A Stomata Classification and Detection System in Microscope Images

Figure 1: An overview of the stomata classification and detection system
Figure 2: In-depth explanation of the stomata classification process.

A Stomata Classification and Detection System in Microscope Images of Maize Cultivars

by Aono A. H., Nagai J. S., Dickel G. S. M., Marinho R. C., de Oliveira P. E. A. M., Faria F. A. (2019)

Alexandre H. Aono a, James S. Nagai a, Gabriella da S. M. Dickel b, Rafaela C. Marinho b, Paulo E. A. M. de Oliveira b, Fabio A. Faria a,∗

a Instituto de Ciencia e Tecnologia, Universidade Federal de S˜ao Paulo – UNIFESP 12247-014, S˜ao Jos´e dos Campos, SP – Brazil
b Instituto de Biologia, Universidade Federal de Uberlˆandia
Uberlndia, MG, Brazil


Figure 3: Examples of stoma (a) and non-stoma (b) subimages/regions, which were manually selected and labeled in this work.

In Biorxiv 2019-02 – https://doi.org/10.1101/538165


Figure 4: In-depth explanation of the stomata identification process


Stomata are morphological structures of plants that have been receiving constant attention. These pores are responsible for the interaction between the internal plant system and the environment, working on different processes such as photosynthesis process and transpiration stream. As evaluated before, understanding the pore mechanism play a key role to explore the evolution and behavior of plants. Although the study of stomata in dicots species of plants have advanced, there is little information about stomata of cereal grasses. In addition, automated detection of these structures have been presented on the literature, but some gaps are still uncovered.

Figure 5: Fifteen different microscope images of Maize Cultivars used in this work.

This fact is motivated by high morphological variation of stomata and the presence of noise from the image acquisition step. Herein, we propose a new methodology of an automatic stomata classification and detection system in microscope images for maize cultivars. In our experiments, we have achieved an approximated accuracy of 97.1% in the identification of stomata regions using classifiers based on deep learning features.

Figure 6: Different types of noise present in the microscopic images. (a) the usage of cyanoacrylate glue can
generate air bubbles; (b) leaves residuals might be captured by the microscope; (c) the leaves might bend and
generate grooves in the image; (d) degradated stomata due to biological factors; and (e) low image quality due
to equipment limitations.
Figure 7: Pos-processing of a microscope image.
Figure 8: Examples of the stomata classification results. (a) True positive subimages; and (b) False positive subimages.

Persistent drought monitoring using a microfluidic-printed electro-mechanical sensor of stomata in planta

Fig. 1 Printing conductive circuits on Spathiphyllum wallisii leaf lamina. (a) Microscope pictures of a stoma in the opened and closed states with
the stomatal aperture indicated. Scale bar: 10 μm. (b) Leaf surface resistance after leaf immersion into the conductive ink (n = 10). Bare leaf has ∼1
MΩ resistance. (c) Change in stomata sizes after 1 h white light illumination (I = 7 mW cm−2
, n = 10), demonstrating that the ink has no effect on
the stomatal aperture. (d) Schematics of conductive circuits printing on the leaf surface. A microfluidic chip is placed on top of the leaf abaxial
surface and clamped in between two holders. (e) Schematic layout of printed microsensor having two contact pads and a stripe going across a
single stoma. Conductive stripe breaks when stoma opens (bottom), increasing sensor resistance. (f) A height profile map demonstrates highly
non-planar leaf surface. Scale bar: 75 μm. (g) Bright-field microscopy images of a microfluidic chip aligned on top of a single stoma (i, iii) and the
same stoma after printing (ii, iv). Scale bars: 30 μm (i, ii) and 10 μm (iii, iv). Red arrows point to individual stomata. (h) Raman map for carbon nanotube G peak intensity (1590 cm−1
) on the leaf surface after the ink printing across a single stoma, demonstrating that the printed ink was confined
in the microfluidic channel.

Persistent drought monitoring using a microfluidic-printed electro-mechanical sensor of stomata in planta

by Koman V. B., Lew T. T. S., Wong M. H., Kwak S.-Y., Giraldo J. P., Strano M. S. (2017)

Volodymyr B. Koman, a Tedrick T. S. Lew, a Min Hao Wong, a Seon-Yeong Kwak, a
Juan P. Giraldo, b and Michael S. Strano*a


