Exploiting spatial and temporal information for extracting burned areas from time series of spot-VGT data
This paper reports on a new methodology for the detection of burned areas in coarse resolution satellite images. The work has been carried out in the framework of the GBA (global burnt areas) 2000 project whose aim is the mapping of burned areas at a global scale from SPOT-VGT for the year 2000. The proposed methodology is based on the use of Multi-Layer Perceptron (MLP) neural network, which allows exploiting not only the spectral information of the observed targets but also the spatial and temporal relationship of the phenomenon. The study area corresponds to the Northern part of the African continent, entirely covered by a mosaic of daily SPOT-VGT images. Validation of the burned area maps obtained at different dates has been conducted by the comparison with visual classification of low resolution data and with automatic classification of Landsat-ETM+ scenes acquired over the study area in correspondence with the SPOT-VGT.
Bibliographic Reference: An oral report given at: The First International Workshop on the Analysis of Multitemporal Remote Sensing Images. Held in: Trento (IT), 13-14 September 2001
Record Number: 200113950 / Last updated on: 2001-10-24
Original language: en
Available languages: en