Speaker
Description
The digitisation of botanical collections and modern optical image analysis expand the opportunities for quantitative assessment of plant morphological traits. Standardising optical measurements and processing parameters reduces observer subjectivity and improves data comparability. Open access to high-resolution images, metadata and algorithms supports reproducible analysis and reuse of bio-optical data.
Leaf venation architecture is relevant to systematics, comparative anatomy, evolutionary biology and plant physiology. The vascular system provides diagnostic traits and forms an optically heterogeneous network associated with mesophyll water supply and with leaf hydraulic and light-scattering properties. Studying it requires high-resolution imaging, reproducible measurements and physical and anatomical interpretation.
This study aims to substantiate quantitative venation analysis in ImageJ/Fiji and to assess its integration with light and scanning electron microscopy for investigating grass leaf structure.
Light micrographs and macro-images of leaves with reticulate venation of Acer L., Quercus L., Betula L., Tilia L. and Populus L. were used to develop image processing algorithms. Computational analysis employed ImageJ/Fiji plugins and custom Java modules. The workflow combines textural, fractal and geometric–topological analyses of optical contrast. Processing comprises region-of-interest segmentation, greyscale conversion, adaptive thresholding and skeletonisation, which reduces vein images to single-pixel-wide centrelines. Networks are represented as graphs, with edges corresponding to vein segments and nodes to branching points, junctions and endpoints. Parameters include vein segment length, tortuosity and orientation, network connectivity and the area of enclosed meshes (areoles). Texture, fractal dimension and lacunarity are also assessed. Application to other network-like structures, such as microcrack networks in materials science, requires separate validation.
Adaptation of the workflow to Poaceae Barnhart accounts for the anisotropy of parallel venation by analysing longitudinal and transverse veins separately. The algorithms are configured to quantify the number of longitudinal veins of each type, intervein spacing, the number of transverse veins per unit area and the total vein length per unit area for each vein group, taking leaf blade geometry into account.
For anatomical validation, cleared-leaf images were compared with transverse sections and epidermal micrographs from adjacent regions. Light microscopy reveals the structure of vascular bundles and surrounding tissues. Scanning electron microscopy complements this analysis by characterising epidermal surface topography, stomatal arrangement and the organisation of costal (over-vein) and intercostal (interveinal) zones.
The proposed integration of digital morphometry and microscopy offers prospects for comprehensive analysis of grass leaf structure, comparison of species and genotypes, and assessment of the diagnostic value of venation traits. Comparing morphoanatomical parameters with statistical analyses of optical measurements, including light transmission and scattering, may clarify the information content and functional significance of these parameters, provided that it is supported by physiological and biophysical data. Implementing this approach requires standardised optical image acquisition, anatomical validation and experimental verification.