Related Experiment Video
Updated: Jun 12, 2025
11:37
RGB and Spectral Root Imaging for Plant Phenotyping and Physiological Research: Experimental Setup and Imaging Protocols
Published on: August 8, 2017
16.1K
Spectral indices with different spatial resolutions in recognizing soybean phenology
Airton Andrade da Silva1, Francisco Charles Dos Santos Silva1, Claudinei Martins Guimarães2
1Universidade Estadual do Maranhão, Balsas, Maranhão, Brazil.
Plos One
|September 18, 2024
Summary
Satellite vegetation indices (VI) effectively identify soybean phenological stages, regardless of sensor spatial resolution. Specific indices accurately pinpoint key growth phases, offering a quick and cost-efficient monitoring method for soybean crops.
Area of Science:
- Agricultural Science
- Remote Sensing
- Crop Monitoring
Background:
- Accurate monitoring of crop phenology is crucial for agricultural management and yield prediction.
- Satellite-derived vegetation indices (VI) offer a scalable approach to assess crop status.
- Understanding the influence of spatial resolution on VI performance is key for effective remote sensing applications.
Purpose of the Study:
- To evaluate the efficiency of various vegetation indices (VI) from Sentinel-2 and Amazônia-1 satellites in distinguishing soybean phenological stages.
- To assess the impact of different spatial resolutions on the performance of VI for soybean phenology.
- To determine the suitability of specific VI for identifying critical soybean growth phases.
Main Methods:
- Weekly field assessments of soybean phenology and leaf area index were conducted.
- Spectral data were acquired using Sentinel-2 and Amazônia-1 satellite sensors.
- Various Near-Infrared (NIR) and Red-Green-Blue (RGB) based VI were calculated and analyzed using discriminant analysis and Artificial Neural Networks (ANN).
Main Results:
- The spatial resolution of satellite sensors did not significantly impact the identification accuracy of soybean phenological stages.
- While no single VI achieved 100% accuracy for all stages, specific indices effectively identified key phenological events.
- The best classification results showed an Apparent Error Rate (APER) of zero using both discriminant analysis and ANN, though overall APER varied significantly between methods.
Conclusions:
- Vegetation indices derived from orbital sensors are effective tools for rapid and economical identification of soybean phenological stages.
- Specific VI can reliably detect critical stages such as flowering (R1, R2), pod development (R4), grain development (R5.1), and physiological maturity (R8).
- The findings support the use of satellite remote sensing for efficient soybean crop management and monitoring.
