Assessing Land Cover Classification Accuracy: Variations in Dataset Combinations and Deep Learning Models

被引:0
作者
Sim, Woo-Dam [1 ]
Yim, Jong-Su [2 ]
Lee, Jung-Soo [1 ]
机构
[1] Kangwon Natl Univ, Coll Forest & Environm Sci, Dept Forest Management, Div Forest Sci, Chunchon 24341, South Korea
[2] Natl Inst Forest Sci, Forest ICT Res Ctr, Seoul 02455, South Korea
关键词
land cover; deep learning; remote sensing; U-Net; gray-level co-occurrence matrix;
D O I
10.3390/rs16142623
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 evaluates land cover classification accuracy through adjustments to the deep learning model (DLM) training process, including variations in loss function, the learning rate scheduler, and the optimizer, along with diverse input dataset compositions. DLM datasets were created by integrating surface reflectance (SR) spectral data from satellite imagery with textural information derived from the gray-level co-occurrence matrix, yielding four distinct datasets. The U-Net model served as the baseline, with models A and B configured by adjusting the training parameters. Eight land cover classifications were generated from four datasets and two deep learning training conditions. Model B, utilizing a dataset comprising spectral, textural, and terrain information, achieved the highest overall accuracy of 90.3% and a kappa coefficient of 0.78. Comparing different dataset compositions, incorporating textural and terrain data alongside SR from satellite imagery significantly enhanced classification accuracy. Furthermore, using a combination of multiple loss functions or dynamically adjusting the learning rate effectively mitigated overfitting issues, enhancing land cover classification accuracy compared to using a single loss function.
引用
收藏
页数:16
相关论文
共 59 条
  • [1] [Anonymous], Kyoto Protocol Kyoto Protocol
  • [2] [Anonymous], Infrastructure, Transport and Tourism Green Infrastructure PortalSite.
  • [3] [Anonymous], 2006, INTERGOVERNMENTAL PA
  • [4] Enhancing Land Cover/Land Use (LCLU) classification through a comparative analysis of hyperparameters optimization approaches for deep neural network (DNN)
    Azedou, Ali
    Amine, Aouatif
    Kisekka, Isaya
    Lahssini, Said
    Bouziani, Youness
    Moukrim, Said
    [J]. ECOLOGICAL INFORMATICS, 2023, 78
  • [5] The Performance Improvement of U-Net Model for Landcover Semantic Segmentation through Data Augmentation
    Baek, Won-Kyung
    Lee, Moung-Jin
    Jung, Hyung-Sup
    [J]. KOREAN JOURNAL OF REMOTE SENSING, 2022, -38 (06) : 1663 - 1676
  • [6] Chen J., 2021, arXiv, DOI [DOI 10.48550/ARXIV.2102.04306, 10.48550/arXiv.2102.04306]
  • [7] Cho KYHY, 2014, Arxiv, DOI [arXiv:1406.1078, DOI 10.48550/ARXIV.1406.1078]
  • [8] chuncheon, Chuncheon-si Introduce Chuncheon
  • [9] An analysis of co-occurrence texture statistics as a function of grey level quantization
    Clausi, DA
    [J]. CANADIAN JOURNAL OF REMOTE SENSING, 2002, 28 (01) : 45 - 62
  • [10] EunBeen Park, 2018, [Journal of Climate Change Research, 한국기후변화학회지], V9, P385, DOI 10.15531/KSCCR.2018.9.4.385