In Lab on a Chip 23 – DOI: 10.1039/c7lc00930e



Stomatal function can be used effectively to monitor plant hydraulics, photosensitivity, and gas exchange. Current approaches to measure single stomatal aperture, such as mold casting or fluorometric techniques, do not allow real time or persistent monitoring of the stomatal function over timescales relevant for long term plant physiological processes, including vegetative growth and abiotic stress. Herein, we utilize a nanoparticle-based conducting ink that preserves stomatal function to print a highly stable, electrical conductometric sensor actuated by the stomata pore itself, repeatedly and reversibly for over 1 week. This stomatal electro-mechanical pore size sensor (SEMPSS) allows for real-time tracking of the latency of single stomatal opening and closing times in planta, which we show vary from 7.0 ± 0.5 to 25.0 ± 0.5 min for the former and from 53.0 ± 0.5 to 45.0 ± 0.5 min for the latter in Spathiphyllum wallisii. These values are shown to correlate with the soil water potential and the onset of the wilting response, in quantitative agreement with a dynamic mathematical model of stomatal function. A single stoma of Spathiphyllum wallisii is shown to distinguish between incident light intensities (up to 12 mW cm−2) with temporal latency slow as 7.0 ± 0.5 min. Over a seven day period, the latency in opening and closing times are stable throughout the plant diurnal cycle and increase gradually with the onset of drought. The monitoring of stomatal function over long term timescales at single stoma level will improve our understanding of plant physiological responses to environmental factors.

A direct relation between the stomatal opening and the intensity of sunlight

Experimental setup (A) Schematic of the portable reflected type microscope having upward facing objective lens and novel leaf holder. Our setup is capable of direct real-time imaging and monitoring of lower epidermis stomata in field (B) Customized compact reflected microscope consists of a commercial finite 40× objective lens, BS and LED-based light source in 3D printed support frame (light source holder and lens holder). (C) Computer-aided design (CAD) of light source holder and lens holder. All the cavities of lens holder were lined with black color velvet paper from inside to reduce the reflection of light from walls of cavities. The whole microscope was covered by aluminum foil to ensure that ambient light should not interfere with the microscopic optical system in sunlight (D) Direct real-time imaging of lower epidermis stomata in field.

In-situ Real-time Field Imaging and Monitoring of Leaf Stomata by High-resolution Portable Microscope

Purwar P., Lee J. (2019)

Prashant Purwar, Junghoon Lee


In BioRxiv – doi: https://doi.org/10.1101/677450 –


Leaf holder: a device to hold and stabilize the leaf for long-term stomata imaging. (A) CAD design of leaf holder components. Upper leaf holder (4) is made of 3 mm thick transparent acryl sheet to allow continuous monitoring of stomata while the upper surface of the leaf exposed to the sunlight. However, 1.5 mm thick lower leaf holder (1, 2 or 3) was fabricated with a 3D printer using black color PLA material with and without a handle. V-shaped notches of width 2 mm have been designed to allow the wind blow and restrict the ambient light at the abaxial layer of the leaf. Small permanent neodymium magnets embedded in the upper and lower leaf holder helps to hold the leaf even if during the wind blow. (B) 3D-printed structures of leaf holder (C) Leaf holder in normal condition (C’) Use of lens cap to avoid ambient sunlight during microscopy. (D) Lower leaf holder with handle. Handle helps to attach the lower leaf holder to the upper part without damaging the biological structures of the lower epidermis of the leaf. (E) A gap of 0.5~1 mm between the lens cap and lower leaf holder to move the sample freely for stomata imaging.


Stomata, functionally specialized micrometer-sized pores on the epidermis of leaves (mainly on the lower epidermis), control the flow of gases and water between the interior of the plant and atmosphere. Real-time monitoring of stomatal dynamics can be used for predicting the plant hydraulics, photosensitivity, and gas exchanges effectively. To date, several techniques offer the direct or indirect measurement of stomatal dynamics, yet none offer real-time, long-term persistent measurement of multiple stomal apertures simultaneously of an intact leaf in a field under natural conditions. Here, we report a high-resolution portable microscope-based technique for in situ real-time field imaging and monitoring of stomata. Our technique is capable of analyzing and quantifying the multiple lower epidermis stomal pore dynamics simultaneously and does not require any physical or chemical manipulation of a leaf. An upward facing objective lens in our portable microscope allows the imaging of lower epidermis stomatal opening of a leaf while upper epidermis being exposed to the natural environment. Small depth of field (~ 1.3 μm) of a high-magnifying objection lens assists in focusing the stomatal plane in highly non-planar tomato leaf having a high density of trichome (hair-like structures). For long-term monitoring, the leaf is fixed mechanically by a novel designed leaf holder providing freedom to expose the upper epidermis to the sunlight and lower epidermis to the wind simultaneously. In our study, a direct relation between the stomatal opening and the intensity of  sunlight illuminating on the upper epidermis has been observed in real-time. In addition, real-time porosity of leaf (ratio between the areas of stomatal opening to the area of a leaf) and stomatal aspect ratio (ratio between the major axis and minor axis of stomatal opening) along with stomatal density have been quantified.

Real-time field monitoring of tomato stomata in sunlight (A) Imaging of stomatal opening at an interval of 15 minutes. Scale bar: 10 μm (B) Stomatal opening as a function of sunlight intensity. Stoma was responsive to the sunlight intensity falling on the upper epidermis compared to the maximum sunlight intensity of ambience inside the green house. The leaf under investigation was facing to sun in morning hours. However, during afternoon and evening it was under shadow of other leaves of the plant. The maximum stomatal opening of 80.3 μm2has been observed after maximum sunlight intensity recorded parallel and close to the leaf surface.

