Dark-field Microscopy Dataset and SVM Code for Agave-Based Mezcal Classification

Dataset Title:Dark-Field Microscopy Images of Evaporated Mezcal Droplets for Agave Species Classification Description:This dataset contains dark-field microscopy images of mezcal samples produced from four agave species: Agave salmiana (salmiana), Agave marmorata (tepeztate), Agave rhodacantha (cuishe), and Agave angustifolia (espadin), as well as an aged salmiana. Each 1 μL droplet of diluted mezcal (20% ABV) was deposited on a cleaned glass slide and allowed to evaporate under ambient conditions to form distinct microstructures. The resulting images were acquired at 4× magnification and used to train and validate a Support Vector Machine (SVM) classifier to distinguish between the first two varietals. The dataset supports research in agave-based spirit authentication, chemometric image analysis, and low-cost classification of artisanal products. Contents: JPEG or PNG image files organized by class (/salmiana/, /tepeztate/, /espadin/, /cuishe/,and /tepeztate/) Python scripts and Jupyter Notebooks for training, evaluation, and model export Pretrained SVM model and label encoder files Format:Images (224×224 pixels), Notebooks (.ipynb) Intended Use:Research in chemometrics, machine learning, and food authentication. May also serve as a benchmark dataset for image-based classification of fermented or distilled products. License:Creative Commons Attribution 4.0 International (CC BY 4.0)