Rotating adaptive Haar wavelet transform for human tracking in thermal omnidirectional vision

In this paper, a novel surveillance system, thermal omnidirectional vision system, is introduced which is robust to illumination and has a global field of view. According to the characteristic of the proposed system, a rotating adaptive Haar wavelet transform is developed for human tracking in thermal omnidirectional vision. The proposed feature can effectively handle the nonisotropic distortion of catadioptric omnidirectional vision (COV). For robust tracking, we develop a rotational kinematic model based adaptive particle filter, which can handle various movements including rapid movement. Since the involvement of the rotational kinematic model, the proposed tracking algorithm can well deal with the short term occlusion. Finally, a series of experiments verify the effectiveness of the proposed rotating adaptive Haar wavelet transform and the rotational kinematic model based adaptive particle filter for human tracking in thermal omnidirectional vision.

[1]  Xin Li,et al.  Pedestrian detection and tracking in infrared imagery using shape and appearance , 2007, Comput. Vis. Image Underst..

[2]  Xiang Gao,et al.  Omni-directional visual surveillance , 2004, Image Vis. Comput..

[3]  Sheng Liu,et al.  Optimal Feature Matching for 3D Reconstruction by Combination of Global and Local Information , 2011, Intell. Autom. Soft Comput..

[4]  Michael Isard,et al.  CONDENSATION—Conditional Density Propagation for Visual Tracking , 1998, International Journal of Computer Vision.

[5]  Nando de Freitas,et al.  An Introduction to Sequential Monte Carlo Methods , 2001, Sequential Monte Carlo Methods in Practice.

[6]  Tomaso A. Poggio,et al.  A Trainable System for Object Detection , 2000, International Journal of Computer Vision.

[7]  Xia Liu,et al.  Pedestrian detection and tracking with night vision , 2005, IEEE Transactions on Intelligent Transportation Systems.

[8]  Zhang Cheng,et al.  Tracking unknown moving targets on omnidirectional vision , 2009, Vision Research.

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

[10]  Grzegorz Cielniak,et al.  Real-time people tracking for mobile robots using thermal vision , 2006, Robotics Auton. Syst..

[11]  Pascal Vasseur,et al.  Particle Filter Approach Adapted to Catadioptric Images for Target Tracking Application , 2009, BMVC.

[12]  S. Y. Chen,et al.  Kalman Filter for Robot Vision: A Survey , 2012, IEEE Transactions on Industrial Electronics.

[13]  Pedro U. Lima,et al.  Omni-directional catadioptric vision for soccer robots , 2001, Robotics Auton. Syst..

[14]  José Santos-Victor,et al.  Vision-based navigation and environmental representations with an omnidirectional camera , 2000, IEEE Trans. Robotics Autom..

[15]  Sheng Liu,et al.  Particle Filter with Affine Transformation for Multiple Key Points Tracking , 2012, Trans. Edutainment.

[16]  R. Douglas Gregory.  Classical Mechanics: Vector angular velocity and rigid body kinematics , 2006 .

[17]  James W. Davis,et al.  A Two-Stage Template Approach to Person Detection in Thermal Imagery , 2005, 2005 Seventh IEEE Workshops on Applications of Computer Vision (WACV/MOTION'05) - Volume 1.

[18]  Neil J. Gordon,et al.  A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking , 2002, IEEE Trans. Signal Process..

[19]  PoggioTomaso,et al.  A Trainable System for Object Detection , 2000 .

[20]  Eduardo Bayro-Corrochano,et al.  Omnidirectional Vision Tracking with Particle Filter , 2006, 18th International Conference on Pattern Recognition (ICPR'06).

[21]  Noboru Yasuda,et al.  Pedestrian Detection and Tracking in Far Infrared Images , 2004, 2004 Conference on Computer Vision and Pattern Recognition Workshop.