Browse > Article
http://dx.doi.org/10.7780/kjrs.2021.37.6.3.5

Classification of Natural and Artificial Forests from KOMPSAT-3/3A/5 Images Using Deep Neural Network  

Baek, Won-Kyung (Department of Geoinformatics, University of Seoul)
Lee, Yong-Suk (Aerial Mapping Team, Shinhan Aerial Survey)
Park, Sung-Hwan (Marine Disaster Research Center, Korea Institute of Ocean Science & Technology)
Jung, Hyung-Sup (Department of Geoinformatics, University of Seoul)
Publication Information
Korean Journal of Remote Sensing / v.37, no.6_3, 2021 , pp. 1965-1974 More about this Journal
Abstract
Satellite remote sensing approach can be actively used for forest monitoring. Especially, it is much meaningful to utilize Korea multi-purpose satellites, an independently operated satellite in Korea, for forest monitoring of Korea, Recently, several studies have been performed to exploit meaningful information from satellite remote sensed data via machine learning approaches. The forest information produced through machine learning approaches can be used to support the efficiency of traditional forest monitoring methods, such as in-situ survey or qualitative analysis of aerial image. The performance of machine learning approaches is greatly depending on the characteristics of study area and data. Thus, it is very important to survey the best model among the various machine learning models. In this study, the performance of deep neural network to classify artificial or natural forests was analyzed in Samcheok, Korea. As a result, the pixel accuracy was about 0.857. F1 scores for natural and artificial forests were about 0.917 and 0.433 respectively. The F1 score of artificial forest was low. However, we can find that the artificial and natural forest classification performance improvement of about 0.06 and 0.10 in F1 scores, compared to the results from single layered sigmoid artificial neural network. Based on these results, it is necessary to find a more appropriate model for the forest type classification by applying additional models based on a convolutional neural network.
Keywords
Korea Multi-Purpose Satellite (KOMPSAT); Machine Learning; Deep Neural Network; Natural Forest; Artificial Forest;
Citations & Related Records
Times Cited By KSCI : 2  (Citation Analysis)
연도 인용수 순위
1 Kim, J., C.-H. Lim, H.-W. Jo, and W.-K. Lee, 2021. Phenological Classification Using Deep Learning and the Sentinel-2 Satellite to Identify Priority Afforestation Sites in North Korea, Remote Sensing, 13(15): 2946.   DOI
2 Lee, Y.-S., W.-K. Baek, and H.-S. Jung, 2019. Forest vertical Structure classification in Gongju city, Korea from optic and RADAR satellite images using artificial neural network, Korean Journal of Remote Sensing, 35(3): 447-455.   DOI
3 Baek, W.-K. and Jung, H.-S. 2019. A review of change detection techniques using multi-temporal synthetic aperture radar images, Korean Journal of Remote Sensing, 35(5-1): 737-750 (in Korea wth English abstract).   DOI
4 Baek, W.-K., J.-W. Yu, Y.-W. Yoon, H.-S. Jung, and J. Lim, 2021. The Relative Height Error Analysis of Digital Elevation Model on South Korea to Determine the TargetVertical Accuracy of CAS500-4, Korean Journal of Remote Sensing, 37(5-1): 1043-1059 (in Korea with English abstract).   DOI
5 Jang, K., 2021. A Study on the Deep Learning-based Tree Species Classification by using High-resolution Orthophoto Images, Journal of the Korean Association of Geographic Information Studies, 24(3): 1-9 (in Korea with English abstract).   DOI
6 Lee, K.-J., K.-Y. Oh, and W.-J. Lee, 2020a. Current Status of Application of KOMPSAT Series, Korean Journal of Remote Sensing, 36(6-2): 1485-1492 (in Korea with English abstract).   DOI
7 Lee, S., M.-J. Lee, H.-S. Jung, and S. Lee, 2020b. Landslide susceptibility mapping using naive bayes and bayesian network models in Umyeonsan, Korea, Geocarto international, 35(15): 1665-1679.   DOI
8 Korea Forest Service, 2021. 2020 Foresty Statics, Korea Forest Service, Daejeon, KOR.
9 Lee, K., E. Kim, and Y. Kim, 2017a. Orthorectification of KOMPSAT optical images using various ground reference data and accuracy assessment, Journal of Sensors, 2017: 6393278.
10 Kwon, S.-K., K.-M. Kim, and J. Lim, 2021. A Study on Pre-evaluation of Tree Species Classification Possibility of CAS500-4 Using RapidEye Satellite Imageries, Korean Journal of Remote Sensing, 37(2): 291-304 (in Korea with English abstract).   DOI
11 Lee, S., M.-J. Lee, and H.-S. Jung, 2017b. Data mining approaches for landslide susceptibility mapping in Umyeonsan, Seoul, South Korea, Applied Sciences, 7(7): 683.   DOI
12 Lee, W.-J., and C.-W. Lee, 2018. Forest canopy height estimation using multiplatform remote sensing dataset, Journal of Sensors, 2018: 1593129.
