StomataCounter: a neural network for automatic stomata identification and counting
StomataCounter: a neural network for automatic stomata identification and counting
复制标题
DOI：
10.1111/nph.15892
复制
发表时间：
2019-08-01
期刊：
影响因子：
9.4
通讯作者：
Keller, Stephen R.
中科院分区：
文献类型：
--
作者：
Fetter, Karl C.;Eberhardt, Sven;Keller, Stephen R.
气孔调节植物的重要生理过程，植物生物学不同领域的研究人员经常对气孔进行表型分析。目前，还没有用户友好的、完全自动化的方法来执行气孔识别和计数的任务，并且气孔密度通常通过手动计数气孔来估计。我们介绍了 StomataCounter，这是一种自动气孔计数系统，使用深度卷积神经网络来识别各种不同显微图像中的气孔。我们使用人机交互的方法来训练和完善基于分类多样化的显微图像集合的神经网络。我们的网络在银杏扫描电子显微镜显微照片上的识别准确率达到了 98.1%，在未经训练的物种上进行测试时，传输准确率达到了 94.2%。为了促进该方法的采用，我们在公开网站上提供了该方法，网址为 。
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 .