A Novel Road Segmentation Technique from Orthophotos Using Deep Convolutional Autoencoders

Alzuhairi, M; Pradhan, B

A Novel Road Segmentation Technique from Orthophotos Using Deep Convolutional Autoencoders

Alzuhairi, M Pradhan, B

Permalink

Publication Type:: Journal Article
Citation:: Korean Journal of Remote Sensing, 2017, 33, (4), pp. 423-436
Issue Date:: 2017-08-22

Open Access

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is open access.

Adobe PDF

Download Published versionAdobe PDF (4.73 MB)

View on publisher's site

View statistics

Full metadata record

Field	Value	Language
dc.contributor.author	Alzuhairi, M
dc.contributor.author	Pradhan, B https://orcid.org/0000-0001-9863-2054
dc.date.accessioned	2022-08-04T04:58:06Z
dc.date.available	2022-08-04T04:58:06Z
dc.date.issued	2017-08-22
dc.identifier.citation	Korean Journal of Remote Sensing, 2017, 33, (4), pp. 423-436
dc.identifier.issn	1225-6161
dc.identifier.issn	2287-9307
dc.identifier.uri	http://hdl.handle.net/10453/159618
dc.description.abstract	This paper presents a deep learning-based road segmentation framework from very high resolution orthophotos.The proposed method usesDeep Convolutional Autoencoders for end-to-end mapping of orthophotos to road segmentations. In addition, a set of post-processing steps were applied to make the model outputs GIS-ready data that could be useful for various applications. The optimization of the model’s parameters is explained whichwas conducted via grid search method.The modelwas trained and implemented in Keras, a high-level deep learning framework run on top of Tensorflow. The results show thatthe proposed model with the best-obtained hyperparameters could segment road objects from orthophotos at an average accuracy of 88.5%. The results of optimization revealed that the best optimization algorithm and activation function for the studied task are Stochastic Gradient Descent (SGD) and Exponential Linear Unit (ELU), respectively. In addition,the best numbers of convolutional filters were found to be 8 for the first and second layers and 128 for the third and fourth layers of the proposed network architecture. Moreover, the analysis on the time complexity of the model showed that the model could be trained in 4 hours and 50 minutes on 1024 high-resolution images of size 106 × 106 pixels, and segment road objects from similar size and resolution images in around 14 minutes.The results show that the deep learning models such as Convolutional Autoencoders could be a best alternative to traditional machine learning models for road segmentation from aerial photographs.
dc.language	en
dc.relation.ispartof	Korean Journal of Remote Sensing
dc.relation.isbasedon	10.7780/kjrs.2017.33.4.8
dc.rights	info:eu-repo/semantics/openAccess
dc.rights	This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons. org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited
dc.title	A Novel Road Segmentation Technique from Orthophotos Using Deep Convolutional Autoencoders
dc.type	Journal Article
utslib.citation.volume	33
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 - CAMGIS - Centre for Advanced Modelling and Geospatial lnformation Systems
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology/School of Information, Systems and Modelling
utslib.copyright.status	open_access	*
pubs.consider-herdc	false
dc.date.updated	2022-08-04T04:58:04Z
pubs.issue	4
pubs.publication-status	Published
pubs.volume	33
utslib.citation.issue	4

Abstract:

This paper presents a deep learning-based road segmentation framework from very high resolution orthophotos.The proposed method usesDeep Convolutional Autoencoders for end-to-end mapping of orthophotos to road segmentations. In addition, a set of post-processing steps were applied to make the model outputs GIS-ready data that could be useful for various applications. The optimization of the model’s parameters is explained whichwas conducted via grid search method.The modelwas trained and implemented in Keras, a high-level deep learning framework run on top of Tensorflow. The results show thatthe proposed model with the best-obtained hyperparameters could segment road objects from orthophotos at an average accuracy of 88.5%. The results of optimization revealed that the best optimization algorithm and activation function for the studied task are Stochastic Gradient Descent (SGD) and Exponential Linear Unit (ELU), respectively. In addition,the best numbers of convolutional filters were found to be 8 for the first and second layers and 128 for the third and fourth layers of the proposed network architecture. Moreover, the analysis on the time complexity of the model showed that the model could be trained in 4 hours and 50 minutes on 1024 high-resolution images of size 106 × 106 pixels, and segment road objects from similar size and resolution images in around 14 minutes.The results show that the deep learning models such as Convolutional Autoencoders could be a best alternative to traditional machine learning models for road segmentation from aerial photographs.

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

http://hdl.handle.net/10453/159618