In Situ Cotton Leaf Area Index by Height Using Three‐Dimensional Point Clouds

Abstract

Three-dimensional (3-D) high-throughput crop phenotyping may benefit plant research and breeding programs by providing a rapid, nondestructive method of determining in-season crop growth and development. In this study, a set of three inexpensive, structured, near-infrared laser projectors mounted on a robotic platform were used to generate high-density color point clouds (PCs) in cotton (Gossypium hirsutum L.) in 2016 and 2017. The PCs were calibrated based on destructive leaf area measurements, and the transformed PCs were analyzed to quantify leaf area by canopy height in 1-cm intervals. We conducted weekly scans of 8-m plots throughout the growing season for two cotton cultivars and two irrigation treatments during the 2 yr. The calibrated PCs were used to accurately measure leaf area index and leaf area index by canopy height. The analysis of leaf area distribution throughout the growing season indicated significant cultivar and irrigation effects by canopy height as well as significant interactions between cultivar and irrigation. Based on the results, it is possible to use a 3-D sensor system to distinguish growth habits that are not easily measured using traditional growth measurements. The information obtained from 3-D measurements may provide additional information about several canopy characteristics that are currently difficult to measure, including radiation capture parameters and biomass partitioning.

Cotton boll distribution and yield estimation using three‐dimensional point cloud data

“…(2019), it was observed that the south‐facing camera measured more leaf area than the north‐facing camera and this was attributed to the south‐facing canopy possibly receiving more sunlight and thus being more productive. Based on these findings, and with leaf production linked to cotton fruit production, it is possible that the south‐facing canopy probably produced more fruit than the north‐facing canopy (Dube et al., 2019). Additionally, it was observed that ST474 exhibited more vegetative growth in the water‐deficit irrigation rate than FM2322, which may have led to the observed yield differences.…”

“…For the side sensors, parts of the top of the plant canopy were hard to capture in some instances. In a companion study on in‐season vegetative growth (Dube et al., 2019) side cameras underestimated leaf area near the top of the plant, due to interference from leaves closer to the sensors. While defoliated cotton is less prone to this error, the boll fraction at the top of the plants may have been somewhat underestimated due to interference of stems and leaves that were not completely defoliated at the time of measurement, as was the case in our 2018 growing season.…”

“…This was also observed in the boll accumulation graphs (Figure 7e, 7g). This effect was, however, minimized in boll analysis, compared to leaf area index analysis (Dube et al., 2019) due to defoliation resulting in little or no leaves that could obscure the bolls located towards the bottom of the canopy. These findings suggest that using the 3D system to measure end‐of‐season boll distribution will be less prohibitive than to measure in‐season leaf area index and other physiological traits.…”

“…Data from the down-facing camera indicated lower boll fraction and accumulation because it occasionally failed to detect the lower parts of the canopy. The failure of the down-facing camera to detect the bottom of the canopy was observed with leaf area detection in a companion study by Dube et al (2019). This was, however, minimized for boll distribution data as the canopy was more than 50% defoliated when the boll data were acquired.…”

“…The south-facing camera data generally correlated better with the measured boll mass than the north-facing camera. In a companion study by Dube et al (2019), it was observed that the south-facing camera measured more leaf area than the north-facing camera and this was attributed to the south-facing canopy possibly receiving more sunlight and thus being more productive. Based on these findings, and with leaf production linked to cotton fruit production, it is possible that the south-facing canopy probably produced more fruit than the north-facing canopy (Dube et al, 2019).…”