Jerawan Saerour. Fractal structure of cloud clusters in satellite images. Master's Degree(Applied Mathematics). King Mongkut's University Technology Thonburi. Library. : King Mongkut's University of Technology Thonburi, 2004.
Fractal structure of cloud clusters in satellite images
Abstract:
Heavy rainfall in the tropics usually occurs from an ensemble of organizedthunderstorm clouds called a "cloud cluster". In this research, fractal dimensions ofcloud clusters in satellite images are identified. The higher the value of the fractaldimension, the rougher the surface. For meteorological satellite images, high values offractal dimension represent high cloud tops, which in the tropics, are usuallythunderstorm clouds. The Differential Box-Counting (DBC) method is improved forappropriately applies to fractal structure or changes in time of fractal dimensions ofcloud clusters in Northeast Thailand which experiences frequent occurrence of cloudclusters. The images used are IR1 and IR2 images from the GMS-5 meteorologicalsatellite. In the first step of improvement, the box size is modified to better suit thesatellite images. This modified DBC method is called the Improved Differential BoxCounting(IDBC) method. The IDBC fractal dimensions during the 3-hour period; 1hour before the mature stage, the mature stage, and 1 hour after the mature stage ofcloud clusters are selected for analysis. The IDBC method is further improved in thesecond step by adjusting the IDBC fractal dimension with the average gray level of theimage. This method is called the Adjusted Differential Box-Counting (ADBC), whichprovides fractal dimensions that are better representations of cloud top heights in thesatellite images.
King Mongkut's University Technology Thonburi. Library