Aboveground biomass reflects the growth of vegetation and the magnitude of carbon storage, and the accuracy of this parameter is crucial for carbon cycle research and climate change mitigation. In this study, a new idea of using deep learning to realize banana canopy detection segmentation and aboveground biomass estimation was proposed. Firstly, the deep learning algorithm YOLOv8s-seg was used as the basic framework improvement, and UAV remote sensing images were applied to realize banana canopy detection segmentation. Then, the canopy coverage area of banana trees was extracted, combined with the measured aboveground biomass data for fitting, and the aboveground biomass estimation model of banana was established by linear regression, K-Nearest Neighbor (KNN), support vector machine, random forest and XGBoost (eXtreme Gradient Boosting). Finally, the model estimation results were compared and analyzed to determine the optimal model. The results showed that the improved YOLOv8s-seg model can quickly and effectively detect and segment banana canopies. Through verification, it was found that the fitting effect and prediction error of the aboveground biomass estimation model based on XGBoost were better than those of other models, with R2 of 0.881 4, root mean square error (RMSE) of 231.37 kg, and mean absolute error (MAE) of 140.47 kg, which could predict the aboveground biomass more accurately and was more suitable for the inversion of the aboveground biomass of bananas, which further verified the feasibility of using UAV and deep learning methods to extract canopy information to estimate the aboveground biomass.
DENGA J, GUOH B, HUJ,et al.Temporal and distribution characteristic of CO2 concentration over China based on GOSAT satellite data[J].National Remote Sensing Bulletin,2020,24(3):319-325.
[3]
郭娜,刘剑秋.植物生物量研究概述(综述)[J].亚热带植物科学,2011,40(2):83-88.
[4]
GUON, LIUJ Q.A review of research on plant biomass[J].Subtropical Plant Science,2011,40(2):83-88.
[5]
MAH Z, MOL D, CROWTHERT W,et al.The global distribution and environmental drivers of aboveground versus belowground plant biomass[J].Nature Ecology & Evolution,2021,5(8):1110-1122.
FUX, ZHANGY X, WANGX J.Prediction of forest biomass carbon pool and carbon sink potential in China before 2060[J].Scientia Silvae Sinicae,2022,58(2):32-41.
ZHANGZ J, LIA N, BIANJ H,et al.Estimating aboveground biomass of grassland in Zoige by visible vegetation index derived from unmanned aerial vehicle image[J].Remote Sensing Technology and Application,2016,31(1):51-62.
[14]
PRAKRITIS, LARRYL, JIYULC,et al.Above-ground biomass estimation in oats using UAV remote sensing and machine learning[J].Sensors,2022,22(2):601.
WUP Q, RENG B, ZHANGC F,et al.Fine identification and biomass estimation of mangroves based on UAV multispectral and LiDAR[J].National Remote Sensing Bulletin,2022,26(6):1169-1181.
YUC Y, GONGH, CAOJ J,et al.Individual tree crown delineation and aboveground biomass estimation of Sonneratia apetala based on unmanned aerial vehicle remote sensing images[J].Tropical Geography,2023,43(1):12-22.
HUANGM Y.Research on automatic error correction of image sensor based on depth learning algorithm[J].Techniques of Automation and Applications,2023,42(5):8-11.
LUQ Y, LIQ J, TONGY K,et al.Instance segmentation method of street trees based on Mask R-CNN[J].Journal of Forestry Engineering,2021,6(5):154-160
[29]
MAR A, JOÃOV, LAMMERTK,et al.Estimation of spinach(Spinacia oleracea) seed yield with 2D UAV data and deep learning[J].Smart Agricultural Technology,2023,3(11):100129.
[30]
BOLYAD, ZHOUC, XIAOF,et al.YOLACT: real-time instance segmentation[C]//2019 IEEE/CVF International Conference on Computer Vision (ICCV).Seoul,Korea (South),2019:9157-9166.
[31]
MUBASHIRUO L.YOLOv5-LiNet: A lightweight network for fruits instance segmentation[J].PloS one,2023,18(3):e0282297.
[32]
YUEX, QIK, NAX,et al.Improved YOLOv8-Seg network for instance segmentation of healthy and diseased tomato plants in the growth stage[J].Agriculture,2023,13(8):1643.
[33]
REIERSENG,DAO D, LÜTJENSB,et al.ReforesTree: A dataset for estimating tropical forest carbon stock with deep learning and aerial imagery[C]//Proceedings of the 36th AAAI Conference on Artificial Intelligence.2022,36(11):12119-12125.
[34]
OUYANGD, HES, ZHANGG,et al.Efficient multi-scale attention module with cross-spatial learning[C]//ICASSP 2023-2023 IEEE International Conference on Acoustics,Speech and Signal Processing (ICASSP).IEEE,2023:1-5.
YANGD F, YANGC J, QIANK D.Estimation of the number of Ginkgo biloba trees and their canopy coverage based on UAV remote sensing[J].Art Science and Technology,2019,32(5):49-50.