Automatic segmentation and measurement methods of living stomata

Flow chart of stomata segmentation and measurement. The red contour line is the segmentation contour from the proposed method, and the yellow contour line is the fitted ellipse

Automatic segmentation and measurement methods of living stomata of plants based on the CV model

Li K., Huang J., Song W., Wang J., Lv S., Wang X. (2019)

In Plant Methods15, Article number: 67 –



The stomata of plants mainly regulate gas exchange and water dispersion between the interior and external environments of plants and play a major role in the plants’ health. The existing methods of stomata segmentation and measurement are mostly for specialized plants. The purpose of this research is to develop a generic method for the fully automated segmentation and measurement of the living stomata of different plants. The proposed method utilizes level set theory and image processing technology and can outperform the existing stomata segmentation and measurement methods based on threshold and skeleton in terms of its versatility.

An Induction System for Clustered Stomata

An Induction System for Clustered Stomata by Sugar Solution Immersion Treatment in Arabidopsis thaliana Seedlings

by Akita K., Higaki T. (2019)


In Journal of Visualized Experiments – DOI: 10.3791/58951-



Stomatal movement mediates plant gas exchange, which is essential for photosynthesis and transpiration. Stomatal opening and closing are accomplished by a significant increase and decrease in guard cell volume, respectively. Because shuttle transport of ions and water occurs between guard cells and larger neighboring epidermal cells during stomatal movement, the spaced distribution of plant stomata is considered an optimal distribution for stomatal movement. Experimental systems for perturbing the spaced pattern of stomata are useful to examine the spacing pattern’s significance. Several key genes associated with the spaced stomatal distribution have been identified, and clustered stomata can be experimentally induced by altering these genes. Alternatively, clustered stomata can be also induced by exogenous treatments without genetic modification. In this article, we describe a simple induction system for clustered stomata in Arabidopsis thaliana seedlings by immersion treatment with a sucrose-containing medium solution. Our method is easy and directly applicable to transgenic or mutant lines. Larger chloroplasts are presented as a cell biological hallmark of sucrose-induced clustered guard cells. In addition, a representative confocal microscopic image of cortical microtubules is shown as an example of intracellular observation of clustered guard cells. The radial orientation of cortical microtubules is maintained in clustered guard cells as in spaced guard cells in control conditions.

Stomata in Adathoda (Acanthaceae) and Ixora (Rubiaceae)

Diacytic stomata in Adathoda vasica Nees

Recent approach to view stomatal anatomy under foldscope

Mownika (2018)

In Microcosmos


Paracytic stomata in Ixora coccinea L.

Here, with the help of Foldscope some of the beautiful stomata are differentiated with specific strains in Dicotyledons plants. They are as follows:


  1. For Diacytic stomatal differentiation, we have used Adathoda vasica
  2. Paracytic stomatal differentiation, we have used Ixora coccinea

Images of Stomata using the Wavelet Spot Detection and the Watershed Transform

Segmenting High-quality Digital Images of Stomata using the Wavelet
Spot Detection and the Watershed Transform

by Duarte K. T. N., de Carvallo M. A.G., Martins P. S. (2017)

Kaue T. N. Duarte, Marco A. G. de Carvalho, Paulo S. Martins,

University of Campinas (UNICAMP), School of Technology, R. Paschoal Marmo, 1888, 13484 Limeira, Brazil


In Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2017), pages 540-547 – ISBN: 978-989-758-225-7 – DOI: 10.5220/0006168105400547 –


Figure 3: Stomata from Ugni Molinae Species in RGB to CIELab (a) RGB-channel R; (b) RGB-channel G; (c) RGB-channel
B; (d) CIELab-channel L; (e) CIELab-channel a; (f) CIELab-channel b.


Stomata are cells mostly found in plant leaves, stems and other organs. They are responsible for controlling the gas exchange process, i.e. the plant absorbs air and water vapor is released through transpiration. Therefore, stomata characteristics such as size and shape are important parameters to be taken into account. In this paper, we present a method (aiming at improved efficiency) to detect and count stomata based on the analysis of the multi-scale properties of the Wavelet, including a spot detection task working in the CIELab colorspace.

Figure 4: Stomata Segmentation Process from UgniMolinae Species (a) Channel a* ; (b) Binary spots from Wavelet Spot
Detection; (c) Morphological Gradient; (d) Open; (e) Erode; (f) Reconstruct; (g) Close; (h) Watershed lines and spots; (i)
Stomata detected. The images (a) and (c-g) had their contrast enhanced for better visualization.

We also segmented stomata images using the Watershed Transform, assigning each spot initially detected as a marker. Experiments with real and high-quality images were conducted and divided in two phases. In the first, the results were compared to both manual enumeration and another recent method existing in the literature, considering the same dataset. In the second, the segmented results were compared to a gold standard provided by a specialist using the F-Measure. The experimental results demonstrate that the proposed method results in better effectiveness for both stomata detection and segmentation.