Research on the Evaluation Model of Fruit Tree Leaf Area Index Based on LiDAR

Abstract

In the process of variable application, acquiring canopy structure parameters of fruit trees is crucial. Traditional manual measurement of these parameters is time-consuming and not conducive to online acquisition. This study presents a method for measuring leaf area index of fruit trees using LiDAR. Six simulated fruit trees with varying leaf area indexes were tested, and real-time collection of canopy data was conducted. The point cloud side view of scanned fruit trees was compared with the contour of simulated fruit trees. When the operation speed was below 1.5 m/s, The point cloud side view of fruit trees closely matched the actual fruit tree contour at less than 1m/s operation speed. It was observed that the point cloud obtained under the same leaf area index was inversely correlated with the operating speed. Mathematical models were developed using Origin to fit the relationships among operation speed, point cloud data, and leaf area index, yielding a coefficient of determination of 0.93. This demonstrates the feasibility of analyzing fruit tree leaf area index using point cloud data. In field tests, the maximum relative error was 12.5% under orchard 0peration speeds (0.5-1 m/s). This method enables real-time measurement of canopy leaf area index, providing technical and model support for variable spray decision-making in orchards.