TACK Tunnel Data (TTD): A Benchmark Dataset for Deep Learning-Based Defect Detection in Tunnels

2025-12-16

AI Summary

To address the bottleneck in tunnel defect detection—namely, the scarcity of high-quality, diverse, and well-annotated datasets hindering deep learning model development—this paper introduces and publicly releases the first multi-class, multi-lining-type visual benchmark dataset specifically designed for tunnel scenarios. The dataset encompasses three prevalent defect categories (cracks, leaching, and water seepage) across three lining materials (concrete, shotcrete, and masonry), comprising thousands of high-resolution images meticulously annotated by domain experts and managed with standardized metadata. It supports semantic segmentation, object detection, and self-supervised/semi-supervised learning paradigms. By systematically integrating multi-defect, multi-material, and high-textural diversity characteristics, the dataset fills a critical gap in domain-specific benchmarks, substantially enhancing model generalization and cross-tunnel transferability, thereby accelerating the practical deployment of automated tunnel inspection systems.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multi-instance/Multi-view Learning
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Tunnels are essential elements of transportation infrastructure, but are increasingly affected by ageing and deterioration mechanisms such as cracking. Regular inspections are required to ensure their safety, yet traditional manual procedures are time-consuming, subjective, and costly. Recent advances in mobile mapping systems and Deep Learning (DL) enable automated visual inspections. However, their effectiveness is limited by the scarcity of tunnel datasets. This paper introduces a new publicly available dataset containing annotated images of three different tunnel linings, capturing typical defects: cracks, leaching, and water infiltration. The dataset is designed to support supervised, semi-supervised, and unsupervised DL methods for defect detection and segmentation. Its diversity in texture and construction techniques also enables investigation of model generalization and transferability across tunnel types. By addressing the critical lack of domain-specific data, this dataset contributes to advancing automated tunnel inspection and promoting safer, more efficient infrastructure maintenance strategies.

Problem

• Addresses the scarcity of annotated tunnel datasets for automated defect detection.
• Introduces a benchmark dataset to support deep learning methods for tunnel inspections.
• Enables research on model generalization across diverse tunnel types and defects.

Innovation

• Introduces a public annotated tunnel defect dataset
• Supports supervised, semi-supervised, unsupervised deep learning methods
• Enables model generalization and transferability across tunnel types