13 Lee, Y.-S., S.-H. Park, H.-S. Jung, and W.-K. Baek, 2018. Classification of natural and artificial forests from KOMPSAT-3/3A/5 images using artificial neural network, Korean Journal of Remote Sensing, 34(6-3): 1399-1414 (in Korea with English abstract).   DOI
14 Lim, J., K.-M. Kim, and M.-K. Kim, 2019. The Development of Major Tree Species Classification Model using Different Satellite Images and Machine Learning in Gwangneung Area, Korean Journal of Remote Sensing, 35(6-2): 1037-1052 (in Korea with English abstract).   DOI
15 Masolele, R.N., V. De Sy, M. Herold, D.M. Gonzalez, J, Verbesselt, F. Gieseke, A.G. Mullissa, and C. Martius 2021. Spatial and temporal deep learning methods for deriving land-use following deforestation: A pan-tropical case study using Landsat time series, Remote Sensing of Environment, 264: 112600.   DOI
16 National Geographic information institute, 2020. The National Atlas of Korea II, National Geographic information institute, Ministry of Land Infrastructure and Transport, Suwon, KOR.
17 Oh, K.-Y. and H.-S. Jung, 2012. Automatic geometric calibration of KOMPSAT-2 stereo pair data, Korean Journal of Remote Sensing, 28(2): 191-202 (in Korea with English abstract)   DOI
18 Fan, Q., F. Chen, M. Cheng, S. Lou, R. Xiao, B. Zhang, C. Wang, J. Li, 2019. Ship detection using a fully convolutional network with compact polarimetric sar images, Remote Sensing 11(18): 2171.   DOI
19 Baek, W.-K., H.-S. Jung, and D. Kim, 2020. Oil spill detection of Kerch strait in November 2007 from dual-polarized TerraSAR-X image using artificial and convolutional neural network regression models, Journal of Coastal Research, 102(SI): 137-144.
20 Bjerreskov, K.S., T. Nord-Larsen, and R. Fensholt, 2021. Classification of Nemoral Forests with Fusion of Multi-Temporal Sentinel-1 and 2 Data, Remote Sensing, 13(5): 950.   DOI
21 Kwon, S.-K., H.-S. Jung, W.-K. Baek, and D. Kim, 2017. Classification of forest vertical structure in south Korea from aerial orthophoto and lidar data using an artificial neural network, Applied Sciences, 7(10): 1046.   DOI
22 Yu, J.-W., Y.-W. Yoon, W.-K. Baek, and H.-S. Jung, 2021. Forest Vertical Structure Mapping Using Two-Seasonal Optic Images and LiDAR DSM Acquired from UAV Platform through Random Forest, XGBoost, and Support Vector Machine Approaches, Remote Sensing, 13(21): 4282.   DOI
23 Oh, K.-Y. and H.-S. Jung, 2016. Automated biascompensation approach for pushbroom sensor modeling using digital elevation model, IEEE Transactions on Geoscience and Remote Sensing, 54(6): 3400-3409.   DOI
24 Park, S.-H. and H.-S. Jung, 2019. Band-based best model selection for topographic normalization of normalized difference vegetation index map, IEEE Access, 8: 4408-4417.   DOI
25 Park, S.-H., G.-S. Lee and H.-S. Jung, 2017. Shadow detection and correction method for urban area using KOMPSAT-3 image, Korean Journal of Remote Sensing, 33(6-3): 1197-1213 (in Korea with English abstract).   DOI
26 Lee, S., W.-K. Baek, H.-S. Jung, and S. Lee, 2020c. Susceptibility Mapping on Urban Landslides Using Deep Learning Approaches in Mt. Umyeon, Applied Sciences, 10(22): 8189.   DOI
27 Baek, W.-K. and H.-S. Jung, 2021. Performance Comparison of Oil Spill and Ship Classification from X-Band Dual-and Single-Polarized SAR Image Using Support Vector Machine, Random Forest, and Deep Neural Network, Remote Sensing, 13(16): 3203.   DOI
28 Kwon, S.-K., 2018. Classification of Natural and Artificial Foretes from Sentinel-2 Images Using Artificial Neural Network, University of Seoul, Seoul, KOR.
29 Onishi, M. and T. Ise, 2021. Explainable identification and mapping of trees using UAV RGB image and deep learning, Scientific Reports, 11(1): 1-15.   DOI