Affiliation:
1. MANİSA CELÂL BAYAR ÜNİVERSİTESİ
Abstract
Water is the most essential requirement for sustaining the life cycle on Earth. These resources are constantly dynamic due to anthropogenic and climatological effects. Therefore, management and consistent water policies are necessary to be followed for the proper management of water resources. Monitoring water resources is possible by accurately determining the water surface boundaries and determining the change in water surface areas. In this context, the normalized difference water index (NDWI) and modified normalized difference water index (MNDWI) were computed using JavaScript on the Google Earth Engine through Landsat-9 and Sentinel-2 satellite images. Water pixels were extracted d from other details using the K-means++ cluster algorithm based on the calculated indices. The water surfaces were determined using the Otsu thresholding method, which is the most preferred method for the NDWI and MNDWI indices calculated from the Sentinel images and was used as verification data. The K-means++ clustering algorithm yielded successful results in detecting water surfaces. In the two indices used, the NDWI index was found to be more successful than the MNDWI index. For Landsat-9 images, OA, Kappa, and F1-scores in the NDWI index were calculated as 99.72%, 0.994, and 99.57%, respectively. The OA, Kappa, and F1-scores in the NDWI index for Sentinel-2 images were calculated as 99.39%, 0.986, and 99.04%, respectively. This study demonstrated that clustering algorithms can be successfully applied to automatically detect water surfaces.
Publisher
Bilge International Journal of Science and Technology Research
Reference39 articles.
1. Agarwal, S., Yadav, S., & Singh, K. (2012). Notice of Violation of IEEE Publication Principles: K-means versus k-means++ clustering technique. In 2012 Students Conference on Engineering and Systems, 1–6.
2. Arthur, D., & Vassilvitskii, S. (2007). K-means++: The advantages of careful seeding In: Proceedings of the Eighteenth Annual ACM-SIAM Symposium on Discrete Algorithms. SODA’07, Society for Industrial and Applied Mathematics, 1027–1035, Philadelphia, PA, USA.
3. Bayram, B., Seker, D. Z., Acar, U., Yuksel, Y., Guner, H. A. A., & Cetin, I. (2013). An integrated approach to temporal monitoring of the shoreline and basin of Terkos Lake. Journal of Coastal Research, 29(6), 1427–1435. https://doi.org/10.2112/JCOASTRES-D-12-00084.1
4. Bouslihim, Y., Kharrou, M. H., Miftah, A., Attou, T., Bouchaou, L., & Chehbouni, A. (2022). Comparing Pan-sharpened Landsat-9 and Sentinel-2 for Land-Use Classification Using Machine Learning Classifiers. Journal of Geovisualization and Spatial Analysis, 6(2), 35.
5. Cordeiro, M. C. R., Martinez, J. M., & Peña-Luque, S. (2021). Automatic water detection from multidimensional hierarchical clustering for Sentinel-2 images and a comparison with Level 2A processors. Remote Sensing of Environment, 253(November 2020). https://doi.org/10.1016/j.rse.2020.112209