Stomatal observation and automatic stomatal detection are useful analyses of stomata for taxonomic, biological, physiological, and eco-physiological studies. We present a new clearing method for improved microscopic imaging of stomata in soybean followed by automated stomatal detection by deep learning. We tested eight clearing agent formulations based upon different ethanol and sodium hypochlorite (NaOCl) concentrations in order to improve the transparency in leaves. An optimal formulation—a 1:1 (v/v) mixture of 95% ethanol and NaOCl (6–14%)—produced better quality images of soybean stomata. Additionally, we evaluated fixatives and dehydrating agents and selected absolute ethanol for both fixation and dehydration. This is a good substitute for formaldehyde, which is more toxic to handle. Using imaging data from this clearing method, we developed an automatic stomatal detector using deep learning and improved a deep-learning algorithm that automatically analyzes stomata through an object detection model using YOLO. The YOLO deep-learning model successfully recognized stomata with high mAP (~0.99). A web-based interface is provided to apply the model of stomatal detection for any soybean data that makes use of the new clearing protocol.

Stomata regulate CO₂ and water vapor exchange between leaves and the atmosphere, serving as a vital indicator of climate change resilience. Therefore, understanding the difference in stomatal numbers and patterns among different soybean cultivars across growth stages is essential to comprehending the complex mechanisms underlying soybean adaptation to climate change. The accurate measurements of stomatal density in soybean leaves are essential to understanding the complexity of stomatal density by environmental conditions. We demonstrated that the five epidermal sections and five microscopic images taken from both sides of each epidermal section at each leaf position (tip, middle, and bottom) were sufficient for stomatal measurements. Furthermore, we investigated variations in stomatal density among leaflet locations (left, right, and central) and leaf position across different growth stages. Notably, while there was no significant variation between the two leaves of the vegetative cotyledon (VC) stage and among the three leaflets of the V1 (first trifoliate) to V4 (fourth trifoliate) growth stages, leaves of the VC stage exhibited the lowest stomatal density, whereas those of the V4 stage exhibited the highest stomatal density. These findings could serve as a valuable tool for evaluating stomatal density, analyzing physiological differences under adverse climatic conditions, and phenotyping a large-scale population to identify the genetic factors responsible for stomatal density variations in soybean genotypes.