plant-stress-detection

1 article

Plant stress early detection through a low-cost multispectral device: Toward safer and more sustainable agricultural practices.

Europe PMC · 2026-06-14

Researchers built a cheap, portable device that uses light sensors and AI to detect plant stress early — before visible damage appears — and can tell apart different levels of severity with 91% accuracy.

The multispectral device achieved an average stress classification accuracy of 91.0 ± 3.1% across different severity levels.

The system automatically selects the most informative spectral features, removing the need to manually pick vegetation indices.

The device uses a simple broadband LED plus a visible-to-near-infrared sensor, keeping costs low enough for small-scale growers and researchers.

Species in Topic

Tobacco

plant-stress-detection

Species in Topic

Related Topics