Stochastic tracking of mesoscale convective systems: evaluation in the West African Sahel
In this work we apply a recently proposed Bayesian multiple target tracking model to mesoscale convective systems tracking. This stochastic model follows the multiple hypothesis tracking paradigm and can handle a varying number of targets while detecting the target birth, death, split, and merge events. The model is tested experimentally with real MCS targets detected from meteosat IR data over the Sahelian region. The performance of the stochastic tracking is evaluated by comparing it qualitatively and quantitatively with well established deterministic methods.
- Bar-Shalom Y, Willett PK, Tian X (2011) Tracking and data fusion: a handbook of algorithms, YBS Publishing, StorrsGoogle Scholar
- Goutte C, Gaussier E (2005) A probabilistic interpretation of precision, recall and f-score, with implication for evaluation. In: Advances in information retrieval. Springer, Berlin, pp 345–359Google Scholar
- Storlie C, Lee T, Hannig J, Nychka D (2009) Tracking of multiple merging and splitting targets: a statistical perspective. Statistica Sinica 19(1):1–52. doi:URL http://nldr.library.ucar.edu/repository/collections/OSGC-000-000-000-072
- Ulmke M, Erdinc O, Willett P (2007) Gaussian mixture cardinalized phd filter for ground moving target tracking. In: Information fusion, 2007 10th international conference on, pp. 1–8. doi: 10.1109/ICIF.2007.4408105Google Scholar
- Vischel T, Lebel T, Massuel S, Cappelaere B (2009) Conditional simulation schemes of rain fields and their application to rainfallrunoff modeling studies in the sahel. J Hydrol 375(12), 273–286. doi:10.1016/j.jhydrol.2009.02.028. Surface processes and water cycle in West Africa, studied from the AMMA-CATCH observing system
- Vo BT, Vo BN, Cantoni A (2006) The cardinalized probability hypothesis density filter for linear gaussian multi-target models. In: Information sciences and systems, 2006 40th annual conference on, pp. 681–686. doi: 10.1109/CISS.2006.286554Google Scholar