Show simple item record

dc.contributor.authorWijethunga, Pavithra A. L.en
dc.contributor.authorSamarasinghe, Sandhyaen
dc.contributor.authorKulasiri, Gamalathge D.en
dc.contributor.authorWoodhead, Ianen
dc.date.accessioned2010-06-16T02:50:09Z
dc.date.issued2009-11en
dc.identifier.citationWijethunga, P., Samarasinghe, S., Kulasiri, D., & Woodhead, I. (2009). Towards a generalized colour image segmentation for kiwifruit detection. In D. Bailey (Ed.), 2009 24th International Conference Image and Vision Computing New Zealand: IVCNZ 2009, James Cook Hotel Grand Chancellor, Wellington, New Zealand, 23-25 November 2009, conference proceedings (pp. 62-66). Piscataway, NJ: IEEE.en
dc.identifier.isbn978-1-4244-4697-1en
dc.identifier.issn2151-2205en
dc.identifier.urihttps://hdl.handle.net/10182/2074
dc.description.abstractDeveloping robust computer vision algorithms to detect fruit in trees is challenging due to less controllable conditions, including variation in illumination within an image as well as between image sets. There are two classes of techniques: local-feature-based techniques and shape-based techniques, which have been used extensively in this application domain. Out of the two classes, the local-feature-based techniques have shown higher accuracies over shape-based techniques, but are less desirable due to the requirement of repeated calibration. In this paper, we investigate the potential of developing a generalized colour pixel classifier that can be employed to detect kiwifruit on vines, under variable fruit maturity levels and imaging conditions. First, we observed the colour data patterns of fruit and nonfruit regions from different image sets. With consistant data patterns it was found that a suitable normalization could produce an invariant colour descriptor. Then, a neural network self-organizing map (SOM) model, which has a hierarchical clustering ability was used to investigate the potential of developing a generalized neural network model to classify pixels under variable conditions. Models were built for colour features extracted in CIELab space for both absolute colour values and relative colour descriptors. The paper presents the positive results of the preliminary investigations. The conditions for a successful application of the approach as well as the potential for extending it for automatic calibration will also be discussed.en
dc.format.extent62-66en
dc.language.isoenen
dc.publisherIEEEen
dc.relationThe original publication is available from - IEEEen
dc.rights© 2009 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.en
dc.sourceProceedings of the 23rd Image and Vision Computing New Zealanden
dc.subjectneural networksen
dc.subjectself organising maps (SOM)en
dc.subjectpixel classifieren
dc.subjectcolour image segmentationen
dc.subjectautomatic calibrationen
dc.subjectkiwifruiten
dc.subjecttree fruit detectionen
dc.subjectshape-based techniquesen
dc.subjectSOMen
dc.titleTowards a generalized colour image segmentation for kiwifruit detectionen
dc.typeConference Contribution - Published
lu.contributor.unitLincoln Universityen
lu.contributor.unitFaculty of Agriculture and Life Sciencesen
lu.contributor.unitDepartment of Wine, Food and Molecular Biosciencesen
lu.contributor.unitFaculty of Environment, Society and Designen
lu.contributor.unitDepartment of Environmental Managementen
lu.contributor.unitLincoln Agritechen
pubs.finish-date2009-11-25en
pubs.organisational-group/LU
pubs.organisational-group/LU/Agriculture and Life Sciences
pubs.organisational-group/LU/Agriculture and Life Sciences/WFMB
pubs.organisational-group/LU/Faculty of Environment, Society and Design
pubs.organisational-group/LU/Faculty of Environment, Society and Design/DEM
pubs.organisational-group/LU/Lincoln Agritech
pubs.organisational-group/LU/Research Management Office
pubs.organisational-group/LU/Research Management Office/QE18
pubs.publication-statusPublisheden
pubs.start-date2009-11-23en
lu.identifier.orcid0000-0002-0282-9811
lu.identifier.orcid0000-0001-8744-1578
lu.identifier.orcid0000-0003-2943-4331
lu.subtypeConference Paperen


Files in this item

Default Thumbnail

This item appears in the following Collection(s)

Show simple item record