Recognition of Urbanized Areas in UAV-Derived Very-High-Resolution Visible-Light Imagery

被引:0
|
作者
Puniach, Edyta [1 ]
Gruszczynski, Wojciech [1 ]
Cwiakala, Pawel [1 ]
Strzabala, Katarzyna [1 ]
Pastucha, Elzbieta [2 ]
机构
[1] AGH Univ Krakow, Fac Geodata Sci Geodesy & Environm Engn, Mickiewicza 30, PL-30059 Krakow, Poland
[2] Univ Southern Denmark, Maersk Mc Kinney Moller Inst, Campusvej 55, DK-5230 Odense, Denmark
关键词
image classification; neural networks; unmanned aerial vehicle (UAV); vegetation indices; visible-light imagery; VEGETATION INDEXES; CLASSIFICATION;
D O I
10.3390/rs16183444
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
This study compared classifiers that differentiate between urbanized and non-urbanized areas based on unmanned aerial vehicle (UAV)-acquired RGB imagery. The tested solutions included numerous vegetation indices (VIs) thresholding and neural networks (NNs). The analysis was conducted for two study areas for which surveys were carried out using different UAVs and cameras. The ground sampling distances for the study areas were 10 mm and 15 mm, respectively. Reference classification was performed manually, obtaining approximately 24 million classified pixels for the first area and approximately 3.8 million for the second. This research study included an analysis of the impact of the season on the threshold values for the tested VIs and the impact of image patch size provided as inputs for the NNs on classification accuracy. The results of the conducted research study indicate a higher classification accuracy using NNs (about 96%) compared with the best of the tested VIs, i.e., Excess Blue (about 87%). Due to the highly imbalanced nature of the used datasets (non-urbanized areas constitute approximately 87% of the total datasets), the Matthews correlation coefficient was also used to assess the correctness of the classification. The analysis based on statistical measures was supplemented with a qualitative assessment of the classification results, which allowed the identification of the most important sources of differences in classification between VIs thresholding and NNs.
引用
收藏
页数:21
相关论文
共 50 条
  • [21] High-resolution imaging of the upper solar chromosphere: First light performance of the very-high-resolution Advanced Ultraviolet Telescope
    Korendyke, CM
    Vourlidas, A
    Cook, JW
    Dere, KP
    Howard, RA
    Morrill, JS
    Moses, JD
    Moulton, NE
    Socker, DG
    SOLAR PHYSICS, 2001, 200 (1-2) : 63 - 73
  • [22] High-resolution Imaging of the Upper Solar Chromosphere: First Light Performance of the Very-high-Resolution Advanced ULtraviolet Telescope
    C.M. Korendyke
    A. Vourlidas
    J.W. Cook
    K.P. Dere
    R.A. Howard
    J.S. Morrill
    J.D. Moses
    N.E. Moulton
    D.G. Socker
    Solar Physics, 2001, 200 : 63 - 73
  • [23] VEHICLE DETECTION FROM VERY-HIGH-RESOLUTION (VHR) AERIAL IMAGERY USING ATTRIBUTE BELIEF PROPAGATION (ABP)
    Wang, Yanli
    Li, Ying
    Zhang, Li
    Huang, Yuchun
    REMOTE SENSING TECHNOLOGIES AND APPLICATIONS IN URBAN ENVIRONMENTS, 2016, 10008
  • [24] Using very-high-resolution satellite imagery and deep learning to detect and count African elephants in heterogeneous landscapes
    Duporge, Isla
    Isupova, Olga
    Reece, Steven
    Macdonald, David W.
    Wang, Tiejun
    REMOTE SENSING IN ECOLOGY AND CONSERVATION, 2021, 7 (03) : 369 - 381
  • [25] An Application of Geographical Random Forests for Population Estimation in Dakar, Senegal using Very-High-Resolution Satellite Imagery
    Georganos, Stefanos
    Grippa, Tais
    Gadiaga, Assane
    Vanhuysse, Sabine
    Kalogirou, Stamatis
    Lennert, Moritz
    Linard, Catherine
    2019 JOINT URBAN REMOTE SENSING EVENT (JURSE), 2019,
  • [26] Transformer and CNN Hybrid Deep Neural Network for Semantic Segmentation of Very-High-Resolution Remote Sensing Imagery
    Zhang, Cheng
    Jiang, Wanshou
    Zhang, Yuan
    Wang, Wei
    Zhao, Qing
    Wang, Chenjie
    IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2022, 60
  • [27] Self-Supervised Variational Autoencoder for Unsupervised Object Counting from Very-High-Resolution Satellite Imagery: Applications in Dwelling Extraction in FDP Settlement Areas
    Gella, Getachew Workineh
    Gangloff, Hugo
    Wendt, Lorenz
    Tiede, Dirk
    Lang, Stefan
    IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2024, 62 : 1 - 18
  • [28] Accuracy evaluation of convolutional neural network classification algorithms for building identification in rural and urban areas from very-high-resolution satellite imagery in Jambi, Indonesia
    Nugroho, Daniel Adi
    Dimyati, Muhammad
    Laswanto
    BULLETIN OF GEOGRAPHY-SOCIO-ECONOMIC SERIES, 2022, 58 (58) : 141 - 154
  • [29] Vehicle detection in urban areas from very high resolution UAV color images
    Altun, Muslum
    Turker, Mustafa
    PAMUKKALE UNIVERSITY JOURNAL OF ENGINEERING SCIENCES-PAMUKKALE UNIVERSITESI MUHENDISLIK BILIMLERI DERGISI, 2020, 26 (02): : 371 - 384
  • [30] Potential of a virtual reality environment based on very-high-resolution satellite imagery for structural geology measurements of lava flows
    Jaud, Marion
    Geoffroy, Laurent
    Chauvet, Francois
    Durand, Erwan
    Civet, Francois
    JOURNAL OF STRUCTURAL GEOLOGY, 2022, 158