A Green Fingerprint of Antarctica: Drones, Hyperspectral Imaging, and Machine Learning for Moss and Lichen Classification

被引:5
|
作者
Sandino, Juan [1 ,2 ]
Bollard, Barbara [3 ,4 ]
Doshi, Ashray [3 ]
Randall, Krystal [3 ,4 ]
Barthelemy, Johan [3 ,5 ]
Robinson, Sharon A. [3 ,4 ]
Gonzalez, Felipe [1 ,2 ]
机构
[1] Queensland Univ Technol, Securing Antarct Environm Future, 2 George St, Brisbane, Qld 4000, Australia
[2] Queensland Univ Technol, QUT Ctr Robot, 2 George St, Brisbane, Qld 4000, Australia
[3] Univ Wollongong, Securing Antarcticas Environm Future, Northfields Ave, Wollongong, NSW 2522, Australia
[4] Univ Wollongong, Sch Earth Atmospher & Life Sci, Northfields Ave, Wollongong, NSW 2522, Australia
[5] NVIDIA, Santa Clara, CA 95051 USA
基金
澳大利亚研究理事会;
关键词
Antarctic Specially Protected Area (ASPA); data fusion; environmental monitoring; hyperspectral imaging (HSI); unmanned aerial system (UAS); unmanned aerial vehicle (UAV); CHLOROPHYLL CONCENTRATION; TERRESTRIAL VEGETATION; REFLECTANCE INDEX; UAV; ECOSYSTEMS; HEALTH; COLOR; LEAF; SOIL;
D O I
10.3390/rs15245658
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Mapping Antarctic Specially Protected Areas (ASPAs) remains a critical yet challenging task, especially in extreme environments like Antarctica. Traditional methods are often cumbersome, expensive, and risky, with limited satellite data further hindering accuracy. This study addresses these challenges by developing a workflow that enables precise mapping and monitoring of vegetation in ASPAs. The processing pipeline of this workflow integrates small unmanned aerial vehicles (UAVs)-or drones-to collect hyperspectral and multispectral imagery (HSI and MSI), global navigation satellite system (GNSS) enhanced with real-time kinematics (RTK) to collect ground control points (GCPs), and supervised machine learning classifiers. This workflow was validated in the field by acquiring ground and aerial data at ASPA 135, Windmill Islands, East Antarctica. The data preparation phase involves a data fusion technique to integrate HSI and MSI data, achieving the collection of georeferenced HSI scans with a resolution of up to 0.3 cm/pixel. From these high-resolution HSI scans, a series of novel spectral indices were proposed to enhance the classification accuracy of the model. Model training was achieved using extreme gradient boosting (XGBoost), with four different combinations tested to identify the best fit for the data. The research results indicate the successful detection and mapping of moss and lichens, with an average accuracy of 95%. Optimised XGBoost models, particularly Model 3 and Model 4, demonstrate the applicability of the custom spectral indices to achieve high accuracy with reduced computing power requirements. The integration of these technologies results in significantly more accurate mapping compared to conventional methods. This workflow serves as a foundational step towards more extensive remote sensing applications in Antarctic and ASPA vegetation mapping, as well as in monitoring the impact of climate change on the Antarctic ecosystem.
引用
收藏
页数:26
相关论文
共 50 条
  • [21] EVALUTION OF MACHINE LEARNING METHODS FOR HYPERSPECTRAL IMAGE CLASSIFICATION
    Kumar, M. Suresh
    Keerthi, V.
    Anjnai, R. N.
    Sarma, M. Manju
    Bothale, Vinod
    2020 IEEE INDIA GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (INGARSS), 2020, : 225 - 228
  • [22] Evaluating the Accuracy of Machine Learning Classification Models for Similar Herbal Medicine Using Hyperspectral Imaging
    Park, So Jin
    Yoon, Hyo In
    Lee, Hyein
    Kim, Min-Chae
    Yang, Jung-Seok
    Jung, Dae-Hyun
    Ahn, Ju Yeon
    Park, Soo Hyun
    JOURNAL OF BIOSYSTEMS ENGINEERING, 2024, 49 (02) : 156 - 166
  • [23] Unified Classification of Bacterial Colonies on Different Agar Media Based on Hyperspectral Imaging and Machine Learning
    Gu, Peng
    Feng, Yao-Ze
    Zhu, Le
    Kong, Li-Qin
    Zhang, Xiu-ling
    Zhang, Sheng
    Li, Shao-Wen
    Jia, Gui-Feng
    MOLECULES, 2020, 25 (08):
  • [24] Supervised Machine Learning Methods and Hyperspectral Imaging Techniques Jointly Applied for Brain Cancer Classification
    Urbanos, Gemma
    Martin, Alberto
    Vazquez, Guillermo
    Villanueva, Marta
    Villa, Manuel
    Jimenez-Roldan, Luis
    Chavarrias, Miguel
    Lagares, Alfonso
    Juarez, Eduardo
    Sanz, Cesar
    SENSORS, 2021, 21 (11)
  • [25] Classification of Citrus Leaf Diseases Using Hyperspectral Reflectance and Fluorescence Imaging and Machine Learning Techniques
    Min, Hyun Jung
    Qin, Jianwei
    Yadav, Pappu Kumar
    Frederick, Quentin
    Burks, Thomas
    Dewdney, Megan
    Baek, Insuck
    Kim, Moon
    HORTICULTURAE, 2024, 10 (11)
  • [26] Combination of hyperspectral imaging and machine learning models for fast characterization and classification of municipal solid waste
    Tao, Junyu
    Gu, Yude
    Hao, Xiaoling
    Liang, Rui
    Wang, Biyu
    Cheng, Zhanjun
    Yan, Beibei
    Chen, Guanyi
    RESOURCES CONSERVATION AND RECYCLING, 2023, 188
  • [27] Classification of Fresh and Frozen-Thawed Beef Using a Hyperspectral Imaging Sensor and Machine Learning
    Park, Seongmin
    Hong, Suk-Ju
    Kim, Sungjay
    Ryu, Jiwon
    Roh, Seungwoo
    Kim, Ghiseok
    AGRICULTURE-BASEL, 2023, 13 (04):
  • [28] On-The-Go Hyperspectral Imaging Under Field Conditions and Machine Learning for the Classification of Grapevine Varieties
    Gutierrez, Salvador
    Fernandez-Novales, Juan
    Diago, Maria P.
    Tardaguila, Javier
    FRONTIERS IN PLANT SCIENCE, 2018, 9
  • [29] Integration of EfficientNetB0 and Machine Learning for Fingerprint Classification
    Alhijaj J.A.
    Khudeyer R.S.
    Informatica (Slovenia), 2023, 47 (05): : 49 - 56
  • [30] Evaluation of Supervised Machine Learning Classification Algorithms for Fingerprint Recognition
    Rojas, Andres
    Dolecek, Gordana Jovanovic
    PROCEEDINGS OF 2021 GLOBAL CONGRESS ON ELECTRICAL ENGINEERING (GC-ELECENG 2021), 2021, : 1 - 4