AUTHOR=Zhu Chuancheng , Hu Yusong , Mao Hude , Li Shumin , Li Fangfang , Zhao Congyuan , Luo Lin , Liu Weizhen , Yuan Xiaohui 

TITLE=A Deep Learning-Based Method for Automatic Assessment of Stomatal Index in Wheat Microscopic Images of Leaf Epidermis

JOURNAL=Frontiers in Plant Science

VOLUME=Volume 12 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2021.716784

DOI=10.3389/fpls.2021.716784

ISSN=1664-462X

ABSTRACT=The stomatal index of the leaf is the ratio of stomata number to the total number of stomata and epidermal cells. Comparing to the stomatal density, the stomatal index is relatively constant in environmental conditions and the age of the leaf and, therefore, of diagnostic characteristics for a given genotype or species. Traditional assessment methods involve manually counting the number of stomata and epidermal cells in microphotographs, which is labor-intensive and time-consuming. Although several automatic measurement algorithms of stomatal density have been proposed, no stomatal index pipelines are currently available. The main aim of this research is to develop an automated stomatal index measurement pipeline. The proposed method employed Faster R-CNN and U-Net and image-processing techniques to count stomata and epidermal cells and subsequently calculate the stomatal index. To improve the labeling speed, a semi-automatic strategy was employed for epidermal cell annotation in each micrograph. Benchmarking our pipeline on 1,000 microscopic images of leaf epidermis in wheat dataset (Triticum aestivum L.), we achieved the average counting accuracies of 98.03% and 95.03% for stomata and epidermal cells, respectively, and the final measurement accuracy of the stomatal index of 95.35%. R2 values between automatic and manual measurement of stomata, epidermal cells, and stomatal index were 0.995, 0.983, and 0.882, respectively. The average running time for the entire pipeline could be as short as 0.32s per microphotograph. The proposed pipeline also achieved a good transferability on other plant families using transfer learning, with the mean counting accuracies of 94.36% and 91.13% for stomata and epidermal cells and the stomatal index accuracy of 89.38% in seven plant families. Our pipeline is an automatic, rapid, and accurate tool for stomatal index measurement, enabling high-throughput phenotyping and facilitating further understanding of the stomatal and epidermal development for the plant physiology community. To our best knowledge, this is the first deep learning-based microphotograph analysis pipeline for stomatal index assessment.