Abstract
In this paper we propose a new method for extracting silhouettes of moving objects in images acquired with a static camera. First the background subtraction algorithm with an adaptive Gaussian mixture model is used to obtain moving regions. The output binary mask is then refined using a region-filtering algorithm based on an adaptive fast-scanning segmentation algorithm. Next, the resulting mask is morphologically processed in order to prepare the input for the GrabCut algorithm. Finally, the GrabCut algorithm leverages spatial and color relationships between pixels in order to improve the background subtraction result. We show through experiments that for certain types of video sequences our approach can perform better than state-of-the-art methods as regards the mask accuracy.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Bouwmans, T.: Traditional and recent approaches in background modeling for foreground detection: An overview. Computer Science Review 11, 31-66 (2014)
Zivkovic, Z., van der Heijden, F.: Efficient adaptive density estimation per image pixel for the task of background subtraction. Pattern recognition letters 27(7), 773-780 (2006)
Bouwmans, T., El Baf, F., Vachon, B.: Background Modeling using Mixture of Gaussians for Foreground Detection - A Survey. Recent Patents on Computer Science, Bentham Science Publishers, 1(3), 219-237 (2008)
Kim, H., et al.: Robust Silhouette Extraction Technique using Background Subtraction. 10th Meeting on Image Recognition and Understand (MIRU), Japan, 1-6 (2007)
Konushin, V., Konushin, A:. Improvement of background subtraction by mask constraints. Proc. GraphiCon., 96-99 (2010)
Rother, C., Kolmogorov, V., Blake, A.: Grabcut - Interactive Foreground Extraction using Iterated Graph Cuts. ACM Transactions on Graphics (TOG), 23(3), 309-314 (2004)
Gulshan, V., Lempitsky, V., Zisserman, A.: Humanising GrabCut: Learning to segment humans using the Kinect. Workshop on Consumer Depth Cameras in Computer Vision, (ICCV), 1-7 (2011)
Hernández-Vela, A., Reyes, M., Ponce, V., Escalera, S.: GrabCut-based Human Segmentation in Video Sequences. Sensors 12 (11), 15376-15393 (2012)
Tomasz Posłuszny among OpenCV patches’ contributors (2014), http://opencv.org/opencv-3-0-beta.html
Ding, J.-J., Kuo C.J., Hong, W.C.: An efficient image segmentation technique by fast scanning and adaptive merging. CVGIP, 1-9 (2009)
Wallflower Test Images, http://research.microsoft.com/users/jckrumm/WallFlower/TestImages.htm
CDnet: a video database for testing change detection algorithms, http://www.changedetection.net
Wang, Y., et al.: CDnet 2014: An Expanded Change Detection Benchmark Dataset. Computer Vision and Pattern Recognition Workshops (CVPRW), 393-400 (2014)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer International Publishing Switzerland
About this paper
Cite this paper
Posłuszny, T., Putz, B. (2016). An Improved Extraction Process of Moving Objects’ Silhouettes in Video Sequences. In: Jabłoński, R., Brezina, T. (eds) Advanced Mechatronics Solutions. Advances in Intelligent Systems and Computing, vol 393. Springer, Cham. https://doi.org/10.1007/978-3-319-23923-1_9
Download citation
DOI: https://doi.org/10.1007/978-3-319-23923-1_9
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-23921-7
Online ISBN: 978-3-319-23923-1
eBook Packages: EngineeringEngineering (R0)