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

Abstract

The stomatal index of the leaf is the ratio of the number of stomata to the total number of stomata and epidermal cells. Comparing with 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 manual counting of 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 regions with convolutional neural networks (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 the pipeline on 1,000 microscopic images of leaf epidermis in the wheat dataset (Triticum aestivum L.), 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% was achieved. R2 values between automatic and manual measurement of stomata, epidermal cells, and stomatal index were 0.995, 0.983, and 0.895, respectively. The average running time (ART) for the entire pipeline could be as short as 0.32 s per microphotograph. The proposed pipeline also achieved a good transferability on the other families of the plant 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 families of the plant. The pipeline is an automatic, rapid, and accurate tool for the stomatal index measurement, enabling high-throughput phenotyping, and facilitating further understanding of the stomatal and epidermal development for the plant physiology community. To the best of our knowledge, this is the first deep learning-based microphotograph analysis pipeline for stomatal index assessment.

Genome‐wide association study revealed TaHXK3‐2A as a candidate gene controlling stomatal index in wheat seedlings

“…Stomatal index was ultimately calculated as the proportion of stomatal cells among total epidermal cells (including both stomatal and nonstomatal cells) (Lee et al, 2015; Sack & Buckley, 2016). In addition, we used ImageJ software to count epidermal cells and stomata, and calculate the stomatal index, as a validation measure to check the statistics obtained by our automated pipeline (Zhu et al, 2021). The linear mixed effect function lmer (in the lme4 package of R Version 3.6.1) was used for best linear unbiased prediction (BLUP) analysis of the stomatal index.…”

“…A fully automated image analysis pipeline that counted both stomatal cells and total cells was used for stomatal index measurement (Zhu et al, 2021). Briefly, the Faster R-CNN algorithm (Ren & Gray, 2015) was used to detect and count stomata in each captured field of view.…”

Cotton Breeding in Australia: Meeting the Challenges of the 21st Century

“…The number of studies using deep learning to extract valuable information from crops and horticulture imagery or image-like data (e.g., LiDAR) is rapidly expanding, in particular around crop classification, weeds or disease detection, and fruit counting (for review, see Kamilaris and Prenafeta-Boldú, 2018). Recent examples also include several studies quantifying stomata in different species, and state-of-the-art cereals biomass prediction from LiDAR data (Fetter et al, 2019;Zhu et al, 2021;Pan et al, 2022).…”

JustDeepIt: Software tool with graphical and character user interfaces for deep learning-based object detection and segmentation in image analysis

“…On an individual scale, studies including species classification, crop disease detection, and weed detection are well researched ( Christin et al., 2019 ; Hasan et al., 2021 ; Liu and Wang, 2021 ). In addition, at the cell level, studies such as cell type identification and stomata identification via microscopic images have been performed ( Moen et al., 2019 ; Zhu et al., 2021 ). The increased availability of these techniques in various fields enhances the importance of the roles they are expected to play in the future.…”