When: Friday, February 25, 2022 1:00 pm EST / 10:00 am PST
Where: United States
Presented by the Ecological Physiology Professional Interest Group, American Society for Horticultural Science
Description Conventional imaging in agriculture involves manual or automatic capturing of images using expensive setups in the field and processing in the lab, which is very time-consuming. As a solution, affordable open-source hardware and software, computer vision algorithms, and crop models can be combined into one system that does everything from capturing images to feature extraction and analysis in real-time. Visible-bands Raspberry Pi camera is easy, cheap to assemble, and widely used in agricultural research. However, there is little information that can be gleaned from RGB images. This led to our current efforts to develop a low-cost, multimodal camera system with the RaspberryPi at its core that combines multispectral and thermal imaging (thermal-VIS-NIR), a light sensor, and a range finder. You can connect the ClimaVUE50 all-in-one microclimate unit (Campbell Scientific) directly to the imager's SDI-12 port. The system has a GUI with embedded computer vision and crop models. The software automatically generates and displays three image layers in real-time (video), and each layer has several options to choose from:
1) Layer 1 (multispectral): R+G+NIR / R / G / NIR (filter dependent)
2) Layer 2 (index): Red NDVI / GNDVI / Normalized Red / Normalized Green / Normalized NIR / Segmented Canopy
3) Layer 3 (thermal + microclimate): ΔT / CWSI / actual Transpiration / Stomatal Conductance
In addition, the software measures plant canopy height and automatically calculates canopy cover, crop coefficient, and crop ET. The imager output is a CSV file with timestamped, geotagged data. This is an ongoing effort and is currently being improved with regard to image segmentation algorithms.
The video below will take you through the slides in just a few min. To watch the whole presentation, please visit the ASHS website: