A Machine-Vision Framework for Automated Egg Variety Recognition, Freshness Assessment, and Nutritional Estimation Using Machine Learning Techniques

This dataset contains image data developed for automated egg variety recognition, storage-duration-based visual freshness assessment, and image-based nutritional estimation. The dataset includes high-resolution RGB images of five egg categories: Bird Koel, Country Chicken, Red Layer Chicken, White Layer Chicken, and Duck. Images were collected from local markets in Dhaka, Bangladesh, under practical lighting conditions to reflect real-world visual variation in shell color, texture, speckling, gloss, and surface condition.
For egg species and freshness recognition, images are organized into five species classes and fifteen freshness-related visual classes formed from three storage-duration stages: Fresh/Day 0, Half-Fresh/Day 25, and Rotten/Day 45. Data augmentation was applied only to the training subset after dataset splitting to prevent data leakage. The dataset also includes 2,250 egg-and-coin images, where each egg is photographed beside a Bangladeshi 5-taka coin for reference-based mass and nutritional estimation.
This dataset supports research in computer vision, deep learning, food quality assessment, egg freshness classification, and non-invasive agricultural product analysis.