Multi-Source Data Fusion Improves Time-Series Phenotype Accuracy in Maize under a Field High-Throughput Phenotyping Platform

Abstract

The field phenotyping platforms that can obtain high-throughput and time-series phenotypes of plant populations at the 3-dimensional level are crucial for plant breeding and management. However, it is difficult to align the point cloud data and extract accurate phenotypic traits of plant populations. In this study, high-throughput, time-series raw data of field maize populations were collected using a field rail-based phenotyping platform with light detection and ranging (LiDAR) and an RGB (red, green, and blue) camera. The orthorectified images and LiDAR point clouds were aligned via the direct linear transformation algorithm. On this basis, time-series point clouds were further registered by the time-series image guidance. The cloth simulation filter algorithm was then used to remove the ground points. Individual plants and plant organs were segmented from maize population by fast displacement and region growth algorithms. The plant heights of 13 maize cultivars obtained using the multi-source fusion data were highly correlated with the manual measurements ( R 2 = 0.98), and the accuracy was higher than only using one source point cloud data ( R 2 = 0.93). It demonstrates that multi-source data fusion can effectively improve the accuracy of time series phenotype extraction, and rail-based field phenotyping platforms can be a practical tool for plant growth dynamic observation of phenotypes in individual plant and organ scales.

Three-Dimensional Modeling of Maize Canopies Based on Computational Intelligence

“…It optimizes the 3D maize canopy model by iteratively calculating each 3D phytomer within the population, combined with a reflective approach and collision detection and response. Given the high workload and low efficiency of high-precision acquisition of maize canopy 3D data [ 17 ], and the insufficient precision and automation level in maize canopy phenotypic analysis [ 15 ], this method enhances the mechanistic nature of 3D maize canopy model construction to some extent and can present the azimuth angle variation characteristics of maize canopies at different densities. Although there is still a difference between the constructed 3D maize canopy model and the actual field population, preventing a 1:1 3D reconstruction, it can still reflect population characteristics to a certain degree, thus promoting the construction of 3D maize canopy models and FSPMs research.…”

“…Data collection was carried out during the stable phase of maize plant structure and canopy formation, specifically during the R3 stage (milk stage). This involved utilizing the field-based phenotyping platform [ 14 , 15 ] to capture top-view images and 3D point clouds of each plot, aiding in the verification of maize canopy models. Within each plot, a 3 ×3 grid of 9 plants was selected for measurement.…”

“…For example, Li et al. [ 15 ] used a field-based phenotyping platform to obtain time-series data of crop canopies, and through the integration of image and point cloud data, they were able to segment plants and organs within the population and extract phenotypic parameters such as plant height, leaf inclination, and azimuth angles. However, due to the high construction and maintenance costs, limited range of action, and difficulty in practical field production application of the phenotyping platforms, their use in analyzing the phenotypic structure of crop canopies remains limited.…”

Drone-Based Harvest Data Prediction Can Reduce On-Farm Food Loss and Improve Farmer Income

“…Li et al. [ 45 ] located the maize positions on the seeding stage, and these positions were used to guide the difficult segmentation tasks when the maize canopy was closed with severe leaf overlapping. Similarly, for the broccoli tasks, the seedling stage was simpler than that during the flowering stage.…”