XAI-Guided Enhancement of Vegetation Indices for Crop Mapping

TL;DR

This work tackles crop mapping with multispectral data by using explainable AI to guide vegetation-index design. It trains an RNN on ten spectral bands, derives SHAP-based attributions to identify informative bands, and replaces raw bands with selected vegetation indices or pairs, retraining models accordingly. The findings show single indices can match the full-band baseline while certain two-index combinations surpass it in several crops, indicating efficient, interpretable inputs can retain or improve accuracy. The approach highlights that the most informative spectral channels align with established vegetation indices and provides a practical pathway to exploit new sensor bands for crop monitoring.

Abstract

Vegetation indices allow to efficiently monitor vegetation growth and agricultural activities. Previous generations of satellites were capturing a limited number of spectral bands, and a few expert-designed vegetation indices were sufficient to harness their potential. New generations of multi- and hyperspectral satellites can however capture additional bands, but are not yet efficiently exploited. In this work, we propose an explainable-AI-based method to select and design suitable vegetation indices. We first train a deep neural network using multispectral satellite data, then extract feature importance to identify the most influential bands. We subsequently select suitable existing vegetation indices or modify them to incorporate the identified bands and retrain our model. We validate our approach on a crop classification task. Our results indicate that models trained on individual indices achieve comparable results to the baseline model trained on all bands, while the combination of two indices surpasses the baseline in certain cases.

Figures (1)

• Figure 1: Global and crop-specific spectral attributions of the model trained on the ten satellite bands.