Combining Sentinel-1 and Sentinel-2 Time Series via RNN for object-based\n land cover classification

2018/12/13 by Dino Ienco, Raffaele Gaetano, Ienco, Dino +7
Engineering · Environmental Science · #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Remote Sensing and LiDAR Applications #Remote Sensing in Agriculture #Remote-Sensing Image Classification

paper · pdf · doi:10.48550/arxiv.1812.05530

openalex publication_date 2018/12/13 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Radar and Optical Satellite Image Time Series (SITS) are sources of\ninformation that are commonly employed to monitor earth surfaces for tasks\nrelated to ecology, agriculture, mobility, land management planning and land\ncover monitoring. Many studies have been conducted using one of the two\nsources, but how to smartly combine the complementary information provided by\nradar and optical SITS is still an open challenge. In this context, we propose\na new neural architecture for the combination of Sentinel-1 (S1) and Sentinel-2\n(S2) imagery at object level, applied to a real-world land cover classification\ntask. Experiments carried out on the Reunion Island, a overseas department of\nFrance in the Indian Ocean, demonstrate the significance of our proposal.\n

Related

• Object-based multi-temporal and multi-source land cover mapping\n leveraging hierarchical class relationships
• DuPLO: A DUal view Point deep Learning architecture for time series classificatiOn
• Deep Neural Networks for automatic extraction of features in time series satellite images
• M3Fusion: A Deep Learning Architecture for Multi-Scale/Modal/Temporal satellite data fusion
• Satellite Image Time Series Classification with Pixel-Set Encoders and\n Temporal Self-Attention