Creation of Digital Microscope Dataset of Fruits and Vegetables for Automatic Quality Detection

The quality of fruits and vegetables is a crucial factor in the food distribution and consumption chain. This study aims to create a digital microscope image dataset of fruit and vegetable surfaces as a foundation for developing an AI-based automatic quality detection system. The research objects include fresh fruits and vegetables obtained from Teluk Dalam Market. The research process consists of seven stages: preparation, sampling, image acquisition, labeling, dataset compilation, validation, and publication. Images were captured using a digital microscope with a resolution of 1280×720 pixels and then labeled into three quality categories: good, slightly damaged, and severely damaged. The final dataset contains a number of images with structured metadata, uploaded to Google Drive, and made accessible for further research. This dataset has the potential to be used for training machine learning models and AI-based agricultural research.