3-D Histogram-Based Segmentation and Leaf Detection for Rosette Plants

Abstract

Recognition and segmentation of plant organs like leaves is one of the major challenges in digital plant phenotyping. Here we present a 3-D histogram-based segmentation and recognition approach for top view images of rosette plants such as Arabidopsis thaliana and tobacco. Furthermore a euclidean-distance-map-based method for the detection of leaves and the corresponding plant leaf segmentation method were developed. An approach for the detection of optimal leaf split points for the separation of overlapping leaf segments was created. We tested and tuned our algorithms for the Leaf Segmentation Challenge (LSC). The results demonstrate that our method is robust and handles demanding imaging situations and different species with high accuracy.

Cite

Text

Pape and Klukas. "3-D Histogram-Based Segmentation and Leaf Detection for Rosette Plants." European Conference on Computer Vision, 2014. doi:10.1007/978-3-319-16220-1_5

Markdown

[Pape and Klukas. "3-D Histogram-Based Segmentation and Leaf Detection for Rosette Plants." European Conference on Computer Vision, 2014.](https://mlanthology.org/eccv/2014/pape2014eccv-d/) doi:10.1007/978-3-319-16220-1_5

BibTeX

@inproceedings{pape2014eccv-d,
  title     = {{3-D Histogram-Based Segmentation and Leaf Detection for Rosette Plants}},
  author    = {Pape, Jean-Michel and Klukas, Christian},
  booktitle = {European Conference on Computer Vision},
  year      = {2014},
  pages     = {61-74},
  doi       = {10.1007/978-3-319-16220-1_5},
  url       = {https://mlanthology.org/eccv/2014/pape2014eccv-d/}
}