Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Cutting-Edge Network Based Concrete Crack Detection and Analysis for Structural Health Monitoring
Structural Health Monitoring (SHM) plays a crucial role in ensuring the safety and longevity of infrastructure, with concrete structures being integral components of our built environment. This chapter introduces DeepCrack, an innovative approach utilizing deep learning techniques for the detection and analysis of concrete cracks in the context of structural health monitoring. DeepCrack influences state-of-the-art Convolutional Neural Networks (CNNs) to automatically identify and characterize cracks in concrete surfaces. The proposed system exhibits high accuracy and efficiency in detecting both visible and subtle cracks, providing a comprehensive solution for structural integrity assessment. By employing a large dataset of annotated concrete images, DeepCrack learns intricate patterns and features associated with different types and stages of cracks. The detection process is complemented by a detailed analysis module that evaluates cross entropy, training and validation accuracy. This analysis not only aids in quantifying the severity of cracks but also contributes valuable insights into potential structural issues. DeepCrack's adaptability enables it to handle various environmental conditions, lighting scenarios, and surface textures commonly encountered in real-world applications. Furthermore, the integration of DeepCrack into existing structural health monitoring systems enhances the overall efficiency of maintenance strategies. Real-time monitoring and continuous analysis of concrete surfaces enable timely identification of structural vulnerabilities, facilitating proactive interventions and preventing potential disasters. The results of extensive experiments demonstrate the superior performance of DeepCrack compared to traditional methods, showcasing its potential as a reliable and accurate tool for concrete crack detection and analysis. The scalability and versatility of DeepCrack make it suitable for a wide range of applications, from routine inspections of buildings and bridges to large-scale infrastructure projects.
Cutting-Edge Network Based Concrete Crack Detection and Analysis for Structural Health Monitoring
Structural Health Monitoring (SHM) plays a crucial role in ensuring the safety and longevity of infrastructure, with concrete structures being integral components of our built environment. This chapter introduces DeepCrack, an innovative approach utilizing deep learning techniques for the detection and analysis of concrete cracks in the context of structural health monitoring. DeepCrack influences state-of-the-art Convolutional Neural Networks (CNNs) to automatically identify and characterize cracks in concrete surfaces. The proposed system exhibits high accuracy and efficiency in detecting both visible and subtle cracks, providing a comprehensive solution for structural integrity assessment. By employing a large dataset of annotated concrete images, DeepCrack learns intricate patterns and features associated with different types and stages of cracks. The detection process is complemented by a detailed analysis module that evaluates cross entropy, training and validation accuracy. This analysis not only aids in quantifying the severity of cracks but also contributes valuable insights into potential structural issues. DeepCrack's adaptability enables it to handle various environmental conditions, lighting scenarios, and surface textures commonly encountered in real-world applications. Furthermore, the integration of DeepCrack into existing structural health monitoring systems enhances the overall efficiency of maintenance strategies. Real-time monitoring and continuous analysis of concrete surfaces enable timely identification of structural vulnerabilities, facilitating proactive interventions and preventing potential disasters. The results of extensive experiments demonstrate the superior performance of DeepCrack compared to traditional methods, showcasing its potential as a reliable and accurate tool for concrete crack detection and analysis. The scalability and versatility of DeepCrack make it suitable for a wide range of applications, from routine inspections of buildings and bridges to large-scale infrastructure projects.
Cutting-Edge Network Based Concrete Crack Detection and Analysis for Structural Health Monitoring
Springer Tracts in Civil Engineering
Jahangir, Hashem (Herausgeber:in) / Arora, Harish Chandra (Herausgeber:in) / Dos Santos, José Viriato Araújo (Herausgeber:in) / Kumar, Krishna (Herausgeber:in) / Kumar, Aman (Herausgeber:in) / Kapoor, Nishant Raj (Herausgeber:in) / Usha, S. Gandhimathi Alias (Autor:in)
Damage Detection and Structural Health Monitoring of Concrete and Masonry Structures ; Kapitel: 5 ; 157-175
22.03.2025
19 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Concrete Crack Detection from Video Footage for Structural Health Monitoring
Springer Verlag | 2021
|British Library Online Contents | 2006
|Sub pixel edge detection for structural health monitoring
Tema Archiv | 2009
|