IC4SD-Wood-Eucalyptus: A macroscopic transverse-section image dataset with metadata, split manifests, and leakage-audit reports for Eucalyptus wood identification

IC4SD-Wood-Eucalyptus is a macroscopic transverse-section image dataset for fine-grained Eucalyptus wood identification. The dataset contains 2,910 images acquired at 50× magnification from eight classes: seven Eucalyptus species (E. camaldulensis, E. cladocalyx, E. deglupta, E. diversicolor, E. grandis, E. microcorys, and E. saligna) and one Myrtaceae outgroup, Syzygium hemisphericum. The release includes raw images, class labels, original image dimensions, SHA-256 file hashes, parsed specimen/acquisition- group identifiers, perceptual-hash component identifiers, two split manifests, and leakage-audit outputs. The two distributed split manifests comprise a filename-derived specimen-stratified reference split and a strict pHash-clean group-disjoint split. The strict split contains 2,029 training images, 486 validation images, and 395 test images, with all eight classes represented in every partition. The accompanying leakage-audit files document parsed-group overlap, exact file-hash overlap, filename overlap, perceptual near-duplicate checks at two Hamming-distance thresholds, and feature nearest-neighbour contact sheets for manual review. The dataset is intended to support reproducible benchmarking of macroscopic woodimage classifiers and to provide an example of leakage-aware data organization for biological image classification.