Urban land-use land-cover extraction for catchment modelling using deep learning techniques

Gong, S; Ball, J; Surawski, N

Urban land-use land-cover extraction for catchment modelling using deep learning techniques

Gong, S Ball, J Surawski, N

Permalink

Publisher:: IWA PUBLISHING
Publication Type:: Journal Article
Citation:: JOURNAL OF HYDROINFORMATICS, 2022, 24, (2), pp. 388-405
Issue Date:: 2022-03-01

Open Access

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is open access.

Adobe PDF

Download Accepted versionAdobe PDF (2.59 MB)

View on publisher's site

View statistics

Full metadata record

Field	Value	Language
dc.contributor.author	Gong, S
dc.contributor.author	Ball, J
dc.contributor.author	Surawski, N https://orcid.org/0000-0003-2584-1044
dc.date.accessioned	2022-04-26T00:40:03Z
dc.date.available	2022-04-26T00:40:03Z
dc.date.issued	2022-03-01
dc.identifier.citation	JOURNAL OF HYDROINFORMATICS, 2022, 24, (2), pp. 388-405
dc.identifier.issn	1464-7141
dc.identifier.issn	1465-1734
dc.identifier.uri	http://hdl.handle.net/10453/156593
dc.description.abstract	<jats:title>Abstract</jats:title> <jats:p>Throughout the world, the likelihood of floods and managing the associated risk are a concern to many catchment managers and the population residing in those catchments. Catchment modelling is a popular approach to predicting the design flood quantiles of a catchment with complex spatial characteristics and limited monitoring data to obtain the necessary information for preparing the flood risk management plan. As an important indicator of urbanisation, land use land cover (LULC) plays a critical role in catchment parameterisation and modelling the rainfall–runoff process. Digitising LULC from remote sensing imagery of urban catchment is becoming increasingly difficult and time-consuming as the variability and diversity of land uses occur during urban development. In recent years, deep learning neural networks (DNNs) have achieved remarkable image classification and segmentation outcomes with the powerful capacity to process complex workflow and features, learn sophisticated relationships and produce superior results. This paper describes end-to-end data assimilation and processing path using U-net and DeepLabV3+, also proposes a novel approach integrated with the clustering algorithm MeanShift. These methods were developed to generate pixel-based LULC semantic segmentation from high-resolution satellite imagery of the Alexandria Canal catchment, Sydney, Australia, and assess the applicability of their outputs as inputs to different catchment modelling systems. A significant innovation is using the MeanShift clustering algorithm to reduce the spatial noise in the raw image and propagate it to the deep learning network to improve prediction. All three methods achieved excellent classification performance, where the MeanShift+U-net has the highest accuracy and consistency on the test imagery. The final suitability assessment illustrates that all three methods are more suitable for the parameterisation of semi-distributed modelling systems rather than the fully distributed modelling systems, where the MeanShift+U-net should be adopted for image-based impervious area extraction of urban catchment due to its superior prediction accuracy of 98.47%.</jats:p>
dc.language	English
dc.publisher	IWA PUBLISHING
dc.relation.ispartof	JOURNAL OF HYDROINFORMATICS
dc.relation.isbasedon	10.2166/hydro.2022.124
dc.rights	info:eu-repo/semantics/openAccess
dc.subject	0905 Civil Engineering
dc.subject.classification	Environmental Engineering
dc.title	Urban land-use land-cover extraction for catchment modelling using deep learning techniques
dc.type	Journal Article
utslib.citation.volume	24
utslib.for	0905 Civil Engineering
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology/School of Civil and Environmental Engineering
pubs.organisational-group	/University of Technology Sydney/Strength - CTWW - Centre for Technology in Water and Wastewater Treatment
utslib.copyright.status	open_access	*
dc.date.updated	2022-04-26T00:39:58Z
pubs.issue	2
pubs.publication-status	Published
pubs.volume	24
utslib.citation.issue	2

Abstract:

Abstract Throughout the world, the likelihood of floods and managing the associated risk are a concern to many catchment managers and the population residing in those catchments. Catchment modelling is a popular approach to predicting the design flood quantiles of a catchment with complex spatial characteristics and limited monitoring data to obtain the necessary information for preparing the flood risk management plan. As an important indicator of urbanisation, land use land cover (LULC) plays a critical role in catchment parameterisation and modelling the rainfall–runoff process. Digitising LULC from remote sensing imagery of urban catchment is becoming increasingly difficult and time-consuming as the variability and diversity of land uses occur during urban development. In recent years, deep learning neural networks (DNNs) have achieved remarkable image classification and segmentation outcomes with the powerful capacity to process complex workflow and features, learn sophisticated relationships and produce superior results. This paper describes end-to-end data assimilation and processing path using U-net and DeepLabV3+, also proposes a novel approach integrated with the clustering algorithm MeanShift. These methods were developed to generate pixel-based LULC semantic segmentation from high-resolution satellite imagery of the Alexandria Canal catchment, Sydney, Australia, and assess the applicability of their outputs as inputs to different catchment modelling systems. A significant innovation is using the MeanShift clustering algorithm to reduce the spatial noise in the raw image and propagate it to the deep learning network to improve prediction. All three methods achieved excellent classification performance, where the MeanShift+U-net has the highest accuracy and consistency on the test imagery. The final suitability assessment illustrates that all three methods are more suitable for the parameterisation of semi-distributed modelling systems rather than the fully distributed modelling systems, where the MeanShift+U-net should be adopted for image-based impervious area extraction of urban catchment due to its superior prediction accuracy of 98.47%.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/156593