Digital Image-Based Characterization of Particle Size and Shape Distribution in Industrial Hemp Shives

This dataset contains original and processed images of industrial hemp shives intended for quantitative analysis of particle size and shape using image analysis techniques. The images were acquired using flatbed scanning and subsequently processed in ImageJ to produce binarized and labeled versions. In total, 99 samples are included, covering three particle size ranges (2–4 mm, >4 mm, and a 50/50 mix of both). The dataset consists of: - **Raw scans** (`/Hemp_Shives_Scans`): high-resolution PNG images of loose hemp shives- **Binary images** (`/Binary_Images`): thresholded versions used for particle segmentation- **Labeled particle images** (`/Labelled_Particles_Images`): images with particle IDs drawn by ImageJ- **Summary_of_Results.xlsx**: spreadsheet containing particle count and geometric descriptors per sample (area, perimeter, Feret diameters) Each sample represents one of the following sieve groups:- `S1_2-4` to `S33_2-4`: particles retained between 2 mm and 4 mm sieves- `S34_4` to `S66_4`: particles retained on a 4 mm sieve- `S67_50-50` to `S99_50-50`: a 50/50 mix of the above two fractions ### Use cases:This dataset is intended for use in:- Training and validation of image-based particle analysis algorithms- Research on bio-based construction materials- Morphological analysis of lignocellulosic aggregates All data were processed using ImageJ (version 1.54p). Manual thresholding and “Analyze Particles” functions were applied with filtering by size and circularity. Full processing parameters are available upon request. **License:** CC BY-NC 4.0 – non-commercial academic use only. Redistribution and adaptation are permitted with proper citation. If you use this dataset, please cite:Piątkiewicz W., Narloch P., Anysz H., Wółczyńska Z., Mańczak J. (2025). *Size and Shape Distribution of Hemp Shives Based on Image Analysis*. Warsaw University of Technology. DOI: 10.5281/zenodo.15664512