Remote Sensing of Particle Absorption Coefficient of Pigments Using a Two-Stage Framework Integrating Optical Classification and Machine Learning

The particle absorption coefficient of pigments (<i>a</i><sub>ph</sub>(λ)), a critical indicator of phytoplankton spectral absorption properties, is essential for bio-optical models and water quality monitoring. To enhance the accuracy of <i>a</i><sub>ph<...

Full description

Saved in:
Bibliographic Details
Main Authors: Xietian Xia, Shaohua Lei, Hui Lu, Zenghui Xu, Xiang Li, Xing Chen, Niancheng Hong, Jie Xu, Kun Shi, Jiacong Huang
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/17/10/1756
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850255951974629376
author Xietian Xia
Shaohua Lei
Hui Lu
Zenghui Xu
Xiang Li
Xing Chen
Niancheng Hong
Jie Xu
Kun Shi
Jiacong Huang
author_facet Xietian Xia
Shaohua Lei
Hui Lu
Zenghui Xu
Xiang Li
Xing Chen
Niancheng Hong
Jie Xu
Kun Shi
Jiacong Huang
author_sort Xietian Xia
collection DOAJ
description The particle absorption coefficient of pigments (<i>a</i><sub>ph</sub>(λ)), a critical indicator of phytoplankton spectral absorption properties, is essential for bio-optical models and water quality monitoring. To enhance the accuracy of <i>a</i><sub>ph</sub>(λ) retrieval in complex aquatic environments, this study proposes a novel two-stage framework integrating optical classification and machine learning regression. Focusing on inland waters—key areas for eutrophication monitoring—we first developed an intelligent clustering method combining Kernel Principal Angle-based Component (KPAC) dimensionality reduction and Chameleon Swarm Algorithm (CSA)-optimized k-medoids to classify water bodies into four optical types based on hyperspectral reflectance features. Subsequently, an XGBoost regression model with L1-norm feature selection was applied to inversely derive <i>a</i><sub>ph</sub>(440), <i>a</i><sub>ph</sub>(555), <i>a</i><sub>ph</sub>(675), and <i>a</i><sub>ph</sub>(709) for each class. Experimental results demonstrated that optical classification significantly improved inversion accuracy: the determination coefficients R<sup>2</sup> all exceeded 0.9 in classified datasets, with RMSE reduced by up to 93.1% compared to unclassified scenarios. This indicates that the strategy based on optical classification and regression inversion can effectively enhance the accuracy of pigment particle absorption coefficient inversions. In summary, this study, with the central objective of accurately measuring the pigment particle absorption coefficient, successfully developed a comprehensive set of optical classification and regression inversion methods applicable to various aquatic environments. This new scientific approach and powerful tool provide a means for monitoring and interpreting the pigment particle absorption characteristics in water bodies using remote sensing technology.
format Article
id doaj-art-65c4fcdc3ccb4af2b3bc93f9fce05fe9
institution OA Journals
issn 2072-4292
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-65c4fcdc3ccb4af2b3bc93f9fce05fe92025-08-20T01:56:45ZengMDPI AGRemote Sensing2072-42922025-05-011710175610.3390/rs17101756Remote Sensing of Particle Absorption Coefficient of Pigments Using a Two-Stage Framework Integrating Optical Classification and Machine LearningXietian Xia0Shaohua Lei1Hui Lu2Zenghui Xu3Xiang Li4Xing Chen5Niancheng Hong6Jie Xu7Kun Shi8Jiacong Huang9China Construction Power and Environment Engineering Co., Ltd., Nanjing 210012, ChinaNational Key Laboratory of Water Disaster Prevention, Nanjing Hydraulic Research Institute, Nanjing 210029, ChinaDepartment of Earth System Science, Ministry of Education Key Laboratory for Earth System Modeling, Institute for Global Change Studies, Tsinghua University, Beijing 100084, ChinaChina Construction Power and Environment Engineering Co., Ltd., Nanjing 210012, ChinaChina Construction Power and Environment Engineering Co., Ltd., Nanjing 210012, ChinaChina Construction Power and Environment Engineering Co., Ltd., Nanjing 210012, ChinaChina Construction Power and Environment Engineering Co., Ltd., Nanjing 210012, ChinaYangtze River Basin Ecological Environment Monitoring and Scientific Research Center, Yangtze River Basin Ecological Environment Supervision and Administration Bureau, Ministry of Ecology and Environment, Wuhan 430010, ChinaKey Laboratory of Lake and Watershed Science for Water Security, Nanjing Institute of Geography and Limnology, Chinese Academy of Sciences, Nanjing 210008, ChinaKey Laboratory of Lake and Watershed Science for Water Security, Nanjing Institute of Geography and Limnology, Chinese Academy of Sciences, Nanjing 210008, ChinaThe particle absorption coefficient of pigments (<i>a</i><sub>ph</sub>(λ)), a critical indicator of phytoplankton spectral absorption properties, is essential for bio-optical models and water quality monitoring. To enhance