Application of deep learning for coherent pixel selection in time series InSAR for urban area and transport infrastructure monitoring

The application of Time-series Interferometry of Synthetic Aperture Radar (TS-InSAR) for monitoring small and narrow infrastructural elements can be limited when using medium spatial resolution SAR sensors, such as Sentinel-1, due to sparse coverage of coherent pixels. Therefore, carefully selecting...

Full description

Saved in:
Bibliographic Details
Main Authors: S. Azadnejad, A. Kandiri, A. Hrysiewicz, F. O’Loughlin, E.P. Holohan, S. Dev, S. Donohue
Format: Article
Language:English
Published: Elsevier 2025-08-01
Series:International Journal of Applied Earth Observations and Geoinformation
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S1569843225003656
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850035104660848640
author S. Azadnejad
A. Kandiri
A. Hrysiewicz
F. O’Loughlin
E.P. Holohan
S. Dev
S. Donohue
author_facet S. Azadnejad
A. Kandiri
A. Hrysiewicz
F. O’Loughlin
E.P. Holohan
S. Dev
S. Donohue
author_sort S. Azadnejad
collection DOAJ
description The application of Time-series Interferometry of Synthetic Aperture Radar (TS-InSAR) for monitoring small and narrow infrastructural elements can be limited when using medium spatial resolution SAR sensors, such as Sentinel-1, due to sparse coverage of coherent pixels. Therefore, carefully selecting coherent pixels is crucial to maximise the density of reliable measurement points and to minimise noisy observations. In this study, a novel framework is proposed for selecting coherent pixels within the Stanford Method for Persistent Scatterers Small Baseline (StaMPS-SB) approach based on the temporal coherence predicted by deep learning. Unlike previous complex deep learning approaches to this problem, a multi-layer perceptron (MLP) model is trained on time-domain and frequency-domain features extracted from a time series of SAR amplitude images to predict the temporal coherence value of each individual pixel, and thus to select coherent pixels. A long short-term memory (LSTM) model, trained directly on the same amplitude time series, is used as a benchmark for evaluating the MLP model’s performance. Predictions of temporal coherence from MLP and LSTM models are similar, but the MLP model training time is much faster. Compared to standard amplitude difference dispersion (ADD) analysis, the MLP model increases the number of coherent pixels over buildings, linear infrastructure, and non-urban classes by 40 %, 68 %, and 45 %, respectively. In addition, the MLP model approach produces fewer unreliable (non-coherent) pixels; over non-urban regions, 70 % to 80 % of pixels selected by the ADD analysis have low temporal coherence (<0.8), compared to only 10 % to 12 % selected by the MLP model.
format Article
id doaj-art-99d1f3b3bb4d469da38794b27e314ada
institution DOAJ
issn 1569-8432
language English
publishDate 2025-08-01
publisher Elsevier
record_format Article
series International Journal of Applied Earth Observations and Geoinformation
spelling doaj-art-99d1f3b3bb4d469da38794b27e314ada2025-08-20T02:57:35ZengElsevierInternational Journal of Applied Earth Observations and Geoinformation1569-84322025-08-0114210471810.1016/j.jag.2025.104718Application of deep learning for coherent pixel selection in time series InSAR for urban area and transport infrastructure monitoringS. Azadnejad0A. Kandiri1A. Hrysiewicz2F. O’Loughlin3E.P. Holohan4S. Dev5S. Donohue6UCD School of Civil Engineering, University College Dublin, Dublin 4, Ireland; Corresponding author.UCD School of Civil Engineering, University College Dublin, Dublin 4, IrelandSFI Research Centre in Applied Geosciences (iCRAG), University College Dublin, Dublin 4, Ireland; UCD School of Earth Sciences, University College Dublin, Dublin 4, IrelandUCD School of Civil Engineering, University College Dublin, Dublin 4, Ireland; SFI Research Centre in Applied Geosciences (iCRAG), University College Dublin, Dublin 4, IrelandSFI Research Centre in Applied Geosciences (iCRAG), University College Dublin, Dublin 4, Ireland; UCD School of Earth Sciences, University College Dublin, Dublin 4, IrelandUCD School of Computer Science, University College Dublin, Dublin 4, IrelandUCD School of