Application of Deep Learning for Delineation of Visible Cadastral Boundaries from Remote Sensing Imagery

Research output: Contribution to journalArticleAcademicpeer-review

14 Downloads (Pure)

Abstract

Cadastral boundaries are often demarcated by objects that are visible in remote sensing imagery. Indirect surveying relies on the delineation of visible parcel boundaries from such images. Despite advances in automated detection and localization of objects from images, indirect surveying is rarely automated and relies on manual on-screen delineation. We have previously introduced a boundary delineation workflow, comprising image segmentation, boundary classification and interactive delineation that we applied on Unmanned Aerial Vehicle (UAV) data to delineate roads. In this study, we improve each of these steps. For image segmentation, we remove the need to reduce the image resolution and we limit over-segmentation by reducing the number of segment lines by 80% through filtering. For boundary classification, we show how Convolutional Neural Networks (CNN) can be used for boundary line classification, thereby eliminating the previous need for Random Forest (RF) feature generation and thus achieving 71% accuracy. For interactive delineation, we develop additional and more intuitive delineation functionalities that cover more application cases. We test our approach on more varied and larger data sets by applying it to UAV and aerial imagery of 0.02–0.25 m resolution from Kenya, Rwanda and Ethiopia. We show that it is more effective in terms of clicks and time compared to manual delineation for parcels surrounded by visible boundaries. Strongest advantages are obtained for rural scenes delineated from aerial imagery, where the delineation effort per parcel requires 38% less time and 80% fewer clicks compared to manual delineation.
Original languageEnglish
Article number2505
Pages (from-to)1-22
Number of pages22
JournalRemote sensing
Volume11
Issue number21
DOIs
Publication statusPublished - 25 Oct 2019

Fingerprint

imagery
learning
remote sensing
segmentation
surveying
boundary line
image resolution
road
vehicle

Keywords

  • indirect surveying
  • RF
  • CNN
  • image analysis
  • deep learning
  • machine learning
  • boundary delineation
  • boundary extraction
  • cadastral mapping
  • ITC-ISI-JOURNAL-ARTICLE
  • ITC-GOLD

Cite this

@article{be9afd73f6db4888a9fa453961c1e7a9,
title = "Application of Deep Learning for Delineation of Visible Cadastral Boundaries from Remote Sensing Imagery",
abstract = "Cadastral boundaries are often demarcated by objects that are visible in remote sensing imagery. Indirect surveying relies on the delineation of visible parcel boundaries from such images. Despite advances in automated detection and localization of objects from images, indirect surveying is rarely automated and relies on manual on-screen delineation. We have previously introduced a boundary delineation workflow, comprising image segmentation, boundary classification and interactive delineation that we applied on Unmanned Aerial Vehicle (UAV) data to delineate roads. In this study, we improve each of these steps. For image segmentation, we remove the need to reduce the image resolution and we limit over-segmentation by reducing the number of segment lines by 80{\%} through filtering. For boundary classification, we show how Convolutional Neural Networks (CNN) can be used for boundary line classification, thereby eliminating the previous need for Random Forest (RF) feature generation and thus achieving 71{\%} accuracy. For interactive delineation, we develop additional and more intuitive delineation functionalities that cover more application cases. We test our approach on more varied and larger data sets by applying it to UAV and aerial imagery of 0.02–0.25 m resolution from Kenya, Rwanda and Ethiopia. We show that it is more effective in terms of clicks and time compared to manual delineation for parcels surrounded by visible boundaries. Strongest advantages are obtained for rural scenes delineated from aerial imagery, where the delineation effort per parcel requires 38{\%} less time and 80{\%} fewer clicks compared to manual delineation.",
keywords = "indirect surveying, RF, CNN, image analysis, deep learning, machine learning, boundary delineation, boundary extraction, cadastral mapping, ITC-ISI-JOURNAL-ARTICLE, ITC-GOLD",
author = "S. Crommelinck and M.N. Koeva and M.Y. Yang and G. Vosselman",
year = "2019",
month = "10",
day = "25",
doi = "10.3390/rs11212505",
language = "English",
volume = "11",
pages = "1--22",
journal = "Remote sensing",
issn = "2072-4292",
publisher = "Multidisciplinary Digital Publishing Institute",
number = "21",

}

Application of Deep Learning for Delineation of Visible Cadastral Boundaries from Remote Sensing Imagery. / Crommelinck, S.; Koeva, M.N.; Yang, M.Y.; Vosselman, G.

