Systems, vol. 11, no. 3, pp. 103-112, 2017. DOI: 10.1049/iet-its.2016.0026
  2. A. Ferrick, J. Fish, E. Venator, G. S. Lee, "UAV Obstacle avoidance using image processing techniques," 2012 IEEE International Conference on Technologies for Practical Robot Applications, pp. 73-78, 2012. DOI: 10.1109/TePRA.2012.6215657
  3. J. Lee, "Implementation of pedestrian recognition based on HOG using ROI for real time processing," Journal of IKEEE, vol. 18, no. 4, pp. 581-585, 2014. DOI: 10.7471/ikeee.2014.18.4.581
  4. A. Smeulders, D. Chu et. al, "Visual tracking: an experimental survey," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 36, no. 7, pp. 1442-1468, 2014. https://doi.org/10.1109/TPAMI.2013.230
  5. S. Kim, H. Kim, and S. Ko, "A vehicle detection and tracking algorithm for supervision of illegal parking," Journal of IKEEE, vol. 13, no. 2, pp. 232-240, 2009. DOI: 10.1109/TPAMI.2013.230
  6. S. Smith and J. Bardy, "SUSAN-A new approach to low-level image processing," International Journal of Computer Vision, vol. 23, pp. 45-48, 1997. DOI: 10.1023/A:1007963824710
  7. D. G. Lowe, "Distinctive image features from scales-invariant key points," International Journal of Computer Vision, vol. 60, no. 2, pp. 91-110, 2004. DOI: 10.1023/B:VISI.0000029664.99615.94
  8. C. Harris and M. Stephens, "A combined corner and edge detector," Proceedings of the fourth alvey vision conference, pp. 147-151, 1988. DOI: 10.1.1.231.1604
  9. W. Jang, S. O and G. Kim, "A hardware implementation of pyramidal KLT feature tracker for driving assistance systems," IEEE Conference on Intelligent Transportation Systems, pp. 220-225, 2009. DOI: 10.1109/ITSC.2009.5309680
  10. T. Cho and K. Wong, "An efficient FPGA implementation of the Harris corner feature detector," 2015 IAPR International Conference of Machine Vision Application, 2015. DOI: 10.1109/MVA.2015.7153140
  11. J. Shi and C. Tomasi, "Good features to track," IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 1994. DOI: 10.1109/CVPR.1994.323794
  12. http://cvlab.hanyang.ac.kr/tracker_benchmark/benchmark_v10.html
  13. T. Dinh et. al, "High throughput FPGA architecture for corner detection in traffic images," 2014 IEEE Fifth ICCE, pp. 297-302, 2014. DOI: 10.1109/CCE.2014.6916718
  14. F. Brenot, P. Fillatreau and J. Piat, "FPGA based accelerator for visual features detection," 2015 IEEE International Workshop of CMSM, 2015. DOI: 10.1109/ECMSM.2015.7208697
  15. A. Aguilar-Gonzalez, M. Arias-Estrada and F. Berry, "Robust feature extraction algorithm suitable for real-time embedded applications," Journal of Real-Time Image Processing, vol. 14, no. 3, pp. 647-665, 2018. DOI: 10.1007/s11554-017-0701-8