Vision-Based Damage Detection for One-Fixed-End Structures Based on Aligned Marker Space and Decision Fusion.

Area of Science:

• Structural Engineering
• Computer Vision
• Non-Destructive Testing

Background:

• Vision-based methods for structural damage detection are limited to beam-like structures.
• A gap exists between current methods and the need to monitor diverse engineering structures.
• Assessing structural deformation under load is a key aspect of damage detection.

Purpose of the Study:

• To introduce a novel vision data morphing method (Aligned Marker Space) for damage detection in arbitrary engineering structures.
• To develop a fusion technique for combining multiple damage detection methods to improve accuracy and sensitivity.
• To validate the proposed methods through simulations and practical experiments on crane structures.

Main Methods:

• Development of the Aligned Marker Space method to generalize damage detection to any object with a fixed support.
• Implementation of a fusion technique to integrate results from various damage detection algorithms.
• Utilizing numerical simulations, blender-based simulations, and experimental testing on damaged crane structures.
• Optimization of damage detection parameters using an evolutionary algorithm to find Pareto-optimal solutions.

Main Results:

• The Aligned Marker Space method successfully enables damage detection in non-beam-like structures by treating them as cantilever beams.
• The fusion technique demonstrated increased accuracy and sensitivity in damage detection compared to individual methods.
• Practical experiments confirmed the effectiveness of the methods in identifying damage of varying sizes and locations in crane structures.
• Analysis provided insights into the influence of factors like camera position and damage location on detection performance.

Conclusions:

• The Aligned Marker Space method offers a versatile solution for vision-based structural damage detection across diverse engineering objects.
• The fusion technique enhances the reliability and precision of structural health monitoring systems.
• The study validates the proposed approach through comprehensive simulations and real-world experiments, paving the way for improved structural integrity assessment.