Adaptive Foreground Extraction for Crowd Analytics Surveillance on Unconstrained Environments

Background modeling is one of the key steps in any visual surveillance system. A good background modeling algorithm should be able to detect objects/targets under any environmental condition. The influence of illumination variance has been a major challenge in many background modeling algorithms. These algorithms produce poor object segmentation or consume substantial amount of computational time, which makes them not implementable at real time. In this paper we propose a novel background modeling method based on Gaussian Mixture Method (GMM). The proposed method uses Phase Congruency (PC) edge features to overcome the effect of illumination variance, while preserving efficient background/foreground segmentation. Moreover, our method uses a combination of pixel information of GMM and the Phase texture information of PC, to construct a foreground invariant of the illumination variance.

[1]  Thierry Bouwmans,et al.  Recent Advanced Statistical Background Modeling for Foreground Detection - A Systematic Survey , 2011 .

[2]  Ferdinand van der Heijden,et al.  Recursive unsupervised learning of finite mixture models , 2004, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[3]  Aamir Saeed Malik,et al.  Comparison of stochastic filtering methods for 3D tracking , 2011, Pattern Recognit..

[4]  Zezhi Chen,et al.  A self-adaptive Gaussian mixture model , 2014, Comput. Vis. Image Underst..

[5]  Peter Kovesi,et al.  Phase Congruency Detects Corners and Edges , 2003, DICTA.

[6]  W. Eric L. Grimson,et al.  Adaptive background mixture models for real-time tracking , 1999, Proceedings. 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No PR00149).

[7]  Mohan M. Trivedi,et al.  A Survey of Vision-Based Trajectory Learning and Analysis for Surveillance , 2008, IEEE Transactions on Circuits and Systems for Video Technology.

[8]  Xiaogang Wang,et al.  Intelligent multi-camera video surveillance: A review , 2013, Pattern Recognit. Lett..

[9]  Aamir Saeed Malik,et al.  Mixture of Gaussian based background modelling for crowd tracking using multiple cameras , 2014, 2014 5th International Conference on Intelligent and Advanced Systems (ICIAS).

[10]  Jen-Hui Chuang,et al.  Regularized Background Adaptation: A Novel Learning Rate Control Scheme for Gaussian Mixture Modeling , 2011, IEEE Transactions on Image Processing.

[11]  Raúl Mohedano,et al.  Versatile Bayesian classifier for moving object detection by non-parametric background-foreground modeling , 2012, 2012 19th IEEE International Conference on Image Processing.

[12]  Álvaro Sánchez Miralles,et al.  Mixture of Merged Gaussian Algorithm using RTDENN , 2013, Machine Vision and Applications.

[13]  Aamir Saeed Malik,et al.  Foreground extraction for real-time crowd analytics in surveillance system , 2014, The 18th IEEE International Symposium on Consumer Electronics (ISCE 2014).

[14]  Li Song,et al.  Background subtraction based on phase feature and distance transform , 2012, Pattern Recognit. Lett..

[15]  Pascal Fua,et al.  Making Background Subtraction Robust to Sudden Illumination Changes , 2008, ECCV.

[16]  M. K. Bhuyan,et al.  Multiple camera-based codebooks for object detection under sudden illumination change , 2013, 2013 International Conference on Communication and Signal Processing.

[17]  Takeshi Ikenaga,et al.  Adaptively Adjusted Gaussian Mixture Models for Surveillance Applications , 2010, MMM.

[18]  Larry S. Davis,et al.  Non-parametric Model for Background Subtraction , 2000, ECCV.

[19]  Erik D. Goodman,et al.  Online background learning for illumination-robust foreground detection , 2010, 2010 11th International Conference on Control Automation Robotics & Vision.