DOI QR코드

DOI QR Code

Vehicle Detection Method Based on Object-Based Point Cloud Analysis Using Vertical Elevation Data

OBPCA 기반의 수직단면 이용 차량 추출 기법

  • Received : 2016.02.03
  • Accepted : 2016.04.15
  • Published : 2016.08.31

Abstract

Among various vehicle extraction techniques, OBPCA (Object-Based Point Cloud Analysis) calculates features quickly by coarse-grained rectangles from top-view of the vehicle candidates. However, it uses only a top-view rectangle to detect a vehicle. Thus, it is hard to extract rectangular objects with similar size. For this reason, accuracy issue has raised on the OBPCA method which influences on DEM generation and traffic monitoring tasks. In this paper, we propose a novel method which uses the most distinguishing vertical elevations to calculate additional features. Our proposed method uses same features with top-view, determines new thresholds, and decides whether the candidate is vehicle or not. We compared the accuracy and execution time between original OBPCA and the proposed one. The experiment result shows that our method produces 6.61% increase of precision and 13.96% decrease of false positive rate despite with marginal increase of execution time. We can see that the proposed method can reduce misclassification.

점 클라우드로부터 차량을 추출하는 다양한 방식 중 OBPCA 방식은 세그먼트 단위의 평가-분류로 정확도가 높고, 단순한 직사각형 평면도에서 특성 값들을 추출하므로 분류가 빠르다. 그러나 이 OBPCA 방식은 차량과 크기가 비슷한 직육면체 모양의 물체를 차량과 구별하지 못하는 문제를 가지므로 이를 극복하고 차량 추출의 정확도를 높이는 방안에 대한 연구가 필요하다. 본 논문에서는 이 문제를 해결하기 위해 수평 단면과 함께 수직 단면을 이용하는 확장 OBPCA 방식을 제안한다. 제안 방법은 수평 단면을 통해 차량 후보를 1차로 선별하고, 각 차량 후보에서 가장 특징적인 수직 단면을 찾아서 그 단면의 특성 값들을 임계값들과 비교하여 차량 여부를 판단한다. 비교실험에서는 본 제안방식이 기존 OBPCA 방식에 비해 정밀도가 6.61% 향상되고 위양성률이 13.96% 감소됨을 확인했으며, 이를 통해 제안 방식이 기존 OBPCA 분류오류 문제에 대해 효과적인 해결방안임을 보였다.

Keywords

References

  1. B. Yang, P. Sharma, and R. Nevatia, "Vehicle detection from low quality aerial LIDAR data," IEEE Workshop on Applications of Computer Vision (WACV), pp.541-548, 2011.
  2. F. Rottensteiner, "Advanced Methods for Automated Object Extraction from LiDAR in Urban Areas," Geoscience and Remote Sensing Symposium (IGARSS), IEEE International, pp.5402-5405, 2012.
  3. X. Feng, "Artificial neural networks forecasting of PM2.5 pollution using air mass trajectory based geographic model and wavelet transformation," Atmospheric Environment, Vol.107, pp.118-128, 2015. https://doi.org/10.1016/j.atmosenv.2015.02.030
  4. J. Han, M. Kamber, and J. Pei, "Data mining concepts and techniques," Morgan Kaufmann Pub., pp.1-5, 2012.
  5. J. Zhang, M. Duan, Q. Yan, and X. Lin, "Automatic vehicle extraction from airborne LiDAR data using an object-based point cloud analysis method," Remote Sensing, Vol.6, No.9, pp.8405-8423, 2014. https://doi.org/10.3390/rs6098405
  6. W. Yao, "Comparison of Two Methods for Vehicle Extraction From Airborne LiDAR Data Toward Motion Analysis," IEEE Geoscience and Remote Sensing Society, Vol.8, Issue 4, pp. 607-611, 2011. https://doi.org/10.1109/LGRS.2010.2097239
  7. J. Secord and A. Zakhor, "Tree Detection in Urban Regions Using Aerial Lidar and Image Data," Geoscience and Remote Sensing Letters, IEEE, Vol.4, Issue 2, pp.196-200, 2007.
  8. W. Yao, S. Hinz, and U. Stilla, "3D object-based classification for vehicle extraction from airborne LiDAR data by combining point shape information with spatial edge," 6th IAPR Workshop Pattern Recognition in Remote Ssensing, pp.1-4, 2010.
  9. B. Guo, X. Huang, F. Zhang, and G. Sohn, "Classification of airborne laser scanning data using JointBoost," ISPRS Journal of Photogrammetry and Remote Sensing, Vol.100, pp.71-83, 2015. https://doi.org/10.1016/j.isprsjprs.2014.04.015
  10. M. Atwah, and J. Baker, "An Associative Implementation Of Graham's Convex Hull Algorithm," Parallel and Distributed Computing, 1995.
  11. I. Jolliffe, "Principal Component Analysis," in Wiley StatsRef: Statistics Reference Online, John Wiley & Sons, Ltd, 2002.