Deep learning (DL) methods have transformed the way we extract plant
traits – both under laboratory as well as field conditions. Evidence
suggests that “well-trained” DL models can significantly simplify and
accelerate trait extraction as well as expand the suite of extractable
traits. Training a DL model typically requires the availability of
copious amounts of annotated data; however, creating large-scale
annotated dataset requires non-trivial efforts, time, and resources.
This has become a major bottleneck in deploying DL tools in practice.
Self-supervised learning (SSL) methods give exciting solution to this
problem, as these methods use unlabeled data to produce pretrained
models for subsequent fine-tuning on labeled data, and have demonstrated
superior transfer learning performance on down-stream classification
tasks. We investigated the application of SSL methods for plant stress
classification using few labels. Plant stress classification is a
fundamentally challenging problem in that (1) disease classification may
depend on abnormalities in a small number of pixels, (2) high data
imbalance across different classes, and (3) there are fewer annotated
and available plant stress images than in other domains. We compared
four different types of SSL methods on two different plant stress
datasets. We report that pre-training on unlabeled plant stress images
significantly outperforms transfer learning methods using random
initialization for plant stress classification. In summary, SSL based
model initialization and data curation improves annotation efficiency
for plant stress classification tasks.