Validation of a low-cost 2D laser scanner in development of a more-affordable mobile terrestrial proximal sensing system for 3D plant structure phenotyping in indoor environment

- A mobile proximal sensing system was developed by integrating low-cost laser scanners.
- This constructed system proved to be able to derive detailed plant structure features.
- This new system can be used for 3D plant structure phenotyping in indoor environments.
- This more affordable system can help from greenhouse agronomy to broad-sense botany.

Plant phenotyping plays a critical role in grasping plant architectures and understanding plant responses to environment changes. Hence, various plant phenotyping techniques have kept being developed for different scenarios, via introducing diverse remote sensing (RS) techniques such as light detection and ranging (LIDAR). Now, one important trend of this field is developing low-cost 3D systems that are affordable by common users, whereas the often-used LIDAR sensors with high costs cannot satisfy this demand. To handle this issue, this study attempted to develop a low-cost 2D laser scanner based mobile terrestrial proximal sensing system for 3D plant structure phenotyping in indoor environment. Specifically, two RPLIDAR laser scanners, as one kind of the lowest-cost 2D LIDAR sensors in the contemporary market, are installed at the two far-ends of the scanner-fixing frame on the mobile platform, with their scan profiles set in an oblique-crossing way. Then, the movement of the platform, after accurate data georeferencing and calibration, can render the two series of 2D scanning profiles in parallel into a full 3D representation of each row of plants of interest. Based on the resulting 3D point clouds, detailed plant structure features can be derived. Tests showed that the proposed solution has been basically validated, in terms of the specific plant structure variables such as leaf area (R2 = 0.92). Overall, this work has pushed forward the development of LIDAR-based plant phenotyping techniques into the real-sense low-cost stage, and this suggests that more and more practical applications of LIDAR for plant phenotyping may occur in the communities such as plant cultivation and precision agriculture.