Browse > Article
http://dx.doi.org/10.7780/kjrs.2021.37.3.4

Accuracy analysis of Multi-series Phenological Landcover Classification Using U-Net-based Deep Learning Model - Focusing on the Seoul, Republic of Korea -  

Kim, Joon (Department of Environmental Science and Ecological Engineering, Korea University)
Song, Yongho (Department of Environmental Science and Ecological Engineering, Korea University)
Lee, Woo-Kyun (Division of Environmental Science and Ecological Engineering, Korea University)
Publication Information
Korean Journal of Remote Sensing / v.37, no.3, 2021 , pp. 409-418 More about this Journal
Abstract
The land cover map is a very important data that is used as a basis for decision-making for land policy and environmental policy. The land cover map is mapped using remote sensing data, and the classification results may vary depending on the acquisition time of the data used even for the same area. In this study, to overcome the classification accuracy limit of single-period data, multi-series satellite images were used to learn the difference in the spectral reflectance characteristics of the land surface according to seasons on a U-Net model, one of the deep learning algorithms, to improve classification accuracy. In addition, the degree of improvement in classification accuracy is compared by comparing the accuracy of single-period data. Seoul, which consists of various land covers including 30% of green space and the Han River within the area, was set as the research target and quarterly Sentinel-2 satellite images for 2020 were aquired. The U-Net model was trained using the sub-class land cover map mapped by the Korean Ministry of Environment. As a result of learning and classifying the model into single-period, double-series, triple-series, and quadruple-series through the learned U-Net model, it showed an accuracy of 81%, 82% and 79%, which exceeds the standard for securing land cover classification accuracy of 75%, except for a single-period. Through this, it was confirmed that classification accuracy can be improved through multi-series classification.
Keywords
Classification; Deep Learning; Multi Series; Vegetation Phenology;
Citations & Related Records
Times Cited By KSCI : 4  (Citation Analysis)
연도 인용수 순위
1 Jo, W., Y. Lim, and K.H. Park, 2019. Deep learning based Land Cover Classification Using Convolutional Neural Network-a case study of Korea, Journal of the Korean Geographical Society, 54(1): 1-16 (in Korean with English abstract)
2 Kim, Y. S., K.D. Nam, W.J. Kim, and J.H. Park, 2001. Organism Phenological Study Using Multi-Series AVHRR Satellite Image, Proc. of 2001 the Korean Institute of Landscape Architecture Conference, Ulsan, KR, Oct. 26, pp. 90-92.
3 Kim, Y.S., K.J. Lee, J.W. Ryu, and J.H. Kim, 2003. Landuse Classification Nomenclature for Urban Growth Analysis using Satellite Imagery, Journal of the Korean Association of Geographic Information Studies, 6(3): 83-94 (in Korean with English abstract).
4 Lee, K.S., Y. Yoon, S.H. Kim, J.I. Shin, J.S. Yoon, and S. Kang, 2009. Analysis of Present Status for the Monitoring of land Use and Land Cover in the Korean Peninsula, Korean Journal of Remote Sensing, 25(1): 71-83 (in Korean with English abstract).   DOI
5 Lee, S. and J. Kim, 2019. Land Cover Classification Using Sematic Image Segmentation with Deep Learning, Korean Journal of Remote Sensing, 35(2): 279-288 (in Korean with English abstract).   DOI
6 Moon, G.S., K.S. Kim, and Y.J. Choung, 2020. Land Cover Classification Based on High Resolution KOMPSAT-3 Satellite Imagery Using Deep Neural Network Model, Journal of the Korean Association of Geographic Information Studies, 23(3): 252-262 (in Korean with English abstract).   DOI
7 Jang, S.S., 2018. Classification of Tree Species Using Sentinel-2 Satellite Image, University of Seoul, Seoul, KR.
8 Jeong, J.C., 2002. Particulate Distribution Map of Tidal Flat using Unsupervised Classification of Multi-Temporary Satellite Data, Korean Journal of Remote Sensing, 18(2): 71-80 (in Korean with English abstract).   DOI
9 Kim, H.O. and J.M. Yeom, 2012. A Study on ObjectBased Image Analysis Methods for Land Cover Classification in Agricultural Areas, Journal of the Korean Association of Geographic Information Studies, 15(4): 26-41 (in Korean with English abstract).   DOI
10 Lee, S.-H. and M. Lee, 2020. A Study on Deep Learning Optimization by Land Cover Classification Item Using Satellite Imagery, Korean Journal of Remote Sensing, 36(6-2): 1591-1604 (in Korean with English abstract).   DOI
11 Ronneberger, O., P. Fischer, and T. Brox, 2015. U-Net: Convolutional Networks for Biomedical Image Segmentation, In: Navab N., Hornegger J., Wells W., Frangi A. (eds), Proc. of Medical Image Computing and Computer-Assisted Intervention - MICCAI 2015, Munich, DE, Oct. 5-9, vol. 9351, pp. 234-241.
12 Park, B.U., 1996. Land Cover Classification Techniques for Large Area using Digital Satellite Data. Journal of the Korean Society of Geodesy, Photogrammetry, and Cartography, 14: 39-47 (in Korean with English abstract).
