International Journal of Science, Engineering and TechnologyAuthors: Assistant Professor, Mrs.P.Lakshmi Satya1, Shaik Fuzaila Farhatunnisa 2,, Mummidi Naga Durga Venkat3, Mohammad Samreen 4,, Sunkara Rajeev Nagendra5
Abstract: Structural cracks in buildings, bridges, and other civil infrastructures pose serious risks to public safety and long-term durability. Early and accurate detection of cracks is essential for effective maintenance and prevention of structural failures. Manual inspection methods are often time-consuming, labour-intensive, and prone to human error, making automated crack detection systems increasingly important.This project presents a deep learning–based framework for image-based crack prediction using a lightweight Convolutional Neural Network architecture. The proposed system utilizes MobileNet as the backbone model and incorporates transfer learning to enhance feature extraction capabilities while reducing computational complexity. Extensive image preprocessing and data augmentation techniques are applied to improve model generalization and handle limited dataset availability.The model is trained and evaluated on a structured crack image dataset consisting of cracked and non-cracked samples. Performance evaluation is carried out using standard metrics such as accuracy, precision, recall, F1-score, and confusion matrix analysis. Experimental results demonstrate that the proposed approach achieves high classification performance while maintaining computational efficiency, making it suitable for real-time infrastructure monitoring applications.The developed framework contributes toward automated structural health monitoring by providing a reliable, scalable, and efficient crack detection solution adaptable to practical engineering environments
