Gaussian Process Shape Models for Bayesian Segmentation of Plant Leaves

Kyle Simek and Kobus Barnard

Abstract

We develop a novel probabilistic model for multi-part shapes based on Gaussian processes, which we apply to model rosette leaves of Arabidopsis plants. Our model incorporates domain knowledge of Arabidopsis leaves in two ways. First, leaves are modeled using two anatomical parts: a blade and a petiole. We model the two regions with separate Gaussian processes, with a smoothness constraint at the boundary. Second, we constrain all leaf petioles to initiate at the rosette center, which is also modeled. This Bayesian prior is combined with a simple likelihood function over foreground pixels to perform image segmentation by optimizing a posterior distribution. A simple data-driven approach is used to over-segment the image, then excess leaves are pruned using a Bayesian model selection criterion. We show that our approach is effective, even with minimal training data.

Session

Workshop: Computer Vision Problems in Plant Phenotyping (CVPPP 2015)

Files

PDF iconPaper (PDF, 906K)

DOI

10.5244/C.29.CVPPP.4
https://dx.doi.org/10.5244/C.29.CVPPP.4

Citation

Kyle Simek and Kobus Barnard. Gaussian Process Shape Models for Bayesian Segmentation of Plant Leaves. In S. A. Tsaftaris, H. Scharr, and T. Pridmore, editors, Proceedings of the Computer Vision Problems in Plant Phenotyping (CVPPP), pages 4.1-4.11. BMVA Press, September 2015.

Bibtex

@inproceedings{CVPP2015_4,
	title={Gaussian Process Shape Models for Bayesian Segmentation of Plant Leaves},
	author={Kyle Simek and Kobus Barnard},
	year={2015},
	month={September},
	pages={4.1-4.11},
	articleno={4},
	numpages={11},
	booktitle={Proceedings of the Computer Vision Problems in Plant Phenotyping (CVPPP)},
	publisher={BMVA Press},
	editor={S. A. Tsaftaris, H. Scharr, and T. Pridmore},
	doi={10.5244/C.29.CVPPP.4},
	isbn={1-901725-55-3},
	url={https://dx.doi.org/10.5244/C.29.CVPPP.4}
}