Towards more efficient agricultural practices via transformer-based crop type classification

TL;DR

Preliminary work is presented showing that it is possible to accurately classify crops from time series derived from Sentinel 1 and 2 satellite imagery in Mexico using a pixel-based binary crop/non-crop time series transformer model and preliminary evidence that meta-learning approaches supplemented with data from similar agro-ecological zones may improve model performance.

Abstract

Machine learning has great potential to increase crop production and resilience to climate change. Accurate maps of where crops are grown are a key input to a number of downstream policy and research applications. In this proposal, we present preliminary work showing that it is possible to accurately classify crops from time series derived from Sentinel 1 and 2 satellite imagery in Mexico using a pixel-based binary crop/non-crop time series transformer model. We also find preliminary evidence that meta-learning approaches supplemented with data from similar agro-ecological zones may improve model performance. Due to these promising results, we propose further development of this method with the goal of accurate multi-class crop classification in Jalisco, Mexico via meta-learning with a dataset comprising similar agro-ecological zones.

Figures (2)

• Figure 1: Locations of ground truth binary training labels are shown in grey. Locations included in the filtered sample of data with similar satellite time series to Mexico are shown in green.
• Figure 2: Actual versus predicted crop classifications. The left plot shows ground truth labels; the right plot shows labels predicted by a transformer model trained on the global dataset.