To see how well your corn is photosynthesizing, you need to look at the angle of the leaves relative to the stem. Scientists normally have to do it manually with a protractor, but the new robotic system can do the job much faster and easier.
Developed by a team from North Carolina State University and Iowa State University, the AngleNet system combines an existing PhenoBot 3.0 wheeled agricultural robot with special machine learning-based software. The robot is equipped with four of his PhenoStereo camera modules, each consisting of two of his cameras and a set of strobe lights. Modules are placed one above the other with a space between them.
As a remote-controlled robot moves along a row of corn, the camera automatically takes stereoscopic side-view photographs of the leaves of each plant at various heights. The software combines these images to form his 3D model of the leaf, from which the angle of the leaf relative to the stem can be calculated.
Additionally, since the camera module is mounted at a known height, we can determine how high the leaves are above the ground. This is also important information.
“In corn, you want relatively vertical upper leaves, but you want lower leaves on more horizontal stems,” said Asst of North Carolina State University.Professor Lirong Xiang, lead author of the study. . “This allows the plants to collect more sunlight. Researchers focused on plant breeding are monitoring the structure of this type of plant. It informs their work.” because it gives.”
In testing this technique, we found that the leaf angles measured by the AngleNet system were within 5 degrees of the manually measured angles. According to scientists, this amount is within the acceptable margin of error for plant breeding purposes.
“We are already working with several crop scientists to take advantage of this technology, and we hope that more researchers will be interested in adopting this technology to inform their research. “Ultimately, our goal is to promote plant breeding research that improves crop yields.”
A paper on this study was recently published in Journal of Field RoboticsFor another example of a bot inspecting leaves, see the University of Illinois’ Crop Phenotyping Robot.
Source: North Carolina State University
