A comparative study of PS-InSAR displacement time series clustering techniques for river levee monitoring

Du Han Lee; Do Jin Kim; Ji Sung Kim

doi:10.3741/JKWRA.2026.59.10.1207

All Issue

2026 Vol.59, Issue 10 Preview Page Next Page

A comparative study of PS-InSAR displacement time series clustering techniques for river levee monitoring 하천제방 모니터링을 위한 PS-InSAR 변위 시계열 군집화 기법 비교 연구

31 October 2026. pp. 1207-1221

PDF XML

Abstract

This study applied six unsupervised clustering methods to PS-InSAR-based displacement time-series data to support river levee stability monitoring and comparatively evaluated the applicability of each method. The methods applied were Raw-KMeans, Raw-MBKMeans, PCA-KMeans, Feature-KMeans, UMAP-HDBSCAN, and PCA-GMM. The results were analyzed based on cluster-wise mean LOS displacement time series, spatial distribution, internal evaluation metrics, reproducibility, computational efficiency, and sensitivity to the number of clusters. The results showed that Raw-KMeans most clearly distinguished stepwise displacement patterns, ranging from stable or slightly displaced clusters to clusters with large negative cumulative LOS displacement, indicating its suitability as a reference clustering method. Raw-MBKMeans was able to perform centroid-based preliminary clustering with a short computation time; however, its reproducibility across repeated runs was relatively low, suggesting that it is more appropriate for preliminary analysis. Feature-KMeans effectively separated major displacement patterns using physically and statistically meaningful features and was evaluated as an interpretable method that can complement Raw-KMeans. UMAP-HDBSCAN was useful for identifying noise points and atypical displacement patterns, demonstrating its applicability to the exploration of local anomalous displacement candidates. PCA-KMeans was highly sensitive to the number of retained components and the post-scaling of PCA scores, and its internal evaluation metrics improved substantially when the original variance of the PCA scores was retained. In contrast, PCA-GMM showed limited cluster separation across the tested PCA conditions. These results demonstrate that time-series interpretability, spatial distribution, reproducibility, computational efficiency, and sensitivity to the number of clusters should be jointly considered when selecting clustering methods for PS-InSAR-based river levee displacement data. They also suggest that different clustering methods can be used complementarily depending on the analytical objective, such as reference classification, rapid preliminary analysis, physical interpretation, and exploration of atypical displacement patterns.

Keywords

PS-InSAR

River levee

LOS displacement

Time-series clustering

KMeans

Displacement monitoring

본 연구에서는 하천제방 안정성 모니터링을 지원하기 위해 PS-InSAR 기반 변위 시계열 자료에 여섯 가지 비지도 군집화 기법을 적용하고, 각 기법의 적용성을 비교·평가하였다. 적용한 기법은 Raw-KMeans, Raw-MBKMeans, PCA-KMeans, Feature-KMeans, UMAP-HDBSCAN, PCA-GMM이며, 군집별 평균 LOS 변위 시계열, 공간분포, 내부 평가지표, 재현성, 계산 효율성 및 군집 수 민감도를 기준으로 분석하였다. 분석 결과, Raw-KMeans는 안정 또는 미소 변위군부터 LOS 방향 음의 누적 변위가 큰 군집까지 단계적인 변위 유형을 가장 명확하게 구분하였으며, 기준 군집화 방법으로 적합한 것으로 나타났다. Raw-MBKMeans는 짧은 계산시간으로 중심 기반 예비 군집화를 수행할 수 있었으나, 반복 실행 재현성은 다소 낮아 예비 분석 방법으로 활용하는 것이 적절하였다. Feature-KMeans는 물리적·통계적으로 의미 있는 특징량을 이용하여 주요 변위 유형을 효과적으로 분리하였으며, Raw-KMeans를 보완할 수 있는 해석 방법으로 평가되었다. UMAP-HDBSCAN은 잡음점과 비정형 변위 양상을 식별하는 데 유용하여 국지적 이상 변위 후보군 탐색에 적용 가능성을 보였다. PCA-KMeans는 유지되는 주성분 수와 PCA 점수의 후속 스케일링 조건에 민감하였으며, PCA 점수의 원래 분산을 유지한 경우 내부 평가지표가 크게 개선되었다. 반면 PCA-GMM은 PCA 조건을 변화시킨 이후에도 분리도 측면에서 한계를 보였다. 이러한 결과는 PS-InSAR 기반 하천제방 변위 자료의 군집화 기법을 선택할 때 시계열 해석성, 공간분포, 재현성, 계산 효율성 및 군집 수 민감도를 함께 고려해야 함을 보여준다. 또한 기준 분류, 빠른 예비 분석, 물리적 해석, 비정형 변위 탐색 등 분석 목적에 따라 서로 다른 군집화 기법을 상호보완적으로 활용할 수 있음을 제시한다.

