Show simple item record

dc.contributor.authorRäsänen, Aleksi
dc.contributor.authorRusanen, Antti
dc.contributor.authorKuitunen, Markku
dc.contributor.authorLensu, Anssi
dc.date.accessioned2013-10-15T09:05:54Z
dc.date.available2014-10-11T21:45:04Z
dc.date.issued2013
dc.identifier.citationRäsänen, A., Rusanen, A., Kuitunen, M., & Lensu, A. (2013). What makes segmentation good? A case study in boreal forest habitat mapping. <i>International Journal of Remote Sensing</i>, <i>34</i>(23), 8603-8627. <a href="https://doi.org/10.1080/01431161.2013.845318" target="_blank">https://doi.org/10.1080/01431161.2013.845318</a>
dc.identifier.otherCONVID_22815602
dc.identifier.otherTUTKAID_57776
dc.identifier.urihttps://jyx.jyu.fi/handle/123456789/42337
dc.description.abstractSegmentation goodness evaluation is a set of approaches meant for deciding which segmentation is good. In this study, we tested different supervised segmentation evaluation measures and visual interpretation in the case of boreal forest habitat mapping in Southern Finland. The data used were WorldView-2 satellite imagery, a lidar digital elevation model (DEM), and a canopy height model (CHM) in 2 m resolution. The segmentation methods tested were the fractal net evolution approach (FNEA) and IDRISI watershed segmentation. Overall, 252 different segmentation methods, layers, and parameter combinations were tested. We also used eight different habitat delineations as reference polygons against which 252 different segmentations were tested. The ranking order of segmentations depended on the chosen supervised evaluation measure; hence, no single segmentation could be ranked as the best. In visual interpretation among the several different segmentations that we found rather good, we selected only one as the best. In the literature, it has been noted that better segmentation leads to higher classification accuracy. We tested this argument by classifying 12 of our segmentations with the random forest classifier. It was found out that there is no straightforward answer to the argument, since the definition of good segmentation is inconsistent. The highest classification accuracy (0.72) was obtained with segmentation that was regarded as one of the best in visual interpretation. However, almost similarly high classification accuracies were obtained with other segmentations. We conclude that one has to decide what one wants from segmentation and use segmentation evaluation measures with care.fi
dc.language.isoeng
dc.publisherTaylor & Francis
dc.relation.ispartofseriesInternational Journal of Remote Sensing
dc.relation.urihttp://www.tandfonline.com/doi/full/10.1080/01431161.2013.845318#.UlZv8nczJ8E
dc.subject.otherluokitus
dc.subject.otherlidar
dc.titleWhat makes segmentation good? A case study in boreal forest habitat mapping
dc.typearticle
dc.identifier.urnURN:NBN:fi:jyu-201310152468
dc.contributor.laitosBio- ja ympäristötieteiden laitosfi
dc.contributor.laitosDepartment of Biological and Environmental Scienceen
dc.contributor.oppiaineYmpäristötiedefi
dc.contributor.oppiaineEnvironmental Scienceen
dc.type.urihttp://purl.org/eprint/type/JournalArticle
dc.date.updated2013-10-15T03:30:03Z
dc.type.coarjournal article
dc.description.reviewstatuspeerReviewed
dc.format.pagerange8603-8627
dc.relation.issn0143-1161
dc.relation.numberinseries23
dc.relation.volume34
dc.type.versionacceptedVersion
dc.rights.copyright© Taylor & Francis. This is an Author's final draft version of an article whose final and definitive form has been published by Taylor & Francis.
dc.rights.accesslevelopenAccessfi
dc.subject.ysokaukokartoitus
dc.subject.ysolaserkeilaus
dc.subject.ysosegmentointi
dc.subject.ysoluokitus (toiminta)
jyx.subject.urihttp://www.yso.fi/onto/yso/p2521
jyx.subject.urihttp://www.yso.fi/onto/yso/p21546
jyx.subject.urihttp://www.yso.fi/onto/yso/p18246
jyx.subject.urihttp://www.yso.fi/onto/yso/p12668
dc.relation.doi10.1080/01431161.2013.845318


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record