A Deep Learning-Based Watershed Feature Fusion Approach for Tunnel Crack Segmentation in Complex Backgrounds

Materials, Год журнала: 2025, Номер 18(1), С. 142 - 142

Опубликована: Янв. 1, 2025

As highway tunnel operations continue over time, structural defects, particularly cracks, have been observed to increase annually. Coupled with the rapid expansion of networks, traditional manual inspection methods proven inadequate meet current demands. In recent years, machine vision and deep learning technologies gained significant attention in civil engineering for detection analysis defects. However, accurate defect identification tunnels presents challenges due complex background conditions, numerous interfering factors, relatively low proportion cracks within structure. Additionally, intensive labor requirements limited efficiency labeling training datasets pose constraints on deployment intelligent crack segmentation algorithms. To address these limitations, this study proposes an automatic optimization algorithm sample sets, utilizing features watershed enable efficient automated minimal human input. Furthermore, learning-based network was optimized through comparative various depths residual structure configurations achieve best possible model performance. Enhanced accuracy attained by incorporating axis extraction filling algorithms refine outcomes. Under diverse lining surface conditions multiple interference proposed approach achieved a 98.78%, Intersection Union (IoU) 72.41%, providing robust solution backgrounds.

Язык: Английский

A review of crack research in concrete structures based on data-driven and intelligent algorithms

Structures, Год журнала: 2025, Номер 75, С. 108800 - 108800

Опубликована: Апрель 8, 2025

Язык: Английский