키워드

PS-InSAR

하천제방

LOS 변위

시계열 군집화

KMeans

변위 모니터링

References

Aggarwal, C.C., Hinneburg, A., and Keim, D.A. (2001). “On the surprising behavior of distance metrics in high dimensional space.” Database Theory — ICDT 2001, Lecture Notes in Computer Science, Vol. 1973, Springer, Berlin, Heidelberg, Germany, pp. 420-434. doi: 10.1007/3-540-44503-X_27.

10.1007/3-540-44503-X_27

Bishop, C.M. (2006). Pattern recognition and machine learning. Springer, New York, NY, U.S.

Caliński, T., and Harabasz, J. (1974). “A dendrite method for cluster analysis.” Communications in Statistics, Vol. 3, No. 1, pp. 1-27. doi: 10.1080/03610927408827101.

10.1080/03610927408827101

Campello, R.J.G.B., Moulavi, D., and Sander, J. (2013). “Density-based clustering based on hierarchical density estimates.” Advances in Knowledge Discovery and Data Mining, Lecture Notes in Computer Science, Vol. 7819, Springer, Berlin, Heidelberg, Germany, pp. 160-172. doi: 10.1007/978-3-642-37456-2_14.

10.1007/978-3-642-37456-2_14

Chong, Y., and Zeng, Q. (2024). “Long-term ground deformation monitoring and quantitative interpretation in Shanghai using multi-platform TS-InSAR, PCA, and K-means clustering.” Remote Sensing, Vol. 16, No. 22, Article No. 4188. doi: 10.3390/rs16224188.

10.3390/rs16224188

Crosetto, M., Monserrat, O., Cuevas-González, M., Devanthéry, N., and Crippa, B. (2016). “Persistent Scatterer Interferometry: A review.” ISPRS Journal of Photogrammetry and Remote Sensing, Vol. 115, pp. 78-89. doi: 10.1016/j.isprsjprs.2015.10.011.

10.1016/j.isprsjprs.2015.10.011

Davies, D.L., and Bouldin, D.W. (1979). “A cluster separation measure.” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. PAMI-1, No. 2, pp. 224-227. doi: 10.1109/TPAMI.1979.4766909.

10.1109/TPAMI.1979.4766909

Dempster, A.P., Laird, N.M., and Rubin, D.B. (1977). “Maximum likelihood from incomplete data via the EM algorithm.” Journal of the Royal Statistical Society: Series B (Methodological), Vol. 39, No. 1, pp. 1-38. doi: 10.1111/j.2517-6161.1977.tb01600.x.

10.1111/j.2517-6161.1977.tb01600.x

Ferretti, A., Prati, C., and Rocca, F. (2001). “Permanent scatterers in SAR interferometry.” IEEE Transactions on Geoscience and Remote Sensing, Vol. 39, No. 1, pp. 8-20. doi: 10.1109/36.898661.

10.1109/36.898661

Festa, D., Novellino, A., Hussain, E., Bateson, L., Casagli, N., Confuorto, P., Del Soldato, M., and Raspini, F. (2023). “Unsupervised detection of InSAR time series patterns based on PCA and K-means clustering.” International Journal of Applied Earth Observation and Geoinformation, Vol. 118, 103276. doi: 10.1016/j.jag.2023.103276.

10.1016/j.jag.2023.103276

Gagolewski, M., Bartoszuk, M., and Cena, A. (2021). “Are cluster validity measures (in)valid?” Information Sciences, Vol. 581, pp. 620-636. doi: 10.1016/j.ins.2021.10.004.

