Fine grained classification for multi-source land cover mapping
This work addresses agricultural monitoring needs by providing more accurate land use mapping, though it is incremental as it builds on existing deep learning techniques.
The paper tackled fine-grained land cover classification using multi-source data by proposing a deep learning approach with an attention-enhanced RNN and hierarchical pretraining, achieving significant improvements over standard remote sensing methods on Reunion island.
Nowadays, there is a general agreement on the need to better characterize agricultural monitoring systems in response to the global changes. Timely and accurate land use/land cover mapping can support this vision by providing useful information at fine scale. Here, a deep learning approach is proposed to deal with multi-source land cover mapping at object level. The approach is based on an extension of Recurrent Neural Network enriched via an attention mechanism dedicated to multi-temporal data context. Moreover, a new hierarchical pretraining strategy designed to exploit specific domain knowledge available under hierarchical relationships within land cover classes is introduced. Experiments carried out on the Reunion island - a french overseas department - demonstrate the significance of the proposal compared to remote sensing standard approaches for land cover mapping.