Quantitative Wood Anatomy (QWA), by providing precise measurements of xylem anatomical traits, gives key mechanistic insights into tree functioning and reconstruction of past environmental variability. Despite major progress in image analysis and artificial intelligence, the lack of open-source and automated tools for the analysis of microscopic images of wood sections, still limits reproducibility and large-scale application in QWA. Here, we introduce qwanamiz , a Python-based open-source program designed to automate the measurement and dating of tracheids in wood anatomical images. The program operates on binarised transversal microsections, combining image analysis with geometry-based algorithms to extract cell- and ring-level traits. A key feature is the use of Region Adjacency Graphs (RAG) to analyze the spatial organisation of tracheids, enabling accurate measurement of cell wall thickness, radial file structure, and detection of tree-ring boundaries. Comparisons with datasets obtained with established software indicate that qwanamiz produces consistent estimates for lumen dimensions and cell wall thickness at both the cell and ring scales. Interannual and decadal trends in anatomical traits are generally preserved, supporting the reliability of the approach. By integrating multiple analytical steps, from image processing to tracheid measurement and dating, within a single reproducible workflow, qwanamiz contributes to the development of more automated and fully open-source pipelines for QWA in conifers. This framework provides a functional foundation toward larger dendroanatomical analyses and aims at promoting future collaborative developments to refine programming methods for anatomical feature detection and measurements.
Bouchut et al. (Fri,) studied this question.