Journal of the Institute of Electronics and Information Engineers (전자공학회논문지)

The Institute of Electronics and Information Engineers (대한전자공학회)
권호 표지
DOI QR코드

DOI QR Code

Rear Vehicle Detection Method in Harsh Environment Using Improved Image Information

개선된 영상 정보를 이용한 가혹한 환경에서의 후방 차량 감지 방법

  • Received : 2016.08.22
  • Accepted : 2016.12.15
  • Published : 2017.01.25
Download PDF
⟨ Previous Next ⟩

Abstract

Most of vehicle detection studies using the existing general lens or wide-angle lens have a blind spot in the rear detection situation, the image is vulnerable to noise and a variety of external environments. In this paper, we propose a method that is detection in harsh external environment with noise, blind spots, etc. First, using a fish-eye lens will help minimize blind spots compared to the wide-angle lens. When angle of the lens is growing because nonlinear radial distortion also increase, calibration was used after initializing and optimizing the distortion constant in order to ensure accuracy. In addition, the original image was analyzed along with calibration to remove fog and calibrate brightness and thereby enable detection even when visibility is obstructed due to light and dark adaptations from foggy situations or sudden changes in illumination. Fog removal generally takes a considerably significant amount of time to calculate. Thus in order to reduce the calculation time, remove the fog used the major fog removal algorithm Dark Channel Prior. While Gamma Correction was used to calibrate brightness, a brightness and contrast evaluation was conducted on the image in order to determine the Gamma Value needed for correction. The evaluation used only a part instead of the entirety of the image in order to reduce the time allotted to calculation. When the brightness and contrast values were calculated, those values were used to decided Gamma value and to correct the entire image. The brightness correction and fog removal were processed in parallel, and the images were registered as a single image to minimize the calculation time needed for all the processes. Then the feature extraction method HOG was used to detect the vehicle in the corrected image. As a result, it took 0.064 seconds per frame to detect the vehicle using image correction as proposed herein, which showed a 7.5% improvement in detection rate compared to the existing vehicle detection method.

기존의 차량 검출 연구들의 대부분은 일반렌즈 또는 광각렌즈를 가지는 후방 카메라를 사용하기 때문에 사각지대가 넓으며, 영상에 노이즈 및 다양한 외부 환경에 취약한 부분이 있다. 본 논문에서는 사각지대를 줄이고, 노이즈 및 가혹한 외부 환경에서도 인식이 가능한 검출 방법을 제안한다. 먼저 광각렌즈보다 더 넓은 화각을 가진 어안렌즈를 이용해 사각지대를 최소화한다. 렌즈의 화각이 커진 만큼 비선형 방사왜곡도 커지게 되므로, 정확한 영상 결과를 얻기 위해서 왜곡 상수 초기화와 최적화를 실시한 후 Calibration을 이용하였다. 그리고 Calibration과 동시에 원본 영상을 분석하여 안개가 자욱한 상황과 갑작스러운 조도 변화로 인해 생기는 명순응, 암순응 현상에 의한 시야 방해 상황에서도 인식이 가능하도록 안개 제거와 밝기 보정을 이용하였다. 안개 제거는 일반적으로 계산 시간이 매우 크다. 따라서 계산 시간을 줄이기 위해 대표적인 안개 제거 알고리즘인 Dark channel prior를 기반으로 안개를 제거하였다. 밝기 보정 시에는 Gamma correction을 이용했고, 보정에 필요한 Gamma value를 결정하기 위해 영상에 대한 밝기 및 명암 평가가 수행하였다. 평가는 영상의 전체가 아닌 일부분을 이용하여 할애되는 계산시간을 줄였다. 밝기 및 명암 값이 계산되면 그 값을 이용해 Gamma value를 결정하고 전체 영상에 보정을 실시하였다. 그리고 밝기 보정과 안개 제거로 나누어 병렬 처리한 후, 영상을 하나로 정합함으로써 전 처리 과정의 연산시간을 줄였다. 이후 보정된 영상으로부터 특징추출법인 HOG를 이용하여 차량을 검출하였다. 그 결과 본 논문에서 제안하는 방법의 영상 보정을 이용한 차량 검출을 하는데 1프레임당 0.064초가 걸렸으며, 기존의 차량 검출 방법에 비해 7.5%의 향상된 검출률을 얻었다.