In: Remote sensing, Vol. 11, No. 21, 2505, 25.10.2019, p. 1-22.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Application of Deep Learning for Delineation of Visible Cadastral Boundaries from Remote Sensing Imagery

AU - Crommelinck, S.

AU - Koeva, M.N.

AU - Yang, M.Y.

AU - Vosselman, G.

PY - 2019/10/25

Y1 - 2019/10/25

N2 - Cadastral boundaries are often demarcated by objects that are visible in remote sensing imagery. Indirect surveying relies on the delineation of visible parcel boundaries from such images. Despite advances in automated detection and localization of objects from images, indirect surveying is rarely automated and relies on manual on-screen delineation. We have previously introduced a boundary delineation workflow, comprising image segmentation, boundary classification and interactive delineation that we applied on Unmanned Aerial Vehicle (UAV) data to delineate roads. In this study, we improve each of these steps. For image segmentation, we remove the need to reduce the image resolution and we limit over-segmentation by reducing the number of segment lines by 80% through filtering. For boundary classification, we show how Convolutional Neural Networks (CNN) can be used for boundary line classification, thereby eliminating the previous need for Random Forest (RF) feature generation and thus achieving 71% accuracy. For interactive delineation, we develop additional and more intuitive delineation functionalities that cover more application cases. We test our approach on more varied and larger data sets by applying it to UAV and aerial imagery of 0.02–0.25 m resolution from Kenya, Rwanda and Ethiopia. We show that it is more effective in terms of clicks and time compared to manual delineation for parcels surrounded by visible boundaries. Strongest advantages are obtained for rural scenes delineated from aerial imagery, where the delineation effort per parcel requires 38% less time and 80% fewer clicks compared to manual delineation.

AB - Cadastral boundaries are often demarcated by objects that are visible in remote sensing imagery. Indirect surveying relies on the delineation of visible parcel boundaries from such images. Despite advances in automated detection and localization of objects from images, indirect surveying is rarely automated and relies on manual on-screen delineation. We have previously introduced a boundary delineation workflow, comprising image segmentation, boundary classification and interactive delineation that we applied on Unmanned Aerial Vehicle (UAV) data to delineate roads. In this study, we improve each of these steps. For image segmentation, we remove the need to reduce the image resolution and we limit over-segmentation by reducing the number of segment lines by 80% through filtering. For boundary classification, we show how Convolutional Neural Networks (CNN) can be used for boundary line classification, thereby eliminating the previous need for Random Forest (RF) feature generation and thus achieving 71% accuracy. For interactive delineation, we develop additional and more intuitive delineation functionalities that cover more application cases. We test our approach on more varied and larger data sets by applying it to UAV and aerial imagery of 0.02–0.25 m resolution from Kenya, Rwanda and Ethiopia. We show that it is more effective in terms of clicks and time compared to manual delineation for parcels surrounded by visible boundaries. Strongest advantages are obtained for rural scenes delineated from aerial imagery, where the delineation effort per parcel requires 38% less time and 80% fewer clicks compared to manual delineation.

KW - indirect surveying

KW - RF

KW - CNN

KW - image analysis

KW - deep learning

KW - machine learning

KW - boundary delineation

KW - boundary extraction

KW - cadastral mapping

KW - ITC-ISI-JOURNAL-ARTICLE

KW - ITC-GOLD

UR - https://ezproxy2.utwente.nl/login?url=https://library.itc.utwente.nl/login/2019/isi/crommelinck_app.pdf

U2 - 10.3390/rs11212505

DO - 10.3390/rs11212505

M3 - Article

VL - 11

SP - 1

EP - 22

JO - Remote sensing

JF - Remote sensing

SN - 2072-4292

IS - 21

M1 - 2505

ER -