The TAM and PSO–SVM methods for assessing the crack damage of concrete slab

Abstract

According to the crack images of concrete slab, two models – the Topographic Attentive Mapping (TAM) Neural Network which simulates the process of visual system and the Support Vector Machine (SVM) which is optimized by the Particle Swarm Optimization (PSO) algorithm – were adopted to assess the structural damages based on the digital image processing technology. Firstly, the TAM and the PSO–SVM assessment model was trained separately according to the crack images of bridge slab collected in advance and the damage level assessed by experts. Then, the established assessment models were used to assess the testing samples of the crack images. The results showed that the structural damage assessment method based on crack images is feasible and practical. The assessment effect of the TAM model is significantly better than that of the PSO–SVM model.