Sugar-Beet Stress Detection using Satellite Image Time Series

TL;DR

This paper presents an unsupervised method using 3D convolutional autoencoders and temporal encodings on satellite image time series to detect stress in sugar-beet fields, enabling practical stress monitoring across different years.

Contribution

It introduces a novel unsupervised approach combining 3D autoencoders and temporal encodings for stress detection in agricultural satellite data.

Findings

Effective separation of stressed and healthy fields achieved

Method generalizes across different years and datasets

Provides a practical tool for farmers and researchers

Abstract

Satellite Image Time Series (SITS) data has proven effective for agricultural tasks due to its rich spectral and temporal nature. In this study, we tackle the task of stress detection in sugar-beet fields using a fully unsupervised approach. We propose a 3D convolutional autoencoder model to extract meaningful features from Sentinel-2 image sequences, combined with acquisition-date-specific temporal encodings to better capture the growth dynamics of sugar-beets. The learned representations are used in a downstream clustering task to separate stressed from healthy fields. The resulting stress detection system can be directly applied to data from different years, offering a practical and accessible tool for stress detection in sugar-beets.