Browsing by Person "He, Miao"

Now showing 1 - 2 of 2

Effect of operational parameters of unmanned aerial vehicle (UAV) on droplet deposition in trellised pear orchard
(2023) Qi, Peng; Zhang, Lanting; Wang, Zhichong; Han, Hu; Müller, Joachim; Li, Tian; Wang, Changling; Huang, Zhan; He, Miao; Liu, Yajia; He, Xiongkui
Background: Unmanned Aerial Vehicles (UAVs) are increasingly being used commercially for crop protection in East Asia as a new type of equipment for pesticide applications, which is receiving more and more attention worldwide. A new model of pear cultivation called the ‘Double Primary Branches Along the Row Flat Type’ standard trellised pear orchards (FT orchard) is widely used in China, Japan, Korea, and other Asian countries because it saves manpower and is suitable for mechanization compared to traditional spindle and open-center cultivation. The disease and pest efficacy of the flat-type trellised canopy structure of this cultivation is a great challenge. Therefore, a UAV spraying trial was conducted in an FT orchard, and a four-factor (SV: Spray application volume rate, FS: Flight speed, FH: Flight height, FD: Flight direction) and 3-level orthogonal test were designed. Results: These data were used to analyze the effect, including spray coverage, deposit density, coefficient of variation, and penetration coefficient on the canopy, to determine the optimal operating parameters of the UAV for pest efficacy in FT orchards. The analysis of extremes of variance showed that factor FD had a significant effect on both spray coverage and deposition density. Followed by factor FS, which had a greater effect on spray coverage (p < 0.05), and factor SV, FH, which had a greater effect on deposition density (p < 0.05). The effects of different factors on spray coverage and deposit density were FD > FS > FH > SV, FD > FH > SV > FS, in that order. The SV3-FS1-FH1-FD3, which flight along the row with an application rate of 90 L/ha, a flight speed of 1.5 m/s, and a flight height of 4.5 m, was the optimal combination, which produced the highest deposit density and spray coverage. It was determined through univariate analysis of all experimental groups, using droplet density of 25/cm2 and spray coverage of 1%, and uniformity of 40% as the measurement index, that T4 and T8 performed the best and could meet the control requirements in different horizontal and vertical directions of the pear canopy. The parameters were as follows: flight along the tree rows, application rate not less than 75 L/ha, flight speed no more than 2 m/s, and flight height not higher than 5 m. Conclusion: This article provides ample data to promote innovation in the use of UAVs for crop protection programs in pergola/vertical trellis system orchards such as FT orchards. At the same time, this project provided a comprehensive analysis of canopy deposition methods and associated recommendations for UAV development and applications.
Effects of different ground segmentation methods on the accuracy of UAV-based canopy volume measurements
(2024) Han, Leng; Wang, Zhichong; He, Miao; He, Xiongkui; Han, Leng; College of Science, China Agricultural University, Beijing, China; Wang, Zhichong; Tropics and Subtropics Group, Institute of Agricultural Engineering, University of Hohenheim, Stuttgart, Germany; He, Miao; College of Science, China Agricultural University, Beijing, China; He, Xiongkui; College of Science, China Agricultural University, Beijing, China
The nonuniform distribution of fruit tree canopies in space poses a challenge for precision management. In recent years, with the development of Structure from Motion (SFM) technology, unmanned aerial vehicle (UAV) remote sensing has been widely used to measure canopy features in orchards to balance efficiency and accuracy. A pipeline of canopy volume measurement based on UAV remote sensing was developed, in which RGB and digital surface model (DSM) orthophotos were constructed from captured RGB images, and then the canopy was segmented using U-Net, OTSU, and RANSAC methods, and the volume was calculated. The accuracy of the segmentation and the canopy volume measurement were compared. The results show that the U-Net trained with RGB and DSM achieves the best accuracy in the segmentation task, with mean intersection of concatenation (MIoU) of 84.75% and mean pixel accuracy (MPA) of 92.58%. However, in the canopy volume estimation task, the U-Net trained with DSM only achieved the best accuracy with Root mean square error (RMSE) of 0.410 m 3 , relative root mean square error (rRMSE) of 6.40%, and mean absolute percentage error (MAPE) of 4.74%. The deep learning-based segmentation method achieved higher accuracy in both the segmentation task and the canopy volume measurement task. For canopy volumes up to 7.50 m 3 , OTSU and RANSAC achieve an RMSE of 0.521 m 3 and 0.580 m 3 , respectively. Therefore, in the case of manually labeled datasets, the use of U-Net to segment the canopy region can achieve higher accuracy of canopy volume measurement. If it is difficult to cover the cost of data labeling, ground segmentation using partitioned OTSU can yield more accurate canopy volumes than RANSAC.