Abstract
In this paper, we address the problem of scene background initialization to define a background model free from foreground objects. The complexity of this task resides in the continuous clutter of the scene by moving and stationary objects. To face this challenge, we propose a robust real-time iterative model completion method based on online block-level processing to initialize the background with low computational cost. First, temporal data analysis is conducted to cluster similar blocks. Meanwhile, a two-folded inter-block spatial neighborhood exploration is performed. It aims to capture relationships among neighboring clusters and reduce the number of candidate clusters employed in the next phase. Then, a smoothness analysis between neighboring locations is performed to iteratively reconstruct the background based on a newly proposed edge matching metric and an inter-block color discontinuity. Extensive evaluations of the proposed approach on the public Scene Background Initialization 2015 dataset and on the Scene Background Modeling Contest 2016 dataset revealed a performance superior or comparable to state-of-the-art methods.
Similar content being viewed by others
Notes
For comparison with RMR, we used the software available at: http://www-vpu.eps.uam.es/publications/BE_RMR/
References
Agarwala A, Dontcheva M, Agrawala M, Drucker S, Colburn A, Curless B, Salesin D, Cohen M (2004) Interactive digital photomontage. In: ACM Transactions on graphics (TOG), vol 23, pp 294–302. ACM
Baltieri D, Vezzani R, Cucchiara R (2010) Fast background initialization with recursive hadamard transform. In: Advanced video and signal based surveillance (AVSS), 2010 seventh IEEE international conference on, pp 165–171. IEEE
Bloisi DD, Grillo A, Pennisi A, Iocchi L, Passaretti C (2015) Multi-modal background model initialization. In: International conference on image analysis and processing, pp 485–492. Springer
Bouwmans T (2014) Traditional and recent approaches in background modeling for foreground detection An overview. Computer Science Review 11:31–66
Bouwmans T, Maddalena L, Petrosino A (2017) Scene background initialization: a taxonomy Pattern Recognition Letters
John C (1986) A computational approach to edge detection. IEEE Transactions on pattern analysis and machine intelligence, (6) pp 679–698
Colombari A, Fusiello A (2010) Patch-based background initialization in heavily cluttered video. IEEE Trans Image Process 19(4):926–933
Davies DL, Bouldin DW (1979) A cluster separation measure. IEEE transactions on pattern analysis and machine intelligence, (2) pp 224–227
Gregorio MD, Giordano M (2017) Background estimation by weightless neural networks Pattern Recognition Letters
Halkidi M, Batistakis Y, Vazirgiannis M (2001) On clustering validation techniques. Journal of intelligent information systems 17(2):107–145
Laugraud B, Piérard S, Droogenbroeck MV (2016) Labgen: A method based on motion detection for generating the background of a scene. Pattern Recognition Letters
Laugraud B, Piérard S, Droogenbroeck MV (2016) Labgen-p: A pixel-level stationary background generation method based on labgen. In: Pattern recognition (ICPR), 2016 23rd international conference on, pp 107–113. IEEE
Maddalena L, Petrosino A (2012) The sobs algorithm: What are the limits? In Computer Vision and Pattern Recognition Workshops (CVPRW), 2012 IEEE Computer Society Conference on, pp 21–26. IEEE
Maddalena L, Petrosino A (2014) The 3dsobs+ algorithm for moving object detection. Comput Vis Image Underst 122:65–73
Maddalena L, Petrosino A (2014) Background model initialization for static cameras. In: Background modeling and foreground detection for video surveillance, pp 3–1. Chapman and hall/CRC
Maddalena L, Petrosino A (2015) Towards benchmarking scene background initialization. In: International conference on image analysis and processing, pp 469–476. Springer
Ortego D, SanMiguel JC, Martínez JM (2016) Rejection based multipath reconstruction for background estimation in video sequences with stationary objects. Comput Vis Image Underst 147:23–37
Park D, Byun H (2013) A unified approach to background adaptation and initialization in public scenes. Pattern Recogn 46(7):1985–1997
Piccardi M (2004) Background subtraction techniques: a review. In: Systems, man and cybernetics, 2004 IEEE international conference on, vol 4, pp 3099–3104. IEEE
Ramirez-Alonso G, Ramirez-Quintana JA, Chacon-Murguia MI (2017) Temporal weighted learning model for background estimation with an automatic re-initialization stage and adaptive parameters upyear Pattern Recognition Letters
Rao KR, Yip P (1990) Discrete cosine transform: algorithms, advantages, applications Academic press
Reddy V, Sanderson C, Lovell BC (2009) An efficient and robust sequential algorithm for background estimation in video surveillance. In: Image processing (ICIP), 2009 16th IEEE International conference on, pp 1109–1112. IEEE
Reddy V, Sanderson C, Lovell BC (2011) A low-complexity algorithm for static background estimation from cluttered image sequences in surveillance contexts. EURASIP Journal on Image and Video Processing 2011(1):1–14
SMBC2016 (ICPR) (2016) Scene background modeling contest http://pione.dinf.usherbrooke.ca/
SMBI2015 (ICIAP) (2015) Scene background modeling and initialization http://sbmi2015.na.icar.cnr.it/
Sobral A, Vacavant A (2014) A comprehensive review of background subtraction algorithms evaluated with synthetic and real videos. Comput Vis Image Underst 122:4–21
St-Charles P-L, Bilodeau G-A, Bergevin R (2015) Subsense: A universal change detection method with local adaptive sensitivity. Image Processing, IEEE Transactions on 24(1):359–373
Wang H, Suter D (2006) A novel robust statistical method for background initialization and visual surveillance. In: Asian conference on computer vision, pp 328–337. Springer
Xun X, Huang TS (2008) A loopy belief propagation approach for robust background estimation. In: Computer vision and pattern recognition, 2008. CVPR 2008. IEEE conference on, pp 1–7. IEEE
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Mseddi, W.S., Jmal, M. & Attia, R. Real-time scene background initialization based on spatio-temporal neighborhood exploration. Multimed Tools Appl 78, 7289–7319 (2019). https://doi.org/10.1007/s11042-018-6399-1
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-018-6399-1