Civil Engineering, University College Dublin, Dublin 4, Ireland; SFI Research Centre in Applied Geosciences (iCRAG), University College Dublin, Dublin 4, IrelandThe application of Time-series Interferometry of Synthetic Aperture Radar (TS-InSAR) for monitoring small and narrow infrastructural elements can be limited when using medium spatial resolution SAR sensors, such as Sentinel-1, due to sparse coverage of coherent pixels. Therefore, carefully selecting coherent pixels is crucial to maximise the density of reliable measurement points and to minimise noisy observations. In this study, a novel framework is proposed for selecting coherent pixels within the Stanford Method for Persistent Scatterers Small Baseline (StaMPS-SB) approach based on the temporal coherence predicted by deep learning. Unlike previous complex deep learning approaches to this problem, a multi-layer perceptron (MLP) model is trained on time-domain and frequency-domain features extracted from a time series of SAR amplitude images to predict the temporal coherence value of each individual pixel, and thus to select coherent pixels. A long short-term memory (LSTM) model, trained directly on the same amplitude time series, is used as a benchmark for evaluating the MLP model’s performance. Predictions of temporal coherence from MLP and LSTM models are similar, but the MLP model training time is much faster. Compared to standard amplitude difference dispersion (ADD) analysis, the MLP model increases the number of coherent pixels over buildings, linear infrastructure, and non-urban classes by 40 %, 68 %, and 45 %, respectively. In addition, the MLP model approach produces fewer unreliable (non-coherent) pixels; over non-urban regions, 70 % to 80 % of pixels selected by the ADD analysis have low temporal coherence (<0.8), compared to only 10 % to 12 % selected by the MLP model.http://www.sciencedirect.com/science/article/pii/S1569843225003656Coherent pixels selectionInSARLSTMMLPTemporal coherence
spellingShingle S. Azadnejad
A. Kandiri
A. Hrysiewicz
F. O’Loughlin
E.P. Holohan
S. Dev
S. Donohue
Application of deep learning for coherent pixel selection in time series InSAR for urban area and transport infrastructure monitoring
International Journal of Applied Earth Observations and Geoinformation
Coherent pixels selection
InSAR
LSTM
MLP
Temporal coherence
title Application of deep learning for coherent pixel selection in time series InSAR for urban area and transport infrastructure monitoring
title_full Application of deep learning for coherent pixel selection in time series InSAR for urban area and transport infrastructure monitoring
title_fullStr Application of deep learning for coherent pixel selection in time series InSAR for urban area and transport infrastructure monitoring
title_full_unstemmed Application of deep learning for coherent pixel selection in time series InSAR for urban area and transport infrastructure monitoring
title_short Application of deep learning for coherent pixel selection in time series InSAR for urban area and transport infrastructure monitoring
title_sort application of deep learning for coherent pixel selection in time series insar for urban area and transport infrastructure monitoring
topic Coherent pixels selection
InSAR
LSTM
MLP
Temporal coherence
url http://www.sciencedirect.com/science/article/pii/S1569843225003656
work_keys_str_mv AT sazadnejad applicationofdeeplearningforcoherentpixelselectionintimeseriesinsarforurbanareaandtransportinfrastructuremonitoring
AT akandiri applicationofdeeplearningforcoherentpixelselectionintimeseriesinsarforurbanareaandtransportinfrastructuremonitoring
AT ahrysiewicz applicationofdeeplearningforcoherentpixelselectionintimeseriesinsarforurbanareaandtransportinfrastructuremonitoring
AT foloughlin applicationofdeeplearningforcoherentpixelselectionintimeseriesinsarforurbanareaandtransportinfrastructuremonitoring
AT epholohan applicationofdeeplearningforcoherentpixelselectionintimeseriesinsarforurbanareaandtransportinfrastructuremonitoring
AT sdev applicationofdeeplearningforcoherentpixelselectionintimeseriesinsarforurbanareaandtransportinfrastructuremonitoring
AT sdonohue applicationofdeeplearningforcoherentpixelselectionintimeseriesinsarforurbanareaandtransportinfrastructuremonitoring