Concrete Surface Crack Detection using Hierarchal Convolutional Neural Network

Davis Agyemang, BSc Software Engineering

Cracks on concrete walls can imply that a building possesses issues with its structural integrity. Surveyors who inspect these defects are expected to provide their customers with excellent evaluation regarding its severity. The process is currently conducted through visual inspection, resulting in occasions of subjective judgements being made on the classification and severity of the concrete surface. This poses danger for customers and the environment as it not being analysed objectively. Many researchers have applied numerous classification techniques to tackle this issue but from the author’s knowledge, their methods do not allow the classification result to be sent back to the user for them look back at the concrete surface, or provide the severity level of the evaluated concrete crack and there is no feedback mechanism in terms of adaptability of when their method classifies a concrete surface incorrectly. In this paper, the author proposes in building a hybrid web mobile application with the capabilities of giving in-depth information about a concrete surface’s severity, users being able to send the classification results to themselves via email and the ability to improve the accuracy of the application via user feedback. The application will be build using Python, Flask, Keras with Tensorflow backend, HTML, CSS and JavaScript, in which a trained Hierarchal-Convolutional Neural Network(H-CNN) will be used to evaluate the concrete surface via images taken by a mobile device or uploaded via a desktop.

This entry was posted in SC2019 and tagged , . Bookmark the permalink.