Active Fire Detection in Multispectral Super-Resolved Sentinel-2 Images by Means of Sam-Based Approach

In the last years, Sentinel-2 data have become extensively used by the remote sensing community due to their relatively fine spatial resolution, high revisit time ensured by the twin satellites Sentinel- 2 and, of course, their free availability. However, not all the bands are provided at the highest resolution (10 meters). For instance, the Short-Wave Infrared (SWIR) bands, very useful for fires monitoring applications, are provided at 20 meters. Therefore, in order to have a more detailed Active Fire maps, we have proposed a super-resolution data fusion method based on Convolutional Neural Network (CNN), hereafter SRNN+. Then we have compared the standard Active Fire Detection (AFDs) based on indices (AFIs) [1], widely used in literature for active fires monitoring purpose, with a method based on the Spectral Angular Mapper (SAM) [2]. The proposed analysis is validated on the widespread fires that damaged the volcano Vesuvius (Italy) during the summer of 2017.

[1]  Kenneth Grogan,et al.  A Review of the Application of Optical and Radar Remote Sensing Data Fusion to Land Use Mapping and Monitoring , 2016, Remote. Sens..

[2]  Mykel J. Kochenderfer,et al.  Neural Network Guidance for UAVs , 2017 .

[3]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[4]  Davide Cozzolino,et al.  Pansharpening by Convolutional Neural Networks , 2016, Remote. Sens..

[5]  Chenghai Yang,et al.  Using spectral distance, spectral angle and plant abundance derived from hyperspectral imagery to characterize crop yield variation , 2012, Precision Agriculture.

[6]  Wei Guo,et al.  Geospatial Object Detection in High Resolution Satellite Images Based on Multi-Scale Convolutional Neural Network , 2018, Remote. Sens..

[7]  Davide Cozzolino,et al.  Target-Adaptive CNN-Based Pansharpening , 2017, IEEE Transactions on Geoscience and Remote Sensing.

[8]  Antonio Iodice,et al.  A CNN-based Super-resolution Technique for Active Fire Detection on Sentinel-2 Data , 2019, 2019 PhotonIcs & Electromagnetics Research Symposium - Spring (PIERS-Spring).

[9]  Jocelyn Chanussot,et al.  A Critical Comparison Among Pansharpening Algorithms , 2015, IEEE Transactions on Geoscience and Remote Sensing.

[10]  S. Sides,et al.  Comparison of three different methods to merge multiresolution and multispectral data: Landsat TM and SPOT panchromatic , 1991 .

[11]  J. Boardman,et al.  Discrimination among semi-arid landscape endmembers using the Spectral Angle Mapper (SAM) algorithm , 1992 .

[12]  Lucien Wald,et al.  Data Fusion. Definitions and Architectures - Fusion of Images of Different Spatial Resolutions , 2002 .

[13]  A. Hegde,et al.  A Review of Quality Metrics for Fused Image , 2015 .

[14]  S. L. Ullo,et al.  Landsat-8 and Sentinel-2 for fire monitoring at a local scale: A case study on Vesuvius , 2018, 2018 IEEE International Conference on Environmental Engineering (EE).

[15]  Giuseppe Scarpa,et al.  A CNN-Based Fusion Method for Super-Resolution of Sentinel-2 Data , 2018, IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium.

[16]  Giovanni Bovio,et al.  Forest fires are changing: let’s change the fire management strategy , 2017 .

[17]  Frédéric Achard,et al.  The Potential of Sentinel Satellites for Burnt Area Mapping and Monitoring in the Congo Basin Forests , 2016, Remote. Sens..