Heritage_cracks

The HeritageCracks dataset is a comprehensive collection of images designed to characterize superficial damage in historical monuments within the city of Morelia, primarily focusing on temples. This dataset enables the automated detection and classification of structural deterioration using computer vision and deep learning techniques, contributing to the preservation and structural health monitoring of cultural heritage. This dataset is structured to facilitate the training and evaluation of machine learning models. The data distribution has been carefully divided to ensure optimal model performance: 75% for Training 15% for Validation 10% for Testing Note: The total dataset size after augmentation is 1,716 images.