Fire Type Detection and Online Quickly Analysis Based on Multi-Sources Fire Data

With the global temperature rising, fires are becoming profound impact on Earth system. Accurate fire type identification based on remote sensing technology is becoming great significance for disaster prevention and reduction. In this study, we built a system framework to quickly detect fire type and analyse spatio-temporal distribution online. First, National Polar-orbiting Partnership (NPP) Visible Infrared Imaging Radiometer (VIIRS) Active Fire (ACF), Landsat 8 fire product, Moderate Resolution Imaging Spectroradiometer (MODIS) fire product (MCD14ML), and NPP-VIIRS Nighttime Fire Data (VNF) were obtained and stored into database. Then, industrial and biological fire were distinguished by taking full advantage of the industrial heat source dataset and land use dataset. Finally, quickly analysis can be realized online directly. The result shows that that system can carry out long time series analysis of multi-source fire data on different regions directly.