Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
CrackDiffusion: A two-stage semantic segmentation framework for pavement crack combining unsupervised and supervised processes
Abstract Achieving precise and reliable automated pavement crack detection using deep learning techniques is vital for intelligent pavement maintenance. This study proposes CrackDiffusion, an enhanced-supervised detection framework for pavement crack, combining two supervised and unsupervised stages. In Stage 1, a multi-blur-based cold diffusion anomaly detection model is proposed, which transforms crack-containing images into crack-free images, while simultaneously extracting pixel-level crack features using the Structural Similarity Index measure (SSIM). In Stage 2, an improved supervised U-Net segmentation model enhances accuracy and robustness by building upon the unsupervised results from Stage 1, ultimately producing highly accurate pixel-level segmentation results for cracks. On four public datasets, both the proposed multi-blur-based cold diffusion model and the comprehensive CrackDiffusion framework attained the highest Intersection over Union (IoU) scores, surpassing the IoU scores of the current state-of-the-practice unsupervised and supervised segmentation models.
Highlights A new unsupervised crack detection method based on diffusion is proposed. A new paradigm for enhancing supervised models with unsupervised information is proposed. An enhanced U-Net module is used to segment cracks from reconstructed images. On four public datasets, proposed CrackDiffusion framework achieves highest mean IoU.
CrackDiffusion: A two-stage semantic segmentation framework for pavement crack combining unsupervised and supervised processes
Abstract Achieving precise and reliable automated pavement crack detection using deep learning techniques is vital for intelligent pavement maintenance. This study proposes CrackDiffusion, an enhanced-supervised detection framework for pavement crack, combining two supervised and unsupervised stages. In Stage 1, a multi-blur-based cold diffusion anomaly detection model is proposed, which transforms crack-containing images into crack-free images, while simultaneously extracting pixel-level crack features using the Structural Similarity Index measure (SSIM). In Stage 2, an improved supervised U-Net segmentation model enhances accuracy and robustness by building upon the unsupervised results from Stage 1, ultimately producing highly accurate pixel-level segmentation results for cracks. On four public datasets, both the proposed multi-blur-based cold diffusion model and the comprehensive CrackDiffusion framework attained the highest Intersection over Union (IoU) scores, surpassing the IoU scores of the current state-of-the-practice unsupervised and supervised segmentation models.
Highlights A new unsupervised crack detection method based on diffusion is proposed. A new paradigm for enhancing supervised models with unsupervised information is proposed. An enhanced U-Net module is used to segment cracks from reconstructed images. On four public datasets, proposed CrackDiffusion framework achieves highest mean IoU.
CrackDiffusion: A two-stage semantic segmentation framework for pavement crack combining unsupervised and supervised processes
Han, Chengjia (Autor:in) / Yang, Handuo (Autor:in) / Ma, Tao (Autor:in) / Wang, Shun (Autor:in) / Zhao, Chaoyang (Autor:in) / Yang, Yaowen (Autor:in)
07.02.2024
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
Semi-supervised semantic segmentation using cross-consistency training for pavement crack detection
| Taylor & Francis Verlag | 2024
A complex scene pavement crack semantic segmentation method based on dual-stream framework
| Taylor & Francis Verlag | 2023
Bayesian-optimized unsupervised semantic segmentation model for structural crack detection
| SAGE Publications | 2024