10.1016/j.ins.2021.10.004

Hubert, L., and Arabie, P. (1985). “Comparing partitions.” Journal of Classification, Vol. 2, pp. 193-218. doi: 10.1007/BF01908075.

10.1007/BF01908075

Javed, A., Lee, B.S., and Rizzo, D.M. (2020). “A benchmark study on time series clustering.” Machine Learning with Applications, Vol. 1, 100001. doi: 10.1016/j.mlwa.2020.100001.

10.1016/j.mlwa.2020.100001

Jolliffe, I.T. (2002). Principal component analysis. 2nd ed., Springer, New York, NY, U.S. doi: 10.1007/b98835.

10.1007/b98835

Jung, S.J., Lee, D.H., and Kim, J.S. (2025). “Assessment of anomaly detection capability of InSAR time-series techniques: A case study of the Jangcheon embankment failure on the Nakdong River.” Journal of Korea Water Resources Association, Vol. 58, No. 11, pp. 1031-1045. doi: 10.3741/JKWRA.2025.58.11.1031.

10.3741/JKWRA.2025.58.11.1031

Kim, K.B., and Kang, H.D. (2023). “Repeated levee collapses at Sangju Weir: ‘A larger accident may occur unless it is removed due to design defects’.”, accessed 15 October 2026, <http://www.khan.co.kr/article/202307261618001>.

Kwon, H.H. (2024). “Relocation of South Mulgeum IC reduces project cost and construction period.”, accessed 15 October 2026, <http://www.yangsanilbo.com/news/articleView.html?idxno=110384>.

Kwon, H.H. (2025). “Yangsan records the highest number of ground subsidence cases in Gyeongsangnam-do.”, accessed 15 October 2026, <https://www.yangsanilbo.com/news/articleView.html?idxno=118502>.

Lee, D.W. (1997). “A study on the characteristics of alluvial clay in Yangsan-Mulgum.” Journal of the Korean Society of Agricultural Engineers, Vol. 39, No. 1, pp. 102-111.

Liu, Y., Li, Z., Xiong, H., Gao, X., and Wu, J. (2010). “Understanding of internal clustering validation measures.” Proceedings of the 2010 IEEE International Conference on Data Mining, IEEE, Sydney, Australia, pp. 911-916.

10.1109/ICDM.2010.35

MacQueen, J. (1967). “Some methods for classification and analysis of multivariate observations.” Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, University of California Press, Berkeley, CA, U.S., Vol. 1, pp. 281-297.

Masciulli, C., Guiduzzi, G., Tiano, D., Zocchi, M., Guerra, F., Mazzanti, P., and Scarascia Mugnozza, G. (2025). “Interpretable clustering of PS-InSAR time series for ground deformation detection.” Computers & Geosciences, Vol. 203, 105959. doi: 10.1016/j.cageo.2025.105959.

10.1016/j.cageo.2025.105959

McInnes, L., Healy, J., Saul, N., and Großberger, L. (2018). “UMAP: Uniform Manifold Approximation and Projection.” Journal of Open Source Software, Vol. 3, No. 29, 861. doi: 10.21105/joss.00861.

10.21105/joss.00861

Moulavi, D., Jaskowiak, P.A., Campello, R.J.G.B., Zimek, A., and Sander, J. (2014). “Density-based clustering validation.” Proceedings of the 2014 SIAM International Conference on Data Mining, SIAM, Philadelphia, PA, U.S., pp. 839-847. doi: 10.1137/1.9781611973440.96.

10.1137/1.9781611973440.96

Nguyen, D.D.C., Bennett, V., and O’Meara, K. (2019). “New Orleans levee displacement study via interferometric point target analysis technique and in situ instrumentation.” Journal of Geotechnical and Geoenvironmental Engineering, Vol. 145, No. 8, 04019032. doi: 10.1061/(ASCE)GT.1943-5606.0002067.

10.1061/(ASCE)GT.1943-5606.0002067

Osmanoglu, B., Sunar, F., Wdowinski, S., and Cabral-Cano, E. (2016). “Time series analysis of InSAR data: Methods and trends.” ISPRS Journal of Photogrammetry and Remote Sensing, Vol. 115, pp. 90-102. doi: 10.1016/j.isprsjprs.2015.10.003.

