Unlocking Comparative Plant Scoring with Siamese Neural Networks and Pairwise Pseudo Labelling

Zane K. J. Hartley, Rob J. Lind, Nicholas Smith, Bob Collison, Andrew P. French

ICCVW 2023 pp. 678-684

doi:10.1109/ICCVW60793.2023.00075 /iccvw/2023/hartley2023iccvw-unlocking/

Abstract

Phenotypic assessment of plants for herbicide discovery is a complex visual task and involves the comparison of a non-treated plant to those treated with herbicides to assign a phytotoxicity score. It is often subjective and difficult to quantify by human observers. Employing novel computer vision approaches using neural networks in order to be non-subjective and truly quantitative offers advantages for data quality, leading to improved decision making.In this paper we present a deep learning approach for comparative plant assessment using Siamese neural networks, an architecture that takes pairs of images as inputs, and we overcome the hurdles of data collection by proposing a novel pseudo-labelling approach for combining different pairs of input images. We demonstrate a high level of accuracy with this method, comparable to human scoring, and present a series of experiments grading Amaranthus retroflexus weeds using our trained model.

ICCVW Semantic Scholar

Cite

Text

Hartley et al. "Unlocking Comparative Plant Scoring with Siamese Neural Networks and Pairwise Pseudo Labelling." IEEE/CVF International Conference on Computer Vision Workshops, 2023. doi:10.1109/ICCVW60793.2023.00075

Markdown

[Hartley et al. "Unlocking Comparative Plant Scoring with Siamese Neural Networks and Pairwise Pseudo Labelling." IEEE/CVF International Conference on Computer Vision Workshops, 2023.](https://mlanthology.org/iccvw/2023/hartley2023iccvw-unlocking/) doi:10.1109/ICCVW60793.2023.00075

BibTeX

@inproceedings{hartley2023iccvw-unlocking,
  title     = {{Unlocking Comparative Plant Scoring with Siamese Neural Networks and Pairwise Pseudo Labelling}},
  author    = {Hartley, Zane K. J. and Lind, Rob J. and Smith, Nicholas and Collison, Bob and French, Andrew P.},
  booktitle = {IEEE/CVF International Conference on Computer Vision Workshops},
  year      = {2023},
  pages     = {678-684},
  doi       = {10.1109/ICCVW60793.2023.00075},
  url       = {https://mlanthology.org/iccvw/2023/hartley2023iccvw-unlocking/}
}