Crack Detection in Civil Infrastructure: A Method-Scenario Review

Abstract

Ensuring the structural safety of civil infrastructure is vital for public welfare and cost-effective maintenance. Crack detection, as a key indicator of structural health, has transitioned from traditional image processing to advanced deep learning methods. This paper presents a systematic review of crack detection technologies organized under a novel “method-scenario” framework that categorizes techniques based on their underlying algorithms and the specific application contexts (e.g., pavements, bridges, tunnels, and specialized materials). By comparing conventional image processing approaches with modern deep learning models and multi-modal fusion techniques, we highlight the strengths and limitations of each method in various real-world scenarios. Our analysis reveals critical challenges—including data scarcity, sensitivity to noise, and the gap between theoretical models and practical deployment—which must be addressed to enhance reliability and generalizability. We conclude by proposing future research directions focused on integrating physics-based constraints with lightweight computational models and establishing unified evaluation protocols to bridge the gap between laboratory precision to engineering implementation.