Batangas Liberica Coffee Bean Morphology Image Dataset with Derived Morphometrics (BaLiCoM), Version 1

Open, well-documented datasets are increasingly treated as first-class research outputs because they enable verification, reuse, and extension of prior work without requiring the original investigators’ full experimental stack. This is the central rationale behind “data articles”/“data descriptors,” which are designed to emphasize dataset generation, structure, access, and reuse rather than hypothesis testing or interpretive claims. Scientific Data’s Data Descriptor template explicitly frames these articles as focused on helping others reuse data, requiring deposition in an appropriate repository prior to submission and public availability upon publication. [1] Data in Brief similarly positions data articles as peer-reviewed descriptions of datasets made available through repositories with persistent identifiers (typically DOIs). [2]

This Data Report describes a curated dataset of high-resolution single-bean images of Philippine Liberica coffee (Coffea liberica, commonly marketed locally as Kapeng Barako) acquired from farms in Batangas Province, Philippines, under controlled imaging conditions, along with derived pixel-based morphological measurements and (optionally) segmentation masks generated through a hybrid deep-learning + deterministic-feature-extraction pipeline. The source manuscript describes collection of approximately 4,000 Liberica bean images, controlled capture conditions (consistent background and lighting), preprocessing to a uniform input size, segmentation using a U-Net–style architecture, and morphometric extraction including area, perimeter, eccentricity, solidity, and centroid coordinates.

Liberica coffee is relevant in the Philippine context as a recognized variety (Barako) within national coffee industry planning and development; PCAARRD’s coffee industry profile notes that the Philippines is one of the few countries producing four commercially viable coffee types (including Liberica/Barako), and highlights the market positioning and production context for Philippine coffee. [3] Documenting Liberica bean morphology at scale supports agricultural quality assessment workflows (e.g., grading standardization, defect screening, or phenotyping-like trait measurement) by providing reproducible image-based measurements that can be reused for both applied and methodological studies.