研究目的
To estimate the leaf area (LA) of maize plants by merging point clouds obtained from different 3-D perspective views using a low-cost time-of-flight camera.
研究成果
The study demonstrated the feasibility of estimating the leaf area of maize plants using a low-cost 3-D time-of-flight camera. The best results were obtained by merging point clouds scanned from the same side of the crop row, with an average mean absolute percentage error of 7.8%.
研究不足
The system relies on a relatively expensive robotic platform and positioning system. Potential environmental difficulties during data acquisition such as dust, rain, direct sunlight, etc., can affect the operation.
1：Experimental Design and Method Selection
The study used a low-cost time-of-flight camera, the Kinect v2, mounted on a robotic platform to acquire 3-D data of maize plants in a greenhouse. Three different maize row reconstruction approaches were compared.
2：Sample Selection and Data Sources
Maize plants in a greenhouse at the University of Hohenheim were used. The seeding was performed in five rows with specific spacing.
3：List of Experimental Equipment and Materials
Kinect v2 camera, robotic platform, SPS930 robotic total station, MT900 Target Prism, Inertial Measurement Unit (IMU) VN-100.
4：Experimental Procedures and Operational Workflow
The robotic platform drove through the maize rows to acquire 3-D images. The images were registered and stitched. Three different approaches for merging point clouds were evaluated.
5：Data Analysis Methods
The point clouds were processed using MATLAB R2016b and CloudCompare. The Poisson surface reconstruction method was applied to estimate the leaf area.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
