Cyclone Tracking using Multiple Heterogeneous Satellite Data Sources
Abstract: It is impractical to use a single Earth orbiting satellite to detect and track events such as cyclones in a continuous manner due to limited spatial and temporal coverage. One solution to alleviate such persistent problems is to utilize heterogeneous sensor data from multiple orbiting satellites. However, this solution requires overcoming other new challenges such as (i) varying spatial and temporal resolution between satellite sensor data, (ii) the need to establish correspondence between features from different satellite sensors, and (iii) the lack of definitive indicators for cyclone events in some sensor data.We describe an automated cyclone discovery and tracking approach using heterogeneous sensor data from multiple satellites. This approach addresses the unique challenges associated with mining and data discovery from heterogeneous satellite data streams. Experimental results on historical hurricane datasets demonstrate the superior performance of our automated approach compared to previous work.
Grant:
NASA Postdoctoral Program Fellowship, "Robust Machine Learning Techniques for Detection and Tracking of Cyclones and Other Events using Multiple Heterogeneous Remote Sensing Satellite Data", Aug. 2007-Aug. 2009, PI: Shen-Shyang Ho.
References:
A. Panangadan, S.-S. Ho, and A. Talukder, Cyclone Tracking using Multiple Satellite Image Sources'', Proc. of 17th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, Seattle, WA, 4-6 Nov, 2009.
S.-S. Ho, Utilizing Spatio-Temporal Text Information for Cyclone Eye Annotation in Satellite Data, IJCAI workshop on Cross-Media Information Access and Mining, Pasadena, CA, 13 July, 2009
A. Talukder and S.-S. Ho, Remote Event Detection and Tracking using Multiple Heterogeneous Satellite Data Fusion'', Proc. of SPIE, Optical Pattern Recognition XX, Orlando, Florida, 16-17 April, 2009. (Invited Paper)
S.-S. Ho and A. Talukder, Automated Cyclone Tracking using Multiple Remote Satellite Data via Knowledge Transfer, IEEE Aerospace Conference 2009, Big Sky, MT, 7-14 March, 2009.
S.-S. Ho, A. Talukder, T. Liu, W. Tang, and A. Bingham, Automated Historical and Real-time Cyclone Discovery with Multi-model Remote Satellite Measurements, American Geophysical Union Fall Meeting, San Francisco, CA, 15-19 December, 2008 [Extended Abstract].
A. Talukder and S.-S. Ho, Classification from Disparate Multiple Streaming Data Sources, NIPS-08 workshop, Learning from Multiple Sources, Vancouver, Canada, 13 Dec, 2008.
S.-S. Ho and A. Talukder, Automated Cyclone Discovery and Tracking using Multiple Heterogeneous Satellite Data, Proc. 14th ACM SIGKDD Int. Conf. on Knowledge Discovery and Data Mining, Las Vegas, NV, 24-27 Aug, 2008.
S.-S. Ho and A. Talukder, Cyclone Tracking Using Multiple Satellite Data Sources via Spatial-Temporal Knowledge Transfer, AAAI-08 workshop, Transfer Learning for Complex Tasks, Chicago, IL, 14 July, 2008.
A. Talukder, S.-S. Ho, T. Liu, W. Tang, A. Bingham, and E. Rigor, Global Cyclone Detection and Tracking using Multiple Remote Satellite Data, NASA Earth Science Technology Conference, Aldephi, MD, 24 June, 2008.
S.-S. Ho and A. Talukder, Automated Cyclone Identification from Remote QuikSCAT Satellite Data, IEEE Aerospace Conference 2008, Big Sky, MT, Mar. 1-8, 2008.