10.1016/j.isprsjprs.2015.10.003

Özer, I.E., Rikkert, S.J.H., van Leijen, F.J., Jonkman, S.N., and Hanssen, R.F. (2019). “Sub-seasonal levee deformation observed using satellite radar interferometry to enhance flood protection.” Scientific Reports, Vol. 9, 2646. doi: 10.1038/s41598-019-39474-x.

10.1038/s41598-019-39474-x30804398PMC6389937

Park, C.S., and Ryu, M.Y. (2019). “Characteristics of long-term settlement in the soft ground of Nakdong River by numerical analysis.” Journal of the Korean Geosynthetics Society, Vol. 18, No. 3, pp. 55-67. doi: 10.12814/jkgss.2019.18.3.055.

10.12814/jkgss.2019.18.3.055

Rand, W.M. (1971). “Objective criteria for the evaluation of clustering methods.” Journal of the American Statistical Association, Vol. 66, No. 336, pp. 846-850. doi: 10.1080/01621459.1971.10482356.

10.1080/01621459.1971.10482356

Rousseeuw, P.J. (1987). “Silhouettes: A graphical aid to the interpretation and validation of cluster analysis.” Journal of Computational and Applied Mathematics, Vol. 20, pp. 53-65. doi: 10.1016/0377-0427(87)90125-7.

10.1016/0377-0427(87)90125-7

Rygus, M., Novellino, A., Hussain, E., Syafiudin, F., Andreas, H., and Meisina, C. (2023). “A clustering approach for the analysis of InSAR time series: Application to the Bandung Basin, Indonesia.” Remote Sensing, Vol. 15, No. 15, 3776. doi: 10.3390/rs15153776.

10.3390/rs15153776

Sarmap (2022). SARscape PS tutorial, version 5.6.2, Purasca, Switzerland, accessed 15 October 2026, <https://https://www.sarmap.ch/tutorials/PS_v562.pdf>.

Sculley, D. (2010). “Web-scale k-means clustering.” Proceedings of the 19th International Conference on World Wide Web, ACM, Raleigh, NC, U.S., pp. 1177-1178. doi: 10.1145/1772690.1772862.

10.1145/1772690.1772862

Seidel, M., Marzahn, P., and Ludwig, R. (2016). “Dyke monitoring by the means of persistent scattering interferometry at the coast of northern Germany.” The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol. XLI-B8, pp. 169-173. doi: 10.5194/isprs-archives-XLI-B8-169-2016.

10.5194/isprs-archives-XLI-B8-169-2016

Seidel, M., Marzahn, P., and Ludwig, R. (2019). “Monitoring of a sea-dike in northern Germany by means of ERS-1, ENVISAT/ASAR, and Sentinel-1 SAR interferometry.” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Vol. 12, No. 11, pp. 4351-4360. doi: 10.1109/JSTARS.2019.2949244.

10.1109/JSTARS.2019.2949244

Shi, Y., Hu, C., Zhao, G., and Hou, T. (2023). “The application of PS-InSAR in the monitoring of the settlement of the Yangtze River levee.” Bulletin of Surveying and Mapping, No. 7, pp. 142-148. doi: 10.13474/j.cnki.11-2246.2023.0215.

10.13474/j.cnki.11-2246.2023.0215

Shin, J.Y. (2018). “No groundwater bathhouse in the new town.” accessed 15 October 2026, <https://www.yangsanilbo.com/news/articleView.html?idxno=64709>.

Information

Publisher :KOREA WATER RESOURECES ASSOCIATION
Publisher(Ko) :한국수자원학회
Journal Title :Journal of Korea Water Resources Association
Journal Title(Ko) :한국수자원학회 논문집
Volume : 59
No :10
Pages :1207-1221
Received Date : 2026-06-22
Revised Date : 2026-08-04
Accepted Date : 2026-08-10
DOI :https://doi.org/10.3741/JKWRA.2026.59.10.1207

Journal of Korea Water Resources Association ISSN:2799-8746(Print) 2799-8754(Online) 한국수자원학회 논문집

All Issue