%0 Journal Article
%4 sid.inpe.br/mtc-m21d/2021/12.27.11.32
%2 sid.inpe.br/mtc-m21d/2021/12.27.11.32.28
%@doi 10.3390/rs13245084
%@issn 2072-4292
%T Deforestation detection with fully convolutional networks in the Amazon forest from Landsat-8 and Sentinel-2 images
%D 2021
%8 Dec.
%9 journal article
%A Torres, Daliana Lobo,
%A Turnes, Javier Noa,
%A Vega, Pedro Juan Soto,
%A Feitosa, Raul Queiroz,
%A Silva, Daniel E.,
%A Marcato Júnior, José,
%A Almeida, Cláudio Aparecido de,
%@affiliation Pontifícia Universidade Católica do Rio de Janeiro (PUC-Rio)
%@affiliation Pontifícia Universidade Católica do Rio de Janeiro (PUC-Rio)
%@affiliation Pontifícia Universidade Católica do Rio de Janeiro (PUC-Rio)
%@affiliation Pontifícia Universidade Católica do Rio de Janeiro (PUC-Rio)
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%@affiliation Universidade Federal do Mato Grosso do Sul (UFMS)
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%@electronicmailaddress daliana0491@aluno.puc-rio.br
%@electronicmailaddress jnoat92@aluno.puc-rio.br
%@electronicmailaddress psoto@ele.puc-rio.br
%@electronicmailaddress raul@ele.puc-rio.br
%@electronicmailaddress daniel.silva@inpe.br
%@electronicmailaddress jose.marcato@ufms.br
%@electronicmailaddress claudio.almeida@inpe.br
%B Remote Sensing
%V 13
%N 24
%P e5084
%K Amazon biome, Change detection, Deep learning, Fully convolutional neural networks, Remote sensing, Semantic segmentation.
%X The availability of remote-sensing multisource data from optical-based satellite sensors has created new opportunities and challenges for forest monitoring in the Amazon Biome. In particular, change-detection analysis has emerged in recent decades to monitor forest-change dynamics, supporting some Brazilian governmental initiatives such as PRODES and DETER projects for biodiversity preservation in threatened areas. In recent years fully convolutional network architectures have witnessed numerous proposals adapted for the change-detection task. This paper comprehensively explores state-of-the-art fully convolutional networks such as U-Net, ResU-Net, SegNet, FC-DenseNet, and two DeepLabv3+ variants on monitoring deforestation in the Brazilian Amazon. The networks performance is evaluated experimentally in terms of Precision, Recall, F1-score, and computational load using satellite images with different spatial and spectral resolution: Landsat-8 and Sentinel-2. We also include the results of an unprecedented auditing process performed by senior specialists to visually evaluate each deforestation polygon derived from the network with the highest accuracy results for both satellites. This assessment allowed estimation of the accuracy of these networks simulating a process in nature and faithful to the PRODES methodology. We conclude that the high resolution of Sentinel-2 images improves the segmentation of deforestation polygons both quantitatively (in terms of F1-score) and qualitatively. Moreover, the study also points to the potential of the operational use of Deep Learning (DL) mapping as products to be consumed in PRODES.
%@language en
%3 remotesensing-13-05084-v2.pdf
%O Prêmio CAPES Elsevier 2023 - ODS 15: Vida terrestre