Background subtraction using depth histogram for object tracking.

Authors

  • E. N. Amirgaliev Аl-Farabi Kazakh National University, Almaty, Kazakhstan
  • A. K. Nussipbekov Аl-Farabi Kazakh National University, Almaty, Kazakhstan
  • G. A. Nussipbekova International IT University, Almaty, Kazakhstan

Keywords:

histogram equalization, background subtraction, object track ing,

Abstract

Background subtraction is an important task in tracking probl ems. It is one of the first steps one has to do in a preprocessing stage. Background subtra ction is used for detecting moving objects on the scene. The problem has been addressed in many research works however it is still challenging task. In this paper we demonstrate how background subtraction can be efficiently done using histogram of depth image. The main difference is that depth histogram shows us how far objects are standing from camera which in turn gives us possibility to separate objects of our interest from background. We use iterative thresholding enhanced with k-means clustering for extraction object of o ur interest from background. Finally we combine depth and RGB images to obtain real picture of object. In our work we create an application written on C# (.NET). Experiment was done using Microsoft Kinect depth camera.

References

C.S. Rao and P. Darwin Frame Difference And Kalman Filter Techniques For Detection Of Moving Vehicles In Video Surveillance. // International Journal of Engineering Research and Applications. – V.2. – No.6. – P. 1168–1170.

S.S. Cheung and C. Kamath Robust techniques for background subtraction in urban traffic video. // Visual Communications and Image Processing 2004. – Proceedings of the SPIE. – 2004. – V.5308. – P. 881–892.

T.W. Ridler and S. Calvard Picture Thresholding Using an Iterative Selection Method. // IEEE Transactions on Systems, Man, and Cybernetics. – V.smc-8. – No.8. – August 1978. – P. 630–632.

Downloads

Published

2014-03-26