Stomata regulate important physiological processes in plants and are often phenotyped by researchers in diverse fields of plant biology. Currently, there are no user-friendly, fully automated methods to perform the task of identifying and counting stomata, and stomata density is generally estimated by manually counting stomata.
We introduce StomataCounter, an automated stomata counting system using a deep convolutional neural network to identify stomata in a variety of different microscopic images. We use a human-in-the-loop approach to train and refine a neural network on a taxonomically diverse collection of microscopic images.
Our network achieves 98.1% identification accuracy on Ginkgo scanning electron microscropy micrographs, and 94.2% transfer accuracy when tested on untrained species.
To facilitate adoption of the method, we provide the method in a publicly available website at http://www.stomata.science/....

StomataCounter: a neural network for automatic stomata identification and counting

10.1111/nph.15892