Characterization of Structural Building Damage in Post-Disaster using GLCM-PCA Analysis Integration

Understanding the characteristics of a building after a natural disaster can be achieved using image analysis techniques. Among these techniques are the Gray-Level Co-occurrence Matrix (GLCM) and Principal Component Analysis (PCA). In the GLCM process, the input image is converted into numerical values using eight different angles and varying pixel distances (1 and 0.5 pixels). The resulting numerical values from GLCM are then fed into the PCA process to reveal information stored within post-disaster building images. Interestingly, the PCA results differ between images processed with GLCM at a 1-pixel distance versus a 0.5-pixel distance. After validation based on surveyor assessments, it was found that the valid and accurate representation of real-world image information corresponds to the GLCM results obtained with a 0.5-pixel distance, indicating severe damage. This conclusion is supported by the fact that PCA results using a GLCM distance of 0.5 produce 2D and 3D visualizations predominantly clustered around severely damaged coordinates, with a range of values (n) where n ≥ 2. Therefore, integrating image analysis techniques such as GLCM and PCA can be used to determine the level of post-disaster building damage.