DOI QR코드

DOI QR Code

Change Area Detection using Color and Edge Gradient Covariance Features

색상과 에지 공분산 특징을 이용한 변화영역 검출

  • Kim, Dong-Keun (Division of Computer Science and Engineering, Kongju National University) ;
  • Hwang, Chi-Jung (Dept. of Computer Engineering, Chungnam National University)
  • 김동근 (공주대학교 컴퓨터공학부) ;
  • 황치정 (충남대학교 컴퓨터공학과)
  • Received : 2015.09.02
  • Accepted : 2016.01.05
  • Published : 2016.01.31

Abstract

This paper proposes a change detection method based on the covariance matrices of color and edge gradient in a color video. The YCbCr color format was used instead of RGB. The color covariance matrix was calculated from the CbCr-channels and the edge gradient covariance matrix was calculated from the Y-channels. The covariance matrices were effectively calculated at each pixel by calculating the sum, squared sum, and sum of two values' multiplication of a rectangle area using the integral images from a background image. The background image was updated by a running the average between the background image and a current frame. The change areas in a current frame image against the background were detected using the Mahalanobis distance, which is a measure of the statistical distance using covariance matrices. The experimental results of an expressway color video showed that the proposed approach can effectively detect change regions for color and edge gradients against the background.

본 논문은 카메라로부터 획득한 컬러 비디오 영상에서 컬러 색상과 에지 그래디언트의 공분산 행렬에 기반한 영상의 변화영역을 검출하는 방법을 제안한다. 컬러 비디오 영상은 RGB 영상 대신에 밝기정보와 색상정보가 분리된 YCbCr 컬러비디오 포맷을 사용한다. CbCr-채널로부터 컬러의 색상분포를 알 수 있는 컬러 공분산 행렬을 계산하며, Y-채널로부터는 영상의 에지 그래디언트 분포를 알 수 있는 공분산 행렬을 계산한다. 컬러 공분산 행렬과 에지 그래디언트 공분산 행렬은 배경영상으로부터 적분영상을 사용하여 사각영역의 합계와 제곱 합계, 곱셈 합계를 효과적으로 계산하여 각 화소에서 빠르게 계산된다. 또한 시간에 따른 변화를 반영하기 위하여 배경영상과 입력영상의 가중평균에 의해 배경영상을 갱신한다. 현재 프레임에서의 배경영상으로부터의 변화영역은 컬러 공분산 행렬과 에지 그래디언트 공분산 행렬을 사용한 통계적 거리측정인 마하라노비스 거리를 이용하여 검출한다. 고속도로의 컬러 비디오 영상의 실험결과에서 컬러색상과 그래디언트의 변화영역을 효과적으로 검출할 수 있었다.

Keywords

References

  1. Richard J. Radke, et., al., "Image Change Detection Algorithms: A Systematic Survey." IEEE Trans. on Image processing, vol.14, no.3(2005), pp.294-307. DOI: http://dx.doi.org/10.1109/TIP.2004.838698
  2. Alan M. McIvor, "Background Subtraction Techniques," In Processings of Image & Vision Computing New Zealand, IVCNZ, 2000.
  3. Massimo Piccardi, "Background subtraction techniques: a review," IEEE International Conference on Systems, Man and Cybernetics, pp.3099-3104, 2004. DOI: http://dx.doi.org/10.1109/icsmc.2004.1400815
  4. C. Stauffer and E. Grimson, "Adaptive background mixture models for real-time tracking", CVPR, vol.2, pp. 246-252, 1998.
  5. Kevin Mader and Gil Reese, "Using Covariance Matrices as Feature Descriptors for Vehicle Detection from a Fixed Camera," Digital Image Processing and Communication, http://arxiv.org/pdf/1202.2528.pdf, Boston University, 2006.
  6. O. Tuzel 2006] O. Tuzel, F. Porikli, P. M. "Region covariance: A fast descriptor for detection and classication," ECCV, 2006
  7. Fatih Porikli and Oncel Tuzel "FAST CONSTRUCTION OF COVARIANCE MATRICES FOR ARBITRARY SIZE IMAGE," WINDOWS
  8. W. Forstner and W. Forstner, B. M. "A metric for covariance matrices. TR, Dept. of Geodesy and Geoinfromatics, Stuttgart University, 1999.
  9. F. Crow, "Summed-area tables for texture mapping", Proceedings of SIGGRAPH, 1984. DOI: http://dx.doi.org/10.1145/800031.808600
  10. Paul Viola and Michael Jones, "Rapid object detection using a boosted cascade of simple features" CVPR, 2001.
  11. F. Porikli, "Integral Histogram: A fast way to extract histograms in Cartesian spaces", in Proceedings of CVPR. 2005. DOI: http://dx.doi.org/10.1109/cvpr.2005.188
  12. Geum-boon Lee and Beom-joon Cho, "A Novel Method for Moving Object Tracking using Covariance Matrix and Riemannian Metric," Journal of KIMICS vol.15, no.2, pp.364-370, 2011.
  13. Geum-boon Lee, "A Fast Moving Object Tracking Method by the Combination of Covariance Matrix and Kalman Filter Algorithm," JKIICE, vol.19, no.6, pp.1477-1484, 2015.
  14. Dongkeun Kim, "Change Detection using Integral Images in Color Video Sequences," Information vol. 18, pp.229-235, 2015.
  15. Geoffrey John McLachlan, "Mahalanobis Distance," RESONANCE, 1999, pp20-26.
  16. R. De Maesschalck, D. Jouan-Rimbaud, D.L. Massart, "The Mahalanobis distance," Chemometrics and Intelligent Laboratory Systems, vol. 50, pp. 1-18, 2000. DOI: http://dx.doi.org/10.1016/S0169-7439(99)00047-7
  17. DongKeun Kim, "Moving Object Detection using Gaussian Pyramid based Subtraction Images in Road Video Sequences," Journal of the Korea Academia-Industrial Cooperation Society, vol.12, no.12, pp.5856-5864. 2011. DOI: http://dx.doi.org/10.5762/KAIS.2011.12.12.5856