Deep Recurrent Neural Networks for Winter Vegetation Quality Mapping via Multitemporal SAR Sentinel-1

• Published in: IEEE geoscience and remote sensing letters Vol. 15; no. 3; pp. 464 - 468
• Main Authors: Ho Tong Minh, Dinh, Ienco, Dino, Gaetano, Raffaele, Lalande, Nathalie, Ndikumana, Emile, Osman, Faycal, Maurel, Pierre
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01-03-2018; The Institute of Electrical and Electronics Engineers, Inc. (IEEE); IEEE - Institute of Electrical and Electronics Engineers
• Subjects: Cloud cover; Computer Science; Data; Environmental Sciences; Gated recurrent unit (GRU); Image quality; Learning algorithms; Logic gates; long short term memory (LSTM); Machine Learning; Mapping; Mathematical model; multitemporal; Neural and Evolutionary Computing; Neural networks; Neurons; Radar; Radar data; Radar imaging; recurrent neural network (RNN); Recurrent neural networks; Remote sensing; SAR (radar); Sentinel-1; Signal and Image Processing; Support vector machines; Synthetic aperture radar; synthetic aperture radar (SAR); Time series analysis; Vegetation; Vegetation mapping; vegetation quality; Winter; Multitemporal; Vegetation quality; Recurrent neural network; Gated recurrent units; Sentinel-1; Synthetic aperture radar; Long short term memory
• ISSN: 1545-598X; 1558-0571
• DOI: 10.1109/LGRS.2018.2794581

Mapping winter vegetation quality is a challenging problem in remote sensing. This is due to cloud coverage in winter periods, leading to a more intensive use of radar rather than optical images. The aim of this letter is to provide a better understanding of the capabilities of Sentinel-1 radar images for winter vegetation quality mapping through the use of deep learning techniques. Analysis is carried out on a multitemporal Sentinel-1 data over an area around Charentes-Maritimes, France. This data set was processed in order to produce an intensity radar data stack from October 2016 to February 2017. Two deep recurrent neural network (RNN)-based classifiers were employed. Our work revealed that the results of the proposed RNN models clearly outperformed classical machine learning approaches (support vector machine and random forest).