Evaluating the performance of pixel-based and object-based multidimensional clustering algorithms for automated surface water mapping

Remote sensing observations of surface water are vital for effective water resource management and sustainable development. Unsupervised classification holds promise for automating large-scale surface water detection, and it helps solve the difficult problem of sample collection in supervised classi...

Full description

Saved in:
Bibliographic Details
Main Authors: Bohao Li, Kai Liu, Ming Wang, Yanfang Wang, Linmei Zhuang, Weihua Zhu, Chenxia Li, Linhao Zhang, Yanan Chen
Format: Article
Language:English
Published: Taylor & Francis Group 2025-07-01
Series:Geo-spatial Information Science
Subjects:
Online Access:https://www.tandfonline.com/doi/10.1080/10095020.2025.2523993
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849423523843407872
author Bohao Li
Kai Liu
Ming Wang
Yanfang Wang
Linmei Zhuang
Weihua Zhu
Chenxia Li
Linhao Zhang
Yanan Chen
author_facet Bohao Li
Kai Liu
Ming Wang
Yanfang Wang
Linmei Zhuang
Weihua Zhu
Chenxia Li
Linhao Zhang
Yanan Chen
author_sort Bohao Li
collection DOAJ
description Remote sensing observations of surface water are vital for effective water resource management and sustainable development. Unsupervised classification holds promise for automating large-scale surface water detection, and it helps solve the difficult problem of sample collection in supervised classification. Here, we refined the water identification rule for automated surface water extraction on the basis of multidimensional clustering and a supervised classifier. We subsequently comprehensively investigated the classification performance of k-means, hierarchical clustering, and spectral clustering with 57 different feature combinations (features consisting of two bands, B8 and B12, along with four water indices: the automated water extraction index (AWEI), multiband water index (MBWI), normalized difference water index (NDWI), and modified normalized difference water index (MNDWI)) in eight challenging scenarios in China. These comparative experiments were performed from both pixel-based and object-based perspectives. The results show that pixel-based hierarchical clustering, which uses the optimal feature combination of B8, the NDWI, and the MBWI, is the algorithm with the best overall performance, with kappa coefficients exceeding 0.9 in each scene. The object-based hierarchical clustering using the optimal feature combination B8, B12, MNDWI, and MBWI achieves a kappa coefficient exceeding 0.85 in almost all scenes. This algorithm is suitable for scenes without small water bodies. In addition, the improved surface water identification rule increases the applicability of the automated surface water extraction method, which is based on multidimensional clustering in scenes with snow cover. Finally, we compared the performance of multidimensional clustering algorithms with that of the modified Otsu thresholding method and the multi-index threshold-based algorithm and found that the former has an advantage in automated surface water extraction.
format Article
id doaj-art-60c406bec3284cef90fe664c09bf583e
institution Kabale University
issn 1009-5020
1993-5153
language English
publishDate 2025-07-01
publisher Taylor & Francis Group
record_format Article
series Geo-spatial Information Science
spelling doaj-art-60c406bec3284cef90fe664c09bf583e2025-08-20T03:30:33ZengTaylor & Francis GroupGeo-spatial Information Science1009-50201993-51532025-07-0112110.1080/10095020.2025.2523993Evaluating the performance of pixel-based and object-based multidimensional clustering algorithms for automated surface water mappingBohao Li0Kai Liu1Ming Wang2Yanfang Wang3Linmei Zhuang4Weihua Zhu5Chenxia Li6Linhao Zhang7Yanan Chen8Joint International Research Laboratory of Catastrophe Simulation and Systemic Risk Governance, Beijing Normal University, Zhuhai, ChinaJoint International Research Laboratory of Catastrophe Simulation and Systemic Risk Governance, Beijing Normal University, Zhuhai, ChinaJoint International Research Laboratory of Catastrophe Simulation and Systemic Risk Governance, Beijing Normal University, Zhuhai, ChinaHebei International Joint Research Centre for Remote Sensing of Agricultural Drought Monitoring, Hebei GEO University, Shijiazhuang, ChinaJoint International Research Laboratory of Catastrophe Simulation and Systemic Risk Governance, Beijing Normal University, Zhuhai, ChinaTransport Planning and Research Institute, Ministry of Transport, Beijing, ChinaCollege of Resources, Environment and Tourism, Capital Normal University, Beijing, ChinaFaculty of Geographical Science, Beijing Normal University, Beijing, ChinaHebei Remote Sensing Centre, Hebei Hydrological Engineering Geological Survey