Mario Valerio Giuffrida, Massimo Minervini, Sotirios A. Tsaftaris

Computer Vision Problems in Plant Phenotyping (CVPPP) workshop, in conjuction with BMVC (2015)

Mario Valerio Giuffrida, Massimo Minervini, Sotirios A. Tsaftaris (2015) “Learning to Count Leaves in Rosette Plants,” Workshop: Computer Vision Problems in Plant Phenotyping (CVPPP), BMVC.

Giuffrida et al. (2015)
Open Externally
wp-content/uploads/2020/10/tex.png
Get Paper
@inproceedings{CVPP2015_1,
title={Learning to Count Leaves in Rosette Plants},
author={Mario Valerio Giuffrida and Massimo Minervini and Sotirios Tsaftaris},
year={2015},
month={September},
pages={1.1-1.13},
articleno={1},
numpages={13},
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.1},
isbn={1-901725-55-3}
}

Abstract

Counting the number of leaves in plants is important for plant phenotyping, since it can be used to assess plant growth stages. We propose a learning-based approach for counting leaves in rosette (model) plants. We relate image-based descriptors learned in an unsupervised fashion to leaf counts using a supervised regression model. To take advantage of the circular and coplanar arrangement of leaves and also to introduce scale and rotation invariance, we learn features in a log-polar representation. Image patches extracted in this log-polar domain are provided to K-means, which builds a codebook in a unsupervised manner. Feature codes are obtained by projecting patches on the codebook using the triangle encoding, introducing both sparsity and specifically designed representation. A global, per-plant image descriptor is obtained by pooling local features in specific regions of the image. Finally, we provide the global descriptors to a support vector regression framework to estimate the number of leaves in a plant. We evaluate our method on datasets of the \textit{Leaf Counting Challenge} (LCC), containing images of Arabidopsis and tobacco plants. Experimental results show that on average we reduce absolute counting error by 40% w.r.t. the winner of the 2014 edition of the challenge -a counting via segmentation method. When compared to state-of-the-art density-based approaches to counting, on Arabidopsis image data ~75% less counting errors are observed. Our findings suggest that it is possible to treat leaf counting as a regression problem, requiring as input only the total leaf count per training image.