Deterioration Mapping of RC Bridge Elements Based on Automated Analysis of GPR Images

Ground-Penetrating Radar (GPR) is a popular non-destructive technique for evaluating RC bridge elements as it can identify major subsurface defects within short span of time. The data interpretation the GPR profiles based on existing amplitude-based approaches not completely reliable when compared to actual condition concrete with destructive measures. An alternative image-based analysis considers an imaging tool wherein experienced analyst marks attenuated areas and generates deterioration maps greater accuracy. However, this approach prone human errors highly subjective. proposed model aims improve through automated detection hyperbolas in classification mathematical modeling. Firstly, are pre-processed, hyperbolic reflections were detected them trained classifier using Viola–Jones Algorithm. false positives eliminated, missing regions identified automatically across top/bottom layer reinforcement user-interactive regional comparison statistical analysis. Subsequently, entropy, textural factor, evaluated differentiate closely equivalent visual system. These finally clustered entropy values K-means algorithm map generated which robust, reliable, corresponds situ state concrete. A case study parking lot demonstrated good correspondence by developed both amplitude- facilitate structural inspectors locally deteriorated zones that require immediate attention repair rehabilitation.