A Vision-Based Technique for Damage Assessment of Civil Structures
Salvatore Salamone Principal Investigator
MetadataShow full item record
Structural health monitoring is an important subject area of research because of the aging infrastructure of the nation and the desire to extend life of buildings and structures. Distress in reinforced concrete structures such as bridges, buildings, and nuclear power plants can be detected by cracks in is concrete. It is advantageous to record characteristics of the cracks remotely and mechanically to reduce use of manpower. The cracks can be recorded using digital camera. The challenge is to quantify the length and size of these cracks in computer based system. Mathematics based fractal analysis has a potential to quantify the cracks. An algorithm will be developed for fundamental fractal analysis of 2D digital visual images and apply the algorithm to surface cracks in reinforced concrete structures. The algorithm will be validated using previously collected data in laboratory tests. Research will be pursued to assess structural damage from the crack patterns.<br/><br/>The objective of this project is to create a vision-based structural health monitoring system for the automatic assessment of damage in reinforced concrete civil infrastructures. The system is based on the fractal analysis of 2D images of the visible spectrum and will be capable of retrieving surface crack patterns that can provide a quantitative measure of damage. The PI will investigate fundamental fractal properties of 2D digital visual images and apply the algorithm to surface cracks in reinforced concrete structures. The database from the Network for Earthquake Engineering Simulation Project Warehouse will be used to validate the algorithm. It is planned to validate the hypothesis that fractal analysis can be used for nondestructive evaluation method and can provide crack pattern. The crack pattern can be used to assess structural damage. It is planned to incorporate results of this project in existing graduate course.