A crowd flow estimation method based on dynamic texture and GRNN

To overcome the deficiencies of the existing methods used in the estimation of the crowd flow with high-density and multi-motion direction, a crowd flow estimation method based on dynamic texture and generalized regression neural network (GRNN) is presented in this paper. The method firstly extracts the dynamic texture features through optical flow, performs the moving crowd segmentation by the dynamic texture features and level set algorithm to achieve ROIs, and then the regression analysis based on GRNN between ROI features and crowd flow is adopted to achieve the real-time crowd flow estimation results in the crowd scene. Experimental results show that the proposed crowd flow estimation algorithm is more suitable than the existing methods to the crowd flow estimation applications with low complexity, high accuracy and high real-time requirements.

[1]  Chunming Li,et al.  Level set evolution without re-initialization: a new variational formulation , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[2]  Yan Chen,et al.  Classifying image texture with statistical landscape features , 2006, Pattern Analysis and Applications.

[3]  Li Zhang,et al.  A dynamic texture-based method for multi-directional motions segmentation of crowd , 2011, 2011 4th International Congress on Image and Signal Processing.

[4]  Paul A. Viola,et al.  Detecting Pedestrians Using Patterns of Motion and Appearance , 2005, International Journal of Computer Vision.

[5]  A. Marana,et al.  On the efficacy of texture analysis for crowd monitoring , 1998, Proceedings SIBGRAPI'98. International Symposium on Computer Graphics, Image Processing, and Vision (Cat. No.98EX237).

[6]  Nikos Paragios,et al.  A MRF-based approach for real-time subway monitoring , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[7]  Luciano da Fontoura Costa,et al.  Estimating crowd density with Minkowski fractal dimension , 1999, 1999 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings. ICASSP99 (Cat. No.99CH36258).

[8]  Sheng-Fuu Lin,et al.  Estimation of number of people in crowded scenes using perspective transformation , 2001, IEEE Trans. Syst. Man Cybern. Part A.

[9]  Nuno Vasconcelos,et al.  Privacy preserving crowd monitoring: Counting people without people models or tracking , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

[10]  Ramakant Nevatia,et al.  Detection of multiple, partially occluded humans in a single image by Bayesian combination of edgelet part detectors , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.

[11]  Hai Tao,et al.  Counting Pedestrians in Crowds Using Viewpoint Invariant Training , 2005, BMVC.

[12]  Sergio A. Velastin,et al.  Crowd monitoring using image processing , 1995 .

[13]  Thomas Brox,et al.  High Accuracy Optical Flow Estimation Based on a Theory for Warping , 2004, ECCV.

[14]  Ramakant Nevatia,et al.  Bayesian human segmentation in crowded situations , 2003, 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings..