Application of neural networks to meteorologic signal and image processing

• Grant number: 96/08422-4
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date: February 01, 1997
• End date: April 30, 1997
• Field of knowledge: Physical Sciences and Mathematics - Geosciences - Meteorology
• Principal Investigator: Li Weigang
• Grantee: Li Weigang
• Host Institution: Instituto Nacional de Pesquisas Espaciais (INPE). Ministério da Ciência, Tecnologia e Inovação (Brasil). São José dos Campos , SP, Brazil

Abstract

The objectives are to continue applying neural networks to meteorologic and climatic signal prediction and to apply neural networks to process the images from geostationary meteorologic satellites to improve the estimations of precipitation from satellite images. The idea is to use satellites images of past convective systems and their respective information of precipitation to teach a neural network to predict precipitation from the image of a convective system. (AU)