13 Park, H.S. and D.H. Jang, 2020. Analysis of Changes in Urbanized Areas in Daejeon Metropolitan City by Detection of Changes in Time Series Landcover: Using Multi-temporal Satellite Images, Journal of the association of Korean Geographers, 9(1): 177-190 (in Korean with English abstract).   DOI
14 Pan, Z., J. Xu, Y. Guo, Y. Hu, and G. Wang, 2020. Deep learning segmentation and classification for urban village using a worldview satellite image based on U-Net, Remote Sensing, 12(10): 1574 (in Korean with English abstract).   DOI
15 Park, J.W., H.S. Na, and J.S. Yim, 2017. Comparison of Land-use Change Assessment Methods for Greenhouse Gas Inventory in Land Sector, Climate Change Research, 8(4): 329-337 (in Korean with English abstract).   DOI
16 Park, H.L., J.W. Choi, and J.H. Oh, 2018. Seasonal Effects Removal of Unsupervised Change Detection based Multitemporal Imagery, Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography, 36(2): 51-58 (in Korean with English abstract).   DOI
17 Park, S.M., J.H. Im, and H.S. Sakong, 2001. Land Cover Classification of a Wide Area through Multi-Scene Landsat Processing, Korean Journal of Remote Sensing, 17(3): 189-197 (in Korean with English abstract).   DOI
18 Sakong, H.S. and J.H. Im, 2003. An Empirical Study on the Land Cover Classification Method using IKONOS Image, Journal of the Korean Association of Geographic Information Studies, 6(3): 107-116 (in Korean with English abstract).
19 Wang, Y., D. Zhang, and G. Dai, 2020. Classification of high resolution satellite images using improved UNet, International Journal of Applied Mathematics and Computer Science, 30(3): 399-143.   DOI
20 Yoo, S.H., J.S. Lee, J.S. Bae, and H.G. Sohn, 2020. Automatic Generation of Land Cover Map Using Residual U-Net, Journal of The Korean Society of Civil Engineers, 40(5): 535-546 (in Korean with English abstract).   DOI
21 Shin, J.S., T.H. Lee, P.M. Jung, and H.S. Kwon, 2015. A Study on Land Cover Map of UAV Imagery using an Object-based Classification Method, Journal of Korean Society for Geospatial Information System, 23(4): 25-33 (in Korean with English abstract).   DOI
22 Jeon, J.J., Y.I. Kim, and K.Y. Yu, 2002. Land-cover Classification for Multi-source Remotely Sensed Data Using Evidence Theory, Journal of the Korean Society of Civil Engineers, 22(1D): 141-150 (in Korean with English abstract).
23 Wei, S., H. Zhang, C. Wang, Y. Wang, and L. Xu, 2019. Multi-temporal SAR data large-scale crop mapping based on U-Net model, Remote Sensing, 11(1): 68.   DOI
24 Yoo, H.Y., K.D. Lee, S.I. Na, C.W. Park, and N.W. Park, 2017. Field crop classification using multi-temporal high-resolution satellite imagery: A case study on garlic/onion field, Korean Journal of Remote Sensing, 33(5-2): 621-630 (in Korean with English abstract).   DOI
25 Ulmas, P. and I. Liiv, 2020. Segmentation of satellite imagery using u-net models for land cover classification, arXiv preprint arXiv(2003.02899v1): 1-11.
26 Lee, J.-M., Y.H. Cho, Y.H. Kim, and S.W. Park, 2019. The Topsoil Characteristics, and Estimation of Topsoil Organic Carbon Storage at Restoration Areas in Riparian Zones of the Han River, Journal of the Korean Institute of Landscape Architecture, 47(4): 12-23 (in Korean with English abstract).
27 Park, H.L., J.W. Choi, and J.H. Oh, 2018. Seasonal effects removal of unsupervised change detection based multitemporal imagery, Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography, 36(2): 51-58 (in Korean with English abstract).   DOI
28 Rakhlin, A., A. Davydow, and S. Nikolenko, 2018. Land cover classification from satellite imagery with u-net and lovasz-softmax loss, Proc. of 2018 the IEEE Conference on Computer Vision and Pattern Recognition Workshops, Salt Lake City, UT, Jun. 18-22, pp. 262-266.
29 Oh, C.Y., S.Y. Park, H.S. Kim, Y.W. Lee, and C.U. Choi, 2010. Comparison of Landcover Map Accuracy Using High Resolution Satellite Imagery, Journal of the Korean Association of Geographic Information Studies, 13(1): 89-100 (in Korean with English abstract).   DOI
30 Choo, I., Y, Seong, B. Shiksha, and Y. Jung, 2021. Calculation of soil carbon changes by administrative district with regard to land cover changes, Journal of the Korean GEO-environmental Society, 22(3): 37-43 (in Korean with English abstract).   DOI
31 Jeon, S.W., 1999. Landcover classification using satellite imagery, KEI, Sejong, KR.
32 Jeong, S.G., J. Park, C.H. Park, and D. Lee, 2016. Terrace Fields Classification in North Korea Using MODIS Multi-temporal Image Data, Journal of the Korea Society of Environmental Restoration Technology, 19(1): 73-83 (in Korean with English abstract).   DOI