SEMO-MAMO, a 3-phase module to compare tropical cyclone satellite images using a modified Hausdorff distance

Tropical cyclones (TC) are natural geoscientific phenomena which affect the daily lives of people around the world. Traditionally, the identification of TCs has been highly dependent on human subjective justification on vast supply of information, such as Dvorak templates. In this paper, we present an efficient three-phase prototype SEMO-MAMO. This uses a modified Hausdorff distance based on the extracted contour edges with predefined weighted points to compare the TC satellite images. A new formulation of the Hausdorff distance is designed for significance-based points matching. The experimental results have shown that the proposed prototype improves the computational speed and matching accuracy. It provides another efficient way to use the Hausdorff distance measure for tropical cyclone satellite images matching and prediction.