ETHZ ASL

Logo

Datasets and open-source code from the Autonomous Systems Lab (ASL) at ETH Zurich, led by Prof. Roland Siegwart.

Eschikon Plant Stress Phenotyping Dataset

Year: 2019
DOI: 10.3929/ethz-c-000788562

Spatio-temporal-spectral dataset of sugar beet crop growth under drought, fertilizer, and weed stress over two months in Eschikon, Switzerland. Includes biweekly RGB images, infrared stereo pairs, multispectral images, treatment parameters, and environmental logs.

Dataset overview

Treatments at a Glance

Multiple box-level treatments with combinations of soil type, water input, nitrogen level, and weed pressure (control, low/med/high N, weed-only variants, drying, mixed stress). See treatment table in dataset for box indices and conditions.

Contents

  • Images: biweekly RGB, infrared stereo, multispectral (organized by date and camera)
  • Spectral point clouds: MATLAB .mat format
  • Calibration images (checkerboards)
  • Greenhouse temperature/humidity logs
  • Box weight measurements (soil moisture proxy)
  • SPAD measurements
  • Above/below ground biomass (yield indicators)

Processing utilities: plant_stress_phenotyping

Downloads

Citation

R. Khanna, L. Schmidt, J. Nieto, R. Siegwart, F. Liebisch,
“A Spatio Temporal Spectral Dataset for Plant Stress Phenotyping”, Plant Methods, 2019.

Tags: agriculture, multispectral, plant-stress, field