Publication: Development of deep learning algorithm using yolov8 to detect crack on concrete surfaces
Loading...
Date
2023-06-19
Authors
Jeremy, Choy Jun Min
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Crack detection plays a crucial role in infrastructure maintenance and safety. Traditional methods for crack detection rely on manual visual inspection which is time-consuming and labor-intensive. In recent year, deep learning techniques have shown promising results in automating crack detection. This paper presents the development of a deep learning algorithm to perform crack detection and segmentation on concrete crack surfaces. There are two models developed which achieve mAP of 79% and 74% respectively, tuned with different hyperparameters. Both models are deployed on multiple platforms including Windows, macOS and Android.