Institute, Shijiazhuang, ChinaRemote sensing observations of surface water are vital for effective water resource management and sustainable development. Unsupervised classification holds promise for automating large-scale surface water detection, and it helps solve the difficult problem of sample collection in supervised classification. Here, we refined the water identification rule for automated surface water extraction on the basis of multidimensional clustering and a supervised classifier. We subsequently comprehensively investigated the classification performance of k-means, hierarchical clustering, and spectral clustering with 57 different feature combinations (features consisting of two bands, B8 and B12, along with four water indices: the automated water extraction index (AWEI), multiband water index (MBWI), normalized difference water index (NDWI), and modified normalized difference water index (MNDWI)) in eight challenging scenarios in China. These comparative experiments were performed from both pixel-based and object-based perspectives. The results show that pixel-based hierarchical clustering, which uses the optimal feature combination of B8, the NDWI, and the MBWI, is the algorithm with the best overall performance, with kappa coefficients exceeding 0.9 in each scene. The object-based hierarchical clustering using the optimal feature combination B8, B12, MNDWI, and MBWI achieves a kappa coefficient exceeding 0.85 in almost all scenes. This algorithm is suitable for scenes without small water bodies. In addition, the improved surface water identification rule increases the applicability of the automated surface water extraction method, which is based on multidimensional clustering in scenes with snow cover. Finally, we compared the performance of multidimensional clustering algorithms with that of the modified Otsu thresholding method and the multi-index threshold-based algorithm and found that the former has an advantage in automated surface water extraction.https://www.tandfonline.com/doi/10.1080/10095020.2025.2523993Water mappingSentinel-2unsupervised classificationmachine learningrandom forestobject-based classification
spellingShingle Bohao Li
Kai Liu
Ming Wang
Yanfang Wang
Linmei Zhuang
Weihua Zhu
Chenxia Li
Linhao Zhang
Yanan Chen
Evaluating the performance of pixel-based and object-based multidimensional clustering algorithms for automated surface water mapping
Geo-spatial Information Science
Water mapping
Sentinel-2
unsupervised classification
machine learning
random forest
object-based classification
title Evaluating the performance of pixel-based and object-based multidimensional clustering algorithms for automated surface water mapping
title_full Evaluating the performance of pixel-based and object-based multidimensional clustering algorithms for automated surface water mapping
title_fullStr Evaluating the performance of pixel-based and object-based multidimensional clustering algorithms for automated surface water mapping
title_full_unstemmed Evaluating the performance of pixel-based and object-based multidimensional clustering algorithms for automated surface water mapping
title_short Evaluating the performance of pixel-based and object-based multidimensional clustering algorithms for automated surface water mapping
title_sort evaluating the performance of pixel based and object based multidimensional clustering algorithms for automated surface water mapping
topic Water mapping
Sentinel-2
unsupervised classification
machine learning
random forest
object-based classification
url https://www.tandfonline.com/doi/10.1080/10095020.2025.2523993
work_keys_str_mv AT bohaoli evaluatingtheperformanceofpixelbasedandobjectbasedmultidimensionalclusteringalgorithmsforautomatedsurfacewatermapping
AT kailiu evaluatingtheperformanceofpixelbasedandobjectbasedmultidimensionalclusteringalgorithmsforautomatedsurfacewatermapping
AT mingwang evaluatingtheperformanceofpixelbasedandobjectbasedmultidimensionalclusteringalgorithmsforautomatedsurfacewatermapping
AT yanfangwang evaluatingtheperformanceofpixelbasedandobjectbasedmultidimensionalclusteringalgorithmsforautomatedsurfacewatermapping
AT linmeizhuang evaluatingtheperformanceofpixelbasedandobjectbasedmultidimensionalclusteringalgorithmsforautomatedsurfacewatermapping
AT weihuazhu evaluatingtheperformanceofpixelbasedandobjectbasedmultidimensionalclusteringalgorithmsforautomatedsurfacewatermapping
AT chenxiali evaluatingtheperformanceofpixelbasedandobjectbasedmultidimensionalclusteringalgorithmsforautomatedsurfacewatermapping
AT linhaozhang evaluatingtheperformanceofpixelbasedandobjectbasedmultidimensionalclusteringalgorithmsforautomatedsurfacewatermapping
AT yananchen evaluatingtheperformanceofpixelbasedandobjectbasedmultidimensionalclusteringalgorithmsforautomatedsurfacewatermapping