Activation Regression for Continuous Domain Generalization with Applications to Crop Classification

• Published in: arXiv.org
• Main Authors: Khanna, Samar, Wallace, Bram, Bala, Kavita, Hariharan, Bharath
• Format: Paper
• Language: English
• Published: Ithaca Cornell University Library, arXiv.org 14.04.2022
• Subjects: Classification; Domains; Landsat satellites; Machine learning; Satellite imagery
• ISSN: 2331-8422

Geographic variance in satellite imagery impacts the ability of machine learning models to generalise to new regions. In this paper, we model geographic generalisation in medium resolution Landsat-8 satellite imagery as a continuous domain adaptation problem, demonstrating how models generalise better with appropriate domain knowledge. We develop a dataset spatially distributed across the entire continental United States, providing macroscopic insight into the effects of geography on crop classification in multi-spectral and temporally distributed satellite imagery. Our method demonstrates improved generalisability from 1) passing geographically correlated climate variables along with the satellite data to a Transformer model and 2) regressing on the model features to reconstruct these domain variables. Combined, we provide a novel perspective on geographic generalisation in satellite imagery and a simple-yet-effective approach to leverage domain knowledge. Code is available at: \url{https://github.com/samar-khanna/cropmap}