In the rapidly evolving world of digital agriculture, precision is key. Farmers and agronomists are increasingly turning to advanced technologies to monitor crops at the plant level, enabling detailed assessments of growth, nutrition, and overall condition. A recent study published in *AgriEngineering* introduces a promising workflow that could revolutionize how we gather and utilize multispectral data in agriculture. The research, led by Isabella Subtil Norberto from the Graduate Program in Cartographic Sciences at São Paulo State University (UNESP), focuses on generating high-resolution multispectral point clouds by integrating terrestrial LiDAR and multispectral imagery.
The study addresses a critical gap in current agricultural monitoring technologies. While RGB point clouds can be generated with commercial terrestrial scanners, multi-band multispectral point clouds are rarely obtained directly. Most existing methods are limited to aerial platforms, which restrict close-range monitoring and plant-level studies. Norberto and her team propose a workflow that combines photogrammetric and computer vision techniques to overcome these limitations. “Our goal was to develop an efficient workflow that ensures geometric accuracy and computational efficiency,” Norberto explains. “By integrating terrestrial LiDAR and multispectral imagery, we can generate high-resolution point clouds that provide detailed insights into plant health and growth.”
The workflow involves several innovative steps. Bundle adjustment estimates the camera’s position and orientation relative to the LiDAR reference system. A frustum-based culling algorithm reduces computational cost by selecting only relevant points, and an occlusion removal algorithm assigns spectral attributes only to visible points. The results were impressive, with the generated multispectral point clouds achieving high geometric consistency between overlapping views, demonstrating stable alignment across perspectives.
The implications for the agriculture sector are significant. High-resolution multispectral point clouds can provide farmers with detailed information about crop health, enabling more precise and timely interventions. This technology can be particularly useful for monitoring nutrient deficiencies, detecting diseases, and assessing the overall condition of crops. “This technology has the potential to transform how we approach plant-level analysis,” Norberto says. “By providing detailed, accurate data, we can help farmers make more informed decisions, ultimately leading to higher yields and more sustainable practices.”
The study also highlights the importance of ground control points in ensuring the accuracy of the data. The researchers found that colourisation was most effective when bundle adjustment used an adequate number of well-distributed ground control points. This emphasizes the need for careful planning and execution in the field to ensure the reliability of the data.
While the study demonstrates the potential of this technology, it also acknowledges some limitations. Wind during data acquisition, for example, can affect the accuracy of the point clouds. However, the researchers believe that with further refinement, these challenges can be overcome.
The research published in *AgriEngineering* by lead author Isabella Subtil Norberto from the Graduate Program in Cartographic Sciences at São Paulo State University (UNESP) represents a significant step forward in the field of digital agriculture. As the technology continues to evolve, it is likely that we will see even more sophisticated applications of multispectral point clouds in agriculture, helping farmers to optimize their operations and achieve better outcomes.
This study not only advances our understanding of how to integrate terrestrial LiDAR and multispectral imagery but also opens up new possibilities for data fusion in agriculture. As the agriculture sector continues to embrace digital technologies, the insights gained from this research could pave the way for more precise, efficient, and sustainable farming practices.