Keywords

References

  1. S. H. Baek, H. S. Kim and K. Suck Boo, "A Method for Rear-side Vehicle Detection and Tracking with Vision System," Journal of the Korean Society for Precision Engineering, Vol. 31, no. 3, pp. 233-241, Mar, 2014. https://doi.org/10.7736/KSPE.2014.31.3.233
  2. P. A. Shyam, H. T. Cho, H. J. Yoo, C. J. Yang and H. S. Kim, "On-Road Succeeding Vehicle Detection using Characteristic Visual Features," The transactions of The Korean Institute of Electrical Engineers, Vol. 59, no. 3, pp. 636-644, Mar, 2010.
  3. S. Kyo, T. Koga, K. Sakurai and S. Okazaki, "A Robust Vehicle Detecting and Tracking System for Wet Weather Conditions using the IMAP-VISION Image Processing Board," Proc. IEEE ITS, pp. 423-428, 1999
  4. S. J. Kim and K. S. Cho, "Design of Efficient Gradient Orientation Bin and Weight Calculation Circuit for HOG Feature Calculation," Journal of the Institute of Electronics and Information Engineers, Vol.51, no. 11, pp. 66-72, Nov, 2014. https://doi.org/10.5573/ieie.2014.51.11.066
  5. N. Dalal and B. Triggs, "Histogram of oriented gradients for human detection," CVPR2005, Vol.1, pp. 886-893, 2005.
  6. S. Denasi and G. Quaglia, "Obstacle Detection using a Deformable Model of Vehicle," Proc. of IEEE IV, pp. 145-150, 2001.
  7. J. H. Rhue and Y. M. Kim, "Using play-back image sequence to detect a vehicle cutting in a line automatically," Journal of the Institute of Electronics and Information Engineers, Vol. 51, no. 2, pp. 95-101, Feb, 2014. https://doi.org/10.5573/ieie.2014.51.2.095
  8. Z. Hu, K. Uchimura, "Tracking Cycle: A New Concept for Simultaneously Tracking of Multiple Moving Objects in a Typical Traffic Scene," Proc. of IEEE 2000 Intelligent Vehicles Symposium, pp. 233-239, 2000.
  9. W. Kruger, W. Enkelmann, and S. Rossle, "Real-time Estimation and Tracking of Optical Flow Vectors for Obstacle Detection," Proc. of Intelligent Vehicles 95' Symposium, pp. 304-309, 1995.
  10. S. M. Smith and J. M. Brady, "ASSET-2 : Real-Time Motion Segmentation and Shape Tracking," IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 17, no. 8, pp. 814-820, 1995. https://doi.org/10.1109/34.400573
  11. K. Inagaki, S. Sato and T. Umezaki, "A Recurrent Neural Network Approach to Rear Vehicle Detection Which Considered State Dependency," Journal of Systemics, Cybernetics and Informatics, Vol. 1, no. 4, pp. 72-77, 2003.
  12. S. H. Kim, Y. J. Cho, J. W. Son, J. R. Lee and M. H. Kim, "Geometric Correction of Vehicle Fish-eye Lens Images," HCI2009, pp. 601-605, Korea, Feb, 2009.
  13. Y. Y. Schechner, S. G. Narasimhan, and S. K. Nayar, "Instant dehazing of images using polarization," in Proc. CVPR, pp. 1984-1991, Dec, 2001.
  14. S. Shwartz, E. Namer, and Y. Y Schechner, "Blind haze separation," in Proc. CVPR, pp. 1984-1991, Oct, 2006.
  15. S. G. Narasimhan and S. K. Nayar, "Chromatic framework for vision in bad weather," in Proc. CVPR, pp. 598-605, June, 2000.
