In September 2023, British scholar John D. Stamford published a paper titled Development of an accurate low cost NDVI imaging system for assessing plant health in Plant Methods, proposing a dual-camera NDVI (Normalized Difference Vegetation Index) imaging system based on Raspberry Pi, named NDVIpi, which can be used to compare the health status of crops.
doi : 10.1186/s13007-023-00981-8
01
NDVIpi uses the following programming languages
.
-
Python
-
OpenCV Library
-
NumPy
02
Spectral imaging is a key method for high-throughput phenotypic analysis, which can be associated with various biological parameters. NDVI (Normalized Difference Vegetation Index), calculated using specific wavelengths, can be used for comparative analysis of crop health.
Figure 1. Schematic diagram of the imaging system connection
03
Any material with known spectral reflectance can be used for calibration. The calibration board used in the experiment is for radiometric correction, as shown in the following process details from captured images to output NDVI images.
Figure 2. Calibration process of the imaging system
04
To compare NDVIpi with commercial imaging systems, a Micasense RedEdge camera was introduced to collect images simultaneously with NDVIpi.
Figure 3. Example data from the imaging system
05
40 images of green beans, wheat, and barley were selected for analysis. The Raspberry Pi system showed a high correlation with actual NDVI (R²>0.89), proving the high accuracy of the system.
Figure 4. RGB image of green bean plants in the greenhouse
Figure 5. Comparison of NDVI data from multiple systems
This study proposes a Raspberry Pi-based NDVI imaging system that consists of low-cost off-the-shelf components and a calibration method that can accurately measure NDVI values with strong robustness. The development of a small, low-cost, and easy-to-install spectral imaging system is of significant importance in precision agriculture.
doi : 10.1186/s13007-023-00981-8
Recommended Reading
