Semi-automatic verification of cropland and grassland using very high resolution mono-temporal satellite images

Many public and private decisions rely on geospatial information stored in a GIS database. For good decisionmaking this information has to be complete, consistent, accurate and up-to-date. In this paper weintroduce a new approach for the semi-automatic verification of a specific part of the, possibl...

Full description

Bibliographic Details
Main Authors: Helmholz, Petra, Rottensteiner, F., Heipke, C.
Format: Journal Article
Published: Elsevier 2014
Subjects:
Online Access:http://hdl.handle.net/20.500.11937/39728
_version_ 1848755669659811840
author Helmholz, Petra
Rottensteiner, F.
Heipke, C.
author_facet Helmholz, Petra
Rottensteiner, F.
Heipke, C.
author_sort Helmholz, Petra
building Curtin Institutional Repository
collection Online Access
description Many public and private decisions rely on geospatial information stored in a GIS database. For good decisionmaking this information has to be complete, consistent, accurate and up-to-date. In this paper weintroduce a new approach for the semi-automatic verification of a specific part of the, possibly outdatedGIS database, namely cropland and grassland objects, using mono-temporal very high resolution (VHR)multispectral satellite images. The approach consists of two steps: first, a supervised pixel-based classificationbased on a Markov Random Field is employed to extract image regions which contain agriculturalareas (without distinction between cropland and grassland), and these regions are intersected withboundaries of the agricultural objects from the GIS database. Subsequently, GIS objects labelled as cropland or grassland in the database and showing agricultural areas in the image are subdivided into differenthomogeneous regions by means of image segmentation, followed by a classification of thesesegments into either cropland or grassland using a Support Vector Machine. The classification result ofall segments belonging to one GIS object are finally merged and compared with the GIS database label.The developed approach was tested on a number of images. The evaluation shows that errors in theGIS database can be significantly reduced while also speeding up the whole verification task when compared to a manual process.
first_indexed 2025-11-14T08:59:59Z
format Journal Article
id curtin-20.500.11937-39728
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T08:59:59Z
publishDate 2014
publisher Elsevier
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-397282017-09-13T15:15:03Z Semi-automatic verification of cropland and grassland using very high resolution mono-temporal satellite images Helmholz, Petra Rottensteiner, F. Heipke, C. Satellite images Quality control Mono-temporal Verification Automation GIS Many public and private decisions rely on geospatial information stored in a GIS database. For good decisionmaking this information has to be complete, consistent, accurate and up-to-date. In this paper weintroduce a new approach for the semi-automatic verification of a specific part of the, possibly outdatedGIS database, namely cropland and grassland objects, using mono-temporal very high resolution (VHR)multispectral satellite images. The approach consists of two steps: first, a supervised pixel-based classificationbased on a Markov Random Field is employed to extract image regions which contain agriculturalareas (without distinction between cropland and grassland), and these regions are intersected withboundaries of the agricultural objects from the GIS database. Subsequently, GIS objects labelled as cropland or grassland in the database and showing agricultural areas in the image are subdivided into differenthomogeneous regions by means of image segmentation, followed by a classification of thesesegments into either cropland or grassland using a Support Vector Machine. The classification result ofall segments belonging to one GIS object are finally merged and compared with the GIS database label.The developed approach was tested on a number of images. The evaluation shows that errors in theGIS database can be significantly reduced while also speeding up the whole verification task when compared to a manual process. 2014 Journal Article http://hdl.handle.net/20.500.11937/39728 10.1016/j.isprsjprs.2014.09.008 Elsevier restricted
spellingShingle Satellite images
Quality control
Mono-temporal
Verification
Automation
GIS
Helmholz, Petra
Rottensteiner, F.
Heipke, C.
Semi-automatic verification of cropland and grassland using very high resolution mono-temporal satellite images
title Semi-automatic verification of cropland and grassland using very high resolution mono-temporal satellite images
title_full Semi-automatic verification of cropland and grassland using very high resolution mono-temporal satellite images
title_fullStr Semi-automatic verification of cropland and grassland using very high resolution mono-temporal satellite images
title_full_unstemmed Semi-automatic verification of cropland and grassland using very high resolution mono-temporal satellite images
title_short Semi-automatic verification of cropland and grassland using very high resolution mono-temporal satellite images
title_sort semi-automatic verification of cropland and grassland using very high resolution mono-temporal satellite images
topic Satellite images
Quality control
Mono-temporal
Verification
Automation
GIS
url http://hdl.handle.net/20.500.11937/39728