Image Analysis of Arabidopsis Trichome Patterning in 4D Confocal Datasets
IEEE International Symposium on Biomedical Imaging (ISBI): 742--745, 2009
Abstract: In this article, we present an approach for the automated extraction of quantitative information about trichome patterning on leaves of Arabidopsis thaliana. Time series of growing rosette leaves (4D confocal datasets, 3D + time) are used for this work. At first, significant anatomical structures, i.e. leaf surface and midplane are extracted robustly. Using the extracted anatomical structures, a biological reference coordinate system is registered to the leaves. The performed registration allows to determine intra- as well as inter-series spatio-temporal correspondences. Trichomes are localized by first detecting candidates using Hough transform. Then, local 3D invariants are extracted and the candidates are validated using a Support Vector Machine (SVM).
Paper
Publisher's link
Images and movies
BibTex reference
@InProceedings{BRB09, author = "R.Bensch and O.Ronneberger and B.Greese and C. Fleck and K.Wester and M.H{\"u}lskamp and H.Burkhardt", title = "Image Analysis of Arabidopsis Trichome Patterning in 4D Confocal Datasets", booktitle = "IEEE International Symposium on Biomedical Imaging (ISBI)", pages = "742--745", year = "2009", url = "http://lmbweb.informatik.uni-freiburg.de/Publications/2009/BRB09" }