Pixel to practice: multi-scale image data for calibrating remote-sensing-based winter wheat monitoring methods

Area of Science:

• Agricultural Science
• Remote Sensing
• Geospatial Data Analysis

Background:

• Site-specific crop management enhances agricultural productivity and environmental sustainability.
• Accurate spatial data on crop growth and health is essential but often lacking for large-scale implementation.
• Interpreting remote sensing signals in commercial crops is challenging due to canopy variability.

Purpose of the Study:

• To present a comprehensive, multi-scale image dataset for improving remote sensing applications in agriculture.
• To enable detailed analysis of field characteristics and crop status across the growing season.
• To address challenges in remote sensing-based trait estimation and precision agriculture.

Main Methods:

• Collected 35,000 high-resolution aerial RGB images, ground-based imagery, and Sentinel-2 satellite data from nine Swiss wheat fields.
• Generated geo-referenced orthomosaics, digital elevation models, and shapefiles.
• Integrated rich metadata including crop husbandry, phenology, and yield maps.

Main Results:

• A comprehensive dataset enabling multi-scale analysis of crop growth and health.
• Facilitated calibration and validation of remote sensing signal interpretation methods.
• Provided resources for addressing key challenges in precision agriculture.

Conclusions:

• The presented dataset is a valuable resource for advancing remote sensing in agriculture.
• Enables more accurate trait estimation and supports site-specific crop management strategies.
• Facilitates research at the intersection of remote sensing, agronomy, and data science.