Combining 2D image and point cloud deep learning to predict wheat above ground biomass

被引:0
|
作者
Zhu, Shaolong [1 ,2 ]
Zhang, Weijun [1 ,2 ]
Yang, Tianle [1 ,2 ]
Wu, Fei [3 ]
Jiang, Yihan [1 ,2 ]
Yang, Guanshuo [1 ,2 ]
Zain, Muhammad [1 ,2 ]
Zhao, Yuanyuan [1 ,2 ]
Yao, Zhaosheng [1 ,2 ]
Liu, Tao [1 ,2 ]
Sun, Chengming [1 ,2 ]
机构
[1] Yangzhou Univ, Coll Agr, Key Lab Crop Genet & Physiol Jiangsu Prov, Key Lab Crop Cultivat & Physiol Jiangsu Prov, Yangzhou 225009, Peoples R China
[2] Yangzhou Univ, Jiangsu Coinnovat Ctr Modern Prod Technol Grain Cr, Yangzhou 225009, Peoples R China
[3] Tech Univ Munich, Sch Life Sci, Precis Agr Lab, D-85354 Freising Weihenstephan, Germany
基金
中国国家自然科学基金;
关键词
Wheat; Biomass prediction; Unmanned aerial vehicle; Point cloud deep learning; Multimodal data fusion; VEGETATION INDEXES; CHLOROPHYLL CONTENT; WINTER-WHEAT; LEAF; RGB; CLASSIFICATION; SEGMENTATION; ALGORITHM; CANOPIES; TEXTURES;
D O I
10.1007/s11119-024-10186-1
中图分类号
S [农业科学];
学科分类号
09 ;
摘要
PurposeThe use of Unmanned aerial vehicle (UAV) data for predicting crop above-ground biomass (AGB) is becoming a more feasible alternative to destructive methods. However, canopy height, vegetation index (VI), and other traditional features can become saturated during the mid to late stages of crop growth, significantly impacting the accuracy of AGB prediction.Methods In 2022 and 2023, UAV multispectral, RGB, and light detection and ranging point cloud data of wheat populations were collected at seven growth stages across two experimental fields. The point cloud depth features were extracted using the improved PointNet++ network, and AGB was predicted by fusion with VI, color index (CI), and texture index (TI) raster image features.ResultsThe findings indicate that when the point cloud depth features were fused, the R2 values predicted from VI, CI, TI, and canopy height model images increased by 0.05, 0.08, 0.06, and 0.07, respectively. For the combination of VI, CI, and TI, R2 increased from 0.86 to a maximum of 0.9, while the root-mean-square error (RMSE) and mean absolute error were 1.80 t ha-1 and 1.36 t ha-1, respectively. Additionally, our findings revealed that the hybrid fusion exhibits the highest accuracy, it demonstrates robust adaptability in predicting AGB across various years, growth stages, crop varieties, nitrogen fertilizer applications, and densities.Conclusion This study effectively addresses the saturation in spectral and chemical information, provides valuable insights for high-precision phenotyping and advanced crop field management, and serves as a reference for studying other crops and phenotypic parameters.
引用
收藏
页码:3139 / 3166
页数:28
相关论文
共 50 条
  • [1] Plant Disease Detection using 2D Point Cloud and Deep Learning
    Subhasri, V. P.
    Grace, R. Kingsy
    2022 IEEE 29TH INTERNATIONAL CONFERENCE ON HIGH PERFORMANCE COMPUTING, DATA AND ANALYTICS WORKSHOP, HIPCW, 2022, : 67 - 67
  • [2] 3D POINT CLOUD SIMULATION FOR ABOVE-GROUND FOREST BIOMASS ESTIMATION
    Song, Qian
    Wang, Yuanyuan
    Zhu, Xiao Xiang
    IGARSS 2023 - 2023 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM, 2023, : 1357 - 1360
  • [3] Deep point cloud regression for above-ground forest biomass estimation from airborne LiDAR
    Oehmcke, Stefan
    Li, Lei
    Trepekli, Katerina
    Revenga, Jaime C.
    Nord-Larsen, Thomas
    Gieseke, Fabian
    Igel, Christian
    REMOTE SENSING OF ENVIRONMENT, 2024, 302
  • [4] Above-ground Biomass Wheat Estimation: Deep Learning with UAV-based RGB Images
    Schreiber, Lincoln Vinicius
    Atkinson Amorim, Joao Gustavo
    Guimaraes, Leticia
    Matos, Debora Motta
    da Costa, Celso Maciel
    Parraga, Adriane
    APPLIED ARTIFICIAL INTELLIGENCE, 2022, 36 (01)
  • [5] A Deep Learning Method for 2D Image Stippling
    Xue, Zhongmin
    Wang, Beibei
    Ma, Lei
    ADVANCES IN COMPUTER GRAPHICS, CGI 2021, 2021, 13002 : 300 - 311
  • [6] A Method for Obtaining 3D Point Cloud Data by Combining 2D Image Segmentation and Depth Information of Pigs
    Wang, Shunli
    Jiang, Honghua
    Qiao, Yongliang
    Jiang, Shuzhen
    ANIMALS, 2023, 13 (15):
  • [7] Point-BLS: 3D Point Cloud Classification Combining Deep Learning and Broad Learning System
    Chen, Yixuan
    Fu, Mengyin
    Shen, Kai
    2022 34TH CHINESE CONTROL AND DECISION CONFERENCE, CCDC, 2022, : 2810 - 2815
  • [8] A novel point cloud registration using 2D image features
    Chien-Chou Lin
    Yen-Chou Tai
    Jhong-Jin Lee
    Yong-Sheng Chen
    EURASIP Journal on Advances in Signal Processing, 2017
  • [9] A novel point cloud registration using 2D image features
    Lin, Chien-Chou
    Tai, Yen-Chou
    Lee, Jhong-Jin
    Chen, Yong-Sheng
    EURASIP JOURNAL ON ADVANCES IN SIGNAL PROCESSING, 2017,
  • [10] Image2Point: 3D Point-Cloud Understanding with 2D Image Pretrained Models
    Xu, Chenfeng
    Yang, Shijia
    Galanti, Tomer
    Wu, Bichen
    Yue, Xiangyu
    Zhai, Bohan
    Zhan, Wei
    Vajda, Peter
    Keutzer, Kurt
    Tomizuka, Masayoshi
    COMPUTER VISION, ECCV 2022, PT XXXVII, 2022, 13697 : 638 - 656