The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XLIII-B2-2021
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLIII-B2-2021, 923–929, 2021
https://doi.org/10.5194/isprs-archives-XLIII-B2-2021-923-2021
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLIII-B2-2021, 923–929, 2021
https://doi.org/10.5194/isprs-archives-XLIII-B2-2021-923-2021

  28 Jun 2021

28 Jun 2021

TOWARDS LIFELONG CROP RECOGNITION USING FULLY CONVOLUTIONAL RECURRENT NETWORKS AND SAR IMAGE SEQUENCES

J. A. Chamorro, R. Q. Feitosa, P. N. Happ, and J. D. Bermudez J. A. Chamorro et al.
  • Dept. of Electrical Engineering, Pontifical Catholic University of Rio de Janeiro (PUC-Rio), Brazil

Keywords: fully convolutional recurrent networks, crop recognition, SAR, remote sensing

Abstract. Recent works have studied crop recognition in regions with highly complex spatio-temporal dynamics typical of a tropical climate. However, most proposals have only been evaluated in a single agricultural year, and their capabilities to generalize to dates outside the temporal sequence have not been properly addressed thus far. This work assesses the generalization capabilities of a recent convolutional recurrent architecture, testing it in a temporal sequence two years ahead of the sequence with which it was trained. Furthermore, a N-to-1 variant of such network is proposed, which is able to produce classification outcomes for every month in the agricultural year, and it is compared with two baselines designed in a more traditional approach, in which a separate specific network is trained for each month of the year. The approaches are evaluated on two public datasets from a tropical region. The first dataset comprehends the period from June 2017 to May 2018, while the second goes from October 2019 to September 2020. Results show a decrease of up to 24.6% in per-date average F1 score when training the network with data of an agricultural year different from the one it is tested on, which indicates a domain shift that demands further research. Additionally, the proposed approach presented only a slight decrease in performance compared to its baseline when trained on the same dataset, with a 2.7% drop in average F1 score. This performance drop is a small cost in exchange for its operational advantages, such as reduced training time and a more straightforward pipeline.