AN IMAGE-BASED APPROACH FOR AUTOMATEDDAMAGE ASSESSMENT OF REINFORCED CONCRETECOLUMNS

Earthquakes can cause significant damage to structural components, such as columns, which canlead to a risk of collapse and extensive repair costs. Traditional damage assessment methods usuallyrely on visual inspection, which can be time-consuming and subjective. This study proposes anautomated damage assessment method using visual damage features of reinforced concrete columns.The visual damage features (i.e., crack length and crushed area) are extracted from a comprehensivedatabase comprising ۱۰۰ images of damaged concrete columns tested under cyclic quasi-staticloading. Then, a predictive equation for the stiffness-based damage index is derived using the linearregression method. The results show a potent correlation between the damage state of the columns andthe extracted visual damage features, with a meaningful correlation coefficient of ۰.۷۲. The proposedmethod can quickly and accurately assess damage, reducing the need for manual inspections andimproving the safety and efficiency of damage assessment processes.

کلیدواژه ها:

Reinforced Concrete Columns ، Visual Damage ، Predictive Equation ، Linear Regression