Limtic-DE : A Comprehensive Dataset for Deglet Nour Date Classification and Benchmarking of Deep Learning Models

the LIMTIC-DE dataset, is a collection of high-resolution images that captur 11 Tunisian Deglet Nour date varieties under controlled conditions. The dataset includes images from six distinct viewpoints, covering different maturity stages (oily, semi-oily, semi-dry, and dry) and treatment states (treated and untreated),making it a versatile resource for a wide range of applications. The dataset comprises 3,684 high-resolution images (4480 × 6720 pixels), pre-processed to focus on the date region and resized to a uniform resolution of 1280 × 1280 pixels. Each image has been manually annotated by domain experts, ensuring accurate classification of date types, maturity stages, and treatment states. The dataset is systematically organized into training, validation, and test sets, with a balanced distribution of classes across subsets to facilitate rigorous evaluation and reproducibility.