Pixel-wise Agricultural Image Time Series Classification: Comparisons and a Deformable Prototype-based Approach

TL;DR

This paper compares methods for pixel-wise agricultural image time series classification and introduces a deformable prototype-based approach that improves unsupervised and low-data classification performance.

Contribution

It presents a deformable prototype-based method adding invariance to spectral and temporal shifts, outperforming existing methods in low-data and unsupervised scenarios.

Findings

Deformable prototype method achieves state-of-the-art unsupervised classification accuracy.

The approach performs well with limited labeled data.

It significantly improves results on four recent SITS datasets.

Abstract

Improvements in Earth observation by satellites allow for imagery of ever higher temporal and spatial resolution. Leveraging this data for agricultural monitoring is key for addressing environmental and economic challenges. Current methods for crop segmentation using temporal data either rely on annotated data or are heavily engineered to compensate the lack of supervision. In this paper, we present and compare datasets and methods for both supervised and unsupervised pixel-wise segmentation of satellite image time series (SITS). We also introduce an approach to add invariance to spectral deformations and temporal shifts to classical prototype-based methods such as K-means and Nearest Centroid Classifier (NCC). We study different levels of supervision and show this simple and highly interpretable method achieves the best performance in the low data regime and significantly improves the state of the art for unsupervised classification of agricultural time series on four recent SITS datasets.