Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
534695 | Pattern Recognition Letters | 2011 | 10 Pages |
A novel spiral band model is presented instead of existing curvilinear ones to locate Tropical Cyclone (TC) centers automatically. Particle swarm optimization is used to optimize the model parameters. Experimental results on infrared satellite cloud images show that the model can achieve the best average error of 0.1223° (about 13.58 km) on both latitude and longitude in comparison with the best track data. It can assist forecasters in locating TC centers by providing objective information.
Research highlights► In order to locate Tropical Cyclones automatically, a spiral band model is designed. The information of spiral pattern of the spiral rain cloud bands can be utilized better than existing curvilinear models. The locating approach converts the locating problem to a pattern-matching one. And the process of searching the best estimates of the model parameters is performed by particle swarm optimization method. The result is encouraging.