Bridge Crack Detection Using Horse Herd Optimization Algorithm

Detection of Bridge Cracks is an important task for ensuring the safety and structural integrity of bridges. Cracks in bridges can lead to serious safety hazards and require frequent inspections to prevent accidents. However, manual inspections can be time-consuming and expensive. Automated detection systems can help in identifying cracks and reducing the cost and time associated with manual inspections. In previous years, several methods have been proposed for automating the crack-detecting procedure using Machine Learning and optimization techniques. In this study, a novel approach to detect bridge cracks is described using the Horse Herd Optimization Algorithm (HHOA). The HHOA is a population-based meta-heuristic algorithm motivated by the behaviour of horses in a herd. It is a relatively new algorithm that has shown promising results in several optimization problems. The Convolution Neural Network (CNN) is a popular deep learning technique used for image classification, object detection, and segmentation tasks. The basic idea behind CNN is to extract important features from an image using convolutional layers and pooling layers, followed by fully connected layers to classify the image. This study proposes an algorithm that can detect cracks in bridge structures efficiently and accurately using HHOA with CNN is giving an accuracy of 97%.