Big Data Scalability Issues in WAAS

Wide Area Aerial Surveillance (WAAS) produces very large images at 1-2 fps or more. This data needs to be processed in real time to produce semantically meaningful information, then queried efficiently. We have designed and implemented a full system to detect and track vehicles, and infer activities. We address here the scalability issues, and propose solutions to have the tracker run in real time using different parallelism strategies. We also describe methods to efficiently query the data in forensic mode. Our methods are validated on large scale real world data, and have been transferred to a National Laboratory for deployment.

[1]  Margaret H. Dunham,et al.  Join processing in relational databases , 1992, CSUR.

[2]  Haroon Idrees,et al.  Detection and Tracking of Large Number of Targets in Wide Area Surveillance , 2010, ECCV.

[3]  Gérard G. Medioni,et al.  Motion pattern interpretation and detection for tracking moving vehicles in airborne video , 2009, 2009 IEEE Conference on Computer Vision and Pattern Recognition.

[4]  M. Keck,et al.  Real-time tracking of low-resolution vehicles for wide-area persistent surveillance , 2013, 2013 IEEE Workshop on Applications of Computer Vision (WACV).

[5]  Gérard G. Medioni,et al.  Inferring tracklets for multi-object tracking , 2011, CVPR 2011 WORKSHOPS.

[6]  Gérard G. Medioni,et al.  Tensor Voting: A Perceptual Organization Approach to Computer Vision and Machine Learning , 2006, Tensor Voting.

[7]  A. G. Amitha Perera,et al.  Multi-Object Tracking Through Simultaneous Long Occlusions and Split-Merge Conditions , 2006, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06).

[8]  Hanan Samet,et al.  Foundations of multidimensional and metric data structures , 2006, Morgan Kaufmann series in data management systems.

[9]  Oliver E. Drummond,et al.  Performance metrics for multiple-sensor multiple-target tracking , 2000, SPIE Defense + Commercial Sensing.

[10]  Rick Cattell,et al.  Scalable SQL and NoSQL data stores , 2011, SGMD.

[11]  Gérard G. Medioni,et al.  Dimensionality Estimation, Manifold Learning and Function Approximation using Tensor Voting , 2010, J. Mach. Learn. Res..

[12]  Harpreet S. Sawhney,et al.  Vehicle detection and tracking in wide field-of-view aerial video , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[13]  Jongmoo Choi,et al.  Activity recognition in wide aerial video surveillance using entity relationship models , 2012, SIGSPATIAL/GIS.

[14]  Gérard G. Medioni,et al.  Robust unsupervised motion pattern inference from video and applications , 2011, 2011 International Conference on Computer Vision.

[15]  Gérard G. Medioni,et al.  Tracking many vehicles in wide area aerial surveillance , 2012, 2012 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops.

[16]  Bernhard P. Wrobel,et al.  Multiple View Geometry in Computer Vision , 2001 .

[17]  Gérard G. Medioni,et al.  Accurate efficient mosaicking for Wide Area Aerial Surveillance , 2012, 2012 IEEE Workshop on the Applications of Computer Vision (WACV).