  16. S. G. Narasimhan and S. K. Nayar, "Contrast restoration of weather degraded images," IEEE Trans. Pattern Anal. Mach, Intell, vol. 25, no. 6, pp. 713-724, June 2003. https://doi.org/10.1109/TPAMI.2003.1201821
  17. S. K. Nayar and S. G. Narasimhan, "Vision in bad weather," in Proc. ICCV, pp. 820-827, Corfu, Greece, Sep. 1999.
  18. J. Kopf, B. Neubert, B. Chen, M. Cohen, D. Cohen-Or, O. Deussen, M. Uyttendaele, and D. Lischinski, "Deep photo: Model-based photograph enhancement and viewing," ACM Trans. Graphics, Vol. 27, no. 5, pp. 1-10, Dec, 2008.
  19. K. He, J. Sun, and X. Tang, "Single image haze removal using dark channel prior," in Proc. CVPR, June 2009.
  20. J. H. Kim, C. S. Kim. "Hierarchical Haze Removal Using Dark Channel Prior," The transactions of The Korean Institute of Electrical Engineers, Vol. 59, no. 2, pp. 457-464, Feb, 2010.
  21. Z. Zhang, "A flexible new technique for camera calibration," IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 22, No. 11, pp. 1330-1334, 2000. https://doi.org/10.1109/34.888718
  22. J. Kannala, "A Generic Camera Calibration Method for Fish-Eye Lenses," IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 28, No. 8, pp. 1335-1340, 2006. https://doi.org/10.1109/TPAMI.2006.153
  23. S. B. Kang, "Semi-automatic methods for recovering radial distortion parameters from a single image," Technical Reports Series CRL 97/3, pp. 1-21, May, 1997.
  24. K. Zuiderveld, "Contrast limited adaptive histogram equalization", Graphics Gems IV, pp. 474-485, 1944.
  25. N. J. Kwak and J. H. Hwang, "Extraction of Representative Color of Digital Images Using Histogram of Hue Area and Non-Hue Area," The Institute of Electronics Engineers of Korea Signal Processing, Vol. 47, no. 2, pp. 1-10, Mar, 2010.
  26. S. M. Pizer et al, "Adaptive histogram equalization and its variations," Computer Vision Graphics and Image Processing, Vol. 39, pp. 355-368, Sep, 1987. https://doi.org/10.1016/S0734-189X(87)80186-X
  27. Restrepo, Alfredo and Giovanni Ramponi, "Word descriptors of image quality based on local dispersion-versus-location distributions," IEEE, Signal Processing Conference, 2008 16th European, pp. 1-5, 2008.
  28. R. Fattal, "Single image dehazing," ACM Trans. Graphics, Vol. 27, no. 3, pp. 1-9, Aug, 2008.
  29. R. Tan, "Visibility in bad weather from a single image", in Proc. CVPR, pp. 1-8, Jun, 2008.
  30. J. P. Tarel and N. Hautiere, "Fast visibility restoration from a single color or gray level image," in Proceedings of IEEE International Conference on Computer Vision(ICCV '09), pp. 2201-2208, Kyoto, Japan, 2009.
  31. S. B. Park, Y. J. Cha, S. H Park and N. Y Kwak. "Fast Dehazing Method Using Variable Window-based Dark Channel Prior," Proceedings of Symposium of the Korean Institute of communications and Information Sciences, pp. 209-212, Jan, 2016.
  32. T. H. Kang, Y. J. Cha, W. C. Oh and N. Y Kwak. "Adaptive Airlight Detection for Dark Channel Prior-based Haze Removal." Proceedings of Symposium of the Korean Institute of communications and Information Sciences, pp. 429-430, Jun, 2014.