Pavement crack detection and classification based on fusion feature of LBP and PCA with SVM

A new crack detection approach based on local binary patterns (LBP) with support vector machine (SVM) was proposed in this paper. The propsed algorithm can extract the LBP feature from each frame of video taken road. Then, dimension spaces be reduced by Principal Component Analysis(PCA). simplified samples are trained to decided type using Support Vector Machine(SVM). In order reflect directional imformation detail, processed image is devided into nine sub-blocks. paper, driving tests were performed 10 times and 12,000 data applied algorithm. average accuracy sub-blocks 91.91%, which about 6.6% higher than without LBP-PCA SVM applying reflects information so that it has high 89.41% 88.24%, especially transverse longitudinal cracks. performance analysis different classifiers, F-Measure, considered balance between precision recall, alligator cracks classifier highest at 0.7601 hence others.