Phenotyping continues to be a bottleneck in genetic research and crop breeding. Drones or unmanned automatic vehicles (UVAs) are not only great toys for all ages, but using them in the field can greatly quicken measuring thousands of plants.
Using drones for analyses includes many steps such as flight planning, a ground control points survey, georeferencing, image acquisition, calibration of the camera and image processing. UAV surveys usually use nadir photography, which means that the images are shot with the camera axis straight below in a vertical position whilst oblique photography using camera moving along the axis tilted at an angle with respect to the vertical. With both methods, 3D point clouds are created to calculate different measurements.
Che and colleagues from the China Agricultural University compared nadir, oblique and manual measurements of plant height and leaf area index (LAI) of maize plants. The research project found that oblique photography can provide more detailed information about plant architecture. The leading scientist, Dr Yingpu Che also recently co-authored a research project on how deep learning provides a chance to improve the accuracy of detecting maize tassels.
The research team set up field trials in Lishu, Jilin, China of 3,600 maize plants. The researchers measured plant height and leaf area index (LAI) of a representative 1,200 plants at four growth stages from 10 inbred lines with an LI-COR device manually and UAV DJI Inspires 2 drone with a ZENMUSE X5S camera above the plots using an open-source flight planning software. For oblique photography, there were three flight plans with a tilt of camera angles by 45°, 90° and 135°. The resolution of the images was 5280×3956 pixels.
The plant height estimation from the nadir and oblique imagery were highly precise compared to manual measurements. The researchers identified inbred maize lines which were significantly taller or shorter at different growth stages. The point clouds obtained by oblique photography were more complete than those by nadir photography. As mentioned before, oblique imagery is obtained from a tilted camera and the researchers have found that nadir imagery is not reliable for estimating leaf area closer to the ground. Whilst nadir imagery has been more commonly used for plant measurements, this study reveals that clear benefits of using oblique imagery.
Che and colleagues explained, “oblique photography allows the reconstruction of vertical or inclined surfaces of the surveyed canopy” but added, “it should be highlighted that oblique photography needs more time and storage in data acquisition, about 3 times than nadir photography in our study”.
The scientists concluded, “the image analysis technology can help to extract crop phenotypic traits at the whole growth stage automatically and efficiently with a limited number of field measurements by UAV observations. In the future, the phenotype information could be combined with Genome Wide Association Studies to design plant growth at genetic level”.
The United Nation’s Food and Agriculture Organisation reviewed and highlighted how the use of drones in agriculture can help countries reaching Sustainable Development Goals by 2030 and even Goldman Sachs predicts that the agriculture sector will be the second largest user of drones in the world.
This research shows how using drones is not just fun but useful, and that camera positioning can have a large effect.