the accuracy of <i>a</i><sub>ph</sub>(λ) retrieval in complex aquatic environments, this study proposes a novel two-stage framework integrating optical classification and machine learning regression. Focusing on inland waters—key areas for eutrophication monitoring—we first developed an intelligent clustering method combining Kernel Principal Angle-based Component (KPAC) dimensionality reduction and Chameleon Swarm Algorithm (CSA)-optimized k-medoids to classify water bodies into four optical types based on hyperspectral reflectance features. Subsequently, an XGBoost regression model with L1-norm feature selection was applied to inversely derive <i>a</i><sub>ph</sub>(440), <i>a</i><sub>ph</sub>(555), <i>a</i><sub>ph</sub>(675), and <i>a</i><sub>ph</sub>(709) for each class. Experimental results demonstrated that optical classification significantly improved inversion accuracy: the determination coefficients R<sup>2</sup> all exceeded 0.9 in classified datasets, with RMSE reduced by up to 93.1% compared to unclassified scenarios. This indicates that the strategy based on optical classification and regression inversion can effectively enhance the accuracy of pigment particle absorption coefficient inversions. In summary, this study, with the central objective of accurately measuring the pigment particle absorption coefficient, successfully developed a comprehensive set of optical classification and regression inversion methods applicable to various aquatic environments. This new scientific approach and powerful tool provide a means for monitoring and interpreting the pigment particle absorption characteristics in water bodies using remote sensing technology.https://www.mdpi.com/2072-4292/17/10/1756inherent optical propertiesphytoplankton absorptionparticle absorption coefficientwater optical classificationmachine learning regressioninland water monitoring
spellingShingle Xietian Xia
Shaohua Lei
Hui Lu
Zenghui Xu
Xiang Li
Xing Chen
Niancheng Hong
Jie Xu
Kun Shi
Jiacong Huang
Remote Sensing of Particle Absorption Coefficient of Pigments Using a Two-Stage Framework Integrating Optical Classification and Machine Learning
Remote Sensing
inherent optical properties
phytoplankton absorption
particle absorption coefficient
water optical classification
machine learning regression
inland water monitoring
title Remote Sensing of Particle Absorption Coefficient of Pigments Using a Two-Stage Framework Integrating Optical Classification and Machine Learning
title_full Remote Sensing of Particle Absorption Coefficient of Pigments Using a Two-Stage Framework Integrating Optical Classification and Machine Learning
title_fullStr Remote Sensing of Particle Absorption Coefficient of Pigments Using a Two-Stage Framework Integrating Optical Classification and Machine Learning
title_full_unstemmed Remote Sensing of Particle Absorption Coefficient of Pigments Using a Two-Stage Framework Integrating Optical Classification and Machine Learning
title_short Remote Sensing of Particle Absorption Coefficient of Pigments Using a Two-Stage Framework Integrating Optical Classification and Machine Learning
title_sort remote sensing of particle absorption coefficient of pigments using a two stage framework integrating optical classification and machine learning
topic inherent optical properties
phytoplankton absorption
particle absorption coefficient
water optical classification
machine learning regression
inland water monitoring
url https://www.mdpi.com/2072-4292/17/10/1756
work_keys_str_mv AT xietianxia remotesensingofparticleabsorptioncoefficientofpigmentsusingatwostageframeworkintegratingopticalclassificationandmachinelearning
AT shaohualei remotesensingofparticleabsorptioncoefficientofpigmentsusingatwostageframeworkintegratingopticalclassificationandmachinelearning
AT huilu remotesensingofparticleabsorptioncoefficientofpigmentsusingatwostageframeworkintegratingopticalclassificationandmachinelearning
AT zenghuixu remotesensingofparticleabsorptioncoefficientofpigmentsusingatwostageframeworkintegratingopticalclassificationandmachinelearning
AT xiangli remotesensingofparticleabsorptioncoefficientofpigmentsusingatwostageframeworkintegratingopticalclassificationandmachinelearning
AT xingchen remotesensingofparticleabsorptioncoefficientofpigmentsusingatwostageframeworkintegratingopticalclassificationandmachinelearning
AT nianchenghong remotesensingofparticleabsorptioncoefficientofpigmentsusingatwostageframeworkintegratingopticalclassificationandmachinelearning
AT jiexu remotesensingofparticleabsorptioncoefficientofpigmentsusingatwostageframeworkintegratingopticalclassificationandmachinelearning
AT kunshi remotesensingofparticleabsorptioncoefficientofpigmentsusingatwostageframeworkintegratingopticalclassificationandmachinelearning
AT jiaconghuang remotesensingofparticleabsorptioncoefficientofpigmentsusingatwostageframeworkintegratingopticalclassificationandmachinelearning