Damage detection using support vector machine for integrity assessment of concrete structures

Abstract

In daily inspection for concrete structures, inspectors assess the integrity of them based on visible information such as cracks first. This study carries out damage detection using Support Vector Machine and Learning Vector Quantization for concrete slabs with cracks for the purpose of the development the integrity assessment system. These pattern recognition methods are used to learn the characteristics of the damage of the concrete slabs. The efficiency and applicability of proposed methods are discussed through the classification of the magnitude of damages extracted from photographed images of concrete slabs with cracks.