Journal
REMOTE SENSING
Volume 9, Issue 4, Pages -Publisher
MDPI
DOI: 10.3390/rs9040377
Keywords
precision agriculture; field robotics; LiDAR; high-throughput phenotyping; crop surface model; plant height
Categories
Funding
- Agricultural Sensing and Robotics Initiative of the College of Engineering, University of Georgia
- College of Agricultural and Environmental Sciences of the University of Georgia
- National Robotics Initiative (NIFA) [2017-67021-25928]
Ask authors/readers for more resources
A LiDAR-based high-throughput phenotyping (HTP) system was developed for cotton plant phenotyping in the field. The HTP system consists of a 2D LiDAR and an RTK-GPS mounted on a high clearance tractor. The LiDAR scanned three rows of cotton plots simultaneously from the top and the RTK-GPS was used to provide the spatial coordinates of the point cloud during data collection. Configuration parameters of the system were optimized to ensure the best data quality. A height profile for each plot was extracted from the dense three dimensional point clouds; then the maximum height and height distribution of each plot were derived. In lab tests, single plants were scanned by LiDAR using 0.5 degrees angular resolution and results showed an R-2 value of 1.00 (RMSE = 3.46 mm) in comparison to manual measurements. In field tests using the same angular resolution; the LiDAR-based HTP system achieved average R-2 values of 0.98 (RMSE = 65 mm) for cotton plot height estimation; compared to manual measurements. This HTP system is particularly useful for large field application because it provides highly accurate measurements; and the efficiency is greatly improved compared to similar studies using the side view scan.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available