Hyperspectral Soil Heavy Metal Prediction via Privileged-Informed Residual Correction

This study integrates hyperspectral remote sensing with chemical and pedological data to estimate Zn, Pb, and Cd concentrations in the upper soil layers. Conducted in agricultural fields east and northeast of Celje, Slovenia, an area impacted by past industrial activities such as zinc ore smelting,...

Full description

Saved in:
Bibliographic Details
Main Authors: Alen Mangafić, Krištof Oštir, Mitja Kolar, Marko Zupan
Format: Article
Language:English
Published: MDPI AG 2025-06-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/17/12/1987
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850164642991570944
author Alen Mangafić
Krištof Oštir
Mitja Kolar
Marko Zupan
author_facet Alen Mangafić
Krištof Oštir
Mitja Kolar
Marko Zupan
author_sort Alen Mangafić
collection DOAJ
description This study integrates hyperspectral remote sensing with chemical and pedological data to estimate Zn, Pb, and Cd concentrations in the upper soil layers. Conducted in agricultural fields east and northeast of Celje, Slovenia, an area impacted by past industrial activities such as zinc ore smelting, the research integrates remote sensing and soil sampling to rapidly identify and map soil pollution over large surfaces. A multi-sensor approach was employed, combining two hyperspectral cameras (VNIR and SWIR, aerial), laboratory spectrometry, soil parameters, and content of chemical covariates measured with portable XRF and ICP-OES with a direct comparison of both techniques for this specific purpose. Accurate atmospheric and signal transformations were performed to improve modeling. The importance of covariates was thoroughly evaluated using conditional permutations to assess their contribution to the prediction of metal concentrations. The proposed framework utilizes spectral data and privileged information during training, improving prediction accuracy through a multi-stage model architecture. Here, a base model trained on spectral data is corrected using privileged information. During inference, the model functions without relying on privileged data providing a scalable and cost-effective solution for large-scale environmental monitoring. Our model achieved a reduction of predicted RMSE for Zn and Cd maps in comparison to the baseline models, translating to more precise identification of possibly polluted zones. However, for Pb, no improvements were observed, potentially due to variability in the data, including spectral issues or imbalances in the training and test datasets.
format Article
id doaj-art-d1daed9ef6914513a6c1d2dd758773fe
institution OA Journals
issn 2072-4292
language English
publishDate 2025-06-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-d1daed9ef6914513a6c1d2dd758773fe2025-08-20T02:21:57ZengMDPI AGRemote Sensing2072-42922025-06-011712198710.3390/rs17121987Hyperspectral Soil Heavy Metal Prediction via Privileged-Informed Residual CorrectionAlen Mangafić0Krištof Oštir1Mitja Kolar2Marko Zupan3Geodetic Institute of Slovenia, Jamova cesta 2, 1000 Ljubljana, SloveniaFaculty of Civil and Geodetic Engineering, University of Ljubljana, Jamova cesta 2, 1000 Ljubljana, SloveniaFaculty of Chemistry and Chemical Technology, University of Ljubljana, Večna pot 113, 1000 Ljubljana, SloveniaBiotechnical Faculty, University of Ljubljana, Jamnikarjeva ulica 101, 1000 Ljubljana, SloveniaThis study integrates hyperspectral remote sensing with chemical and pedological data to estimate Zn, Pb, and Cd concentrations in the upper soil layers. Conducted in agricultural fields east and northeast of Celje, Slovenia, an area impacted by past industrial activities such as zinc ore smelting, the research integrates remote sensing and soil sampling to rapidly identify and map soil pollution over large surfaces. A multi-sensor approach was employed, combining two hyperspectral cameras (VNIR and SWIR, aerial), laboratory spectrometry, soil parameters, and content of chemical covariates measured with portable XRF and ICP-OES with a direct comparison of both techniques for this specific purpose. Accurate atmospheric and signal transformations were performed to improve modeling. The importance of covariates was thoroughly evaluated using conditional permutations to assess their contribution to the prediction of metal concentrations. The proposed framework utilizes spectral data and privileged information during training, improving prediction accuracy through a multi-stage model architecture. Here, a base model trained on spectral data is corrected using privileged information. During inference, the model functions without relying on privileged data providing a scalable and cost-effective solution for large-scale environmental monitoring. Our model achieved a reduction of predicted RMSE for Zn and Cd maps in comparison to the baseline models, translating to more precise identification of possibly polluted zones. However, for Pb, no improvements were observed, potentially due to variability in the data, including spectral issues or imbalances in the training and test datasets.https://www.mdpi.com/2072-4292/17/12/1987hyperspectral imageryaerial remote sensingimaging spectroscopydigital soil mappingpedometric mappingsoil contamination
spellingShingle Alen Mangafić
Krištof Oštir
Mitja Kolar
Marko Zupan
Hyperspectral Soil Heavy Metal Prediction via Privileged-Informed Residual Correction
Remote Sensing
hyperspectral imagery
aerial remote sensing
imaging spectroscopy
digital soil mapping
pedometric mapping
soil contamination
title Hyperspectral Soil Heavy Metal Prediction via Privileged-Informed Residual Correction
title_full Hyperspectral Soil Heavy Metal Prediction via Privileged-Informed Residual Correction
title_fullStr Hyperspectral Soil Heavy Metal Prediction via Privileged-Informed Residual Correction
title_full_unstemmed Hyperspectral Soil Heavy Metal Prediction via Privileged-Informed Residual Correction
title_short Hyperspectral Soil Heavy Metal Prediction via Privileged-Informed Residual Correction
title_sort hyperspectral soil heavy metal prediction via privileged informed residual correction
topic hyperspectral imagery
aerial remote sensing
imaging spectroscopy
digital soil mapping
pedometric mapping
soil contamination
url https://www.mdpi.com/2072-4292/17/12/1987
work_keys_str_mv AT alenmangafic hyperspectralsoilheavymetalpredictionviaprivilegedinformedresidualcorrection
AT kristofostir hyperspectralsoilheavymetalpredictionviaprivilegedinformedresidualcorrection
AT mitjakolar hyperspectralsoilheavymetalpredictionviaprivilegedinformedresidualcorrection
AT markozupan hyperspectralsoilheavymetalpredictionviaprivilegedinformedresidualcorrection