Recent development of unmanned aerial systems has created opportunities in automation of field-based high-throughput phenotyping by lowering flight operational cost and complexity and allowing flexible re-visit time and higher image resolution than satellite or manned airborne remote sensing. In this study, flights were conducted over corn and sorghum breeding trials in College Station, Texas, with a fixed-wing unmanned aerial vehicle (UAV) carrying two multispectral cameras and a high-resolution digital camera. The objectives were to establish the workflow and investigate the ability of UAV-based remote sensing for automating data collection of plant traits to develop genetic and physiological models. Most important among these traits were plant height and number of plants which are currently manually collected with high labor costs. Vegetation indices were calculated for each breeding cultivar from mosaicked and radiometrically calibrated multi-band imagery in order to be correlated with ground-measured plant heights, populations and yield across high genetic-diversity breeding cultivars. Growth curves were profiled with the aerial measured time-series height and vegetation index data. The next step of this study will be to investigate the correlations between aerial measurements and ground truth measured manually in field and from lab tests.
Chlorophyll is essential to plant photosynthesis, and chlorophyll content is an important indicator of a plant’s growth status. During the past few decades, various types of spectral indices have been used to estimate chlorophyll content. Here we used a continuous wavelet transform (CWT) to estimate the chlorophyll content of maize leaves in different layers from visible to near-infrared (400 to 1000 nm) spectra. The dataset comprised 186 spectra from three leaf layers of plants under different nitrogen treatments. To identify the most sensitive wavelet features, wavelet power scalograms were generated by the CWT, then linear regression models were established between the wavelet power coefficients and chlorophyll content. Two individual wavelet features in the red-edge region were chosen for estimating the chlorophyll content of middle and lower layer, and all their determination coefficients (R 2 ) were better than the spectral indices. For the whole dataset, the most sensitive wavelet feature (724 nm, scale 4) was located near the red edge position, with better correlation (R 2 =90.50% ) than the best spectral index (R 2 =81.85% ). All the predicted models showed good consistency between the calibration and validation datasets, indicating that the chlorophyll content of different maize leaf layers can be accurately estimated by use of a CWT.
Ground-based hyperspectral imaging has a unique advantage in analyzing the component information of field crop due to
its characteristics of combining image with spectrum. However, how to fully utilize its data advantages need to be
studied specifically. This paper collected the spectral reflectance of corn leaves using the Pushbroom Imaging
Spectrometer (PIS) in different growth stages. Then, the red edge position (REP) were identified through six algorithms:
first derivative reflectance (FDR), polynomial function fitting (POLY), four points inserting (FPI), line extrapolate
method(LEM), inverted gauss (IG), Lagrange interpolation (LAGR); and the correlation between REP and chlorophyll
content was explored on the basis of studying the red edge amplitude changes. The results showed that: 1) The REP
obtained by different algorithms changed between 690 nm and 740 nm in which the amplitude changes of red edge for
the FDR, POLY and LAGR were maximum and varied from 692 nm to 730nm; the amplitude changes of the FPI and
LEM varied from 713 nm to 740nm; while the IG algorithm was the narrowest and varied only between 702 nm and 710
nm. 2) Considering the relationship between REP and chlorophyll concentration under different conditions (i.e. growth
stages, species, fertilization and leaf positions), the FDR and LAGR performed well in maize under different conditions;
the IG was suitable for different growth stages; the FPI had a good effect in distinguishing different varieties; the POLY
was suitable for different fertilization; the LEM had wider changes for red edge amplitude and a significant correlation
with chlorophyll content, but the correlation coefficient was smaller than other algorithms and this phenomenon needed
to be further studied. The above research results provided some references for quantitatively retrieving crop nutrients
using ground-based hyperspectral imaging data.