Classification of Road Pavement Defects Based on Convolution Neural Network in Keras

The aim of this paper is to propose the convolution neural network-VGG16 structure in Keras to classify pavement defects. In this paper, we present a method to build an automated system to classify different types of defects such as block cracks, longitudinal cracks and potholes. A region of interest is found and features are extracted using image processing techniques and machine learning methods. This system includes the following steps. The first step is to detect the defect location (ROI), then the defect is described by its features. Finally, each defect is classified based on these different features. The system ensures stable operation in the presence of limited lighting conditions, shadowing, and complex shaped defects.

Automated summary

This paper proposes a method using convolution neural network-VGG16 structure in Keras to classify pavement defects, and establishes an automated system that can operate stably under various lighting conditions, shadowing, and complex shaped defects.