XAI-Guided Enhancement of Vegetation Indices for Crop Mapping

TL;DR

This paper introduces an explainable AI method to select and modify vegetation indices using multispectral satellite data, improving crop classification accuracy by leveraging influential spectral bands.

Contribution

It presents a novel approach combining deep learning and explainability to enhance vegetation indices for better crop mapping with multispectral satellites.

Findings

Models with selected indices perform comparably to all-band models.

Combining two indices can outperform baseline models.

The method effectively identifies influential spectral bands for crop classification.

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.