Detection of Highway Pavement Damage Based on a CNN Using Grayscale and HOG Features

Aiming at the demand for rapid detection of highway pavement damage, many deep learning methods based on convolutional neural networks (CNNs) have been developed. However, CNN methods with raw image data require a high-performance hardware configuration and cost machine time. To reduce machine time and to apply the detection methods in common scenarios, the CNN structure with preprocessed image data needs to be simplified. In this work, a detection method based on a CNN and the combination of the grayscale and histogram of oriented gradients (HOG) features is proposed. First, the Gamma correction was employed to highlight the grayscale distribution of the damage area, which compresses the space of normal pavement. The preprocessed image was then divided into several unit cells, whose grayscale and HOG were calculated, respectively. The grayscale and HOG of each unit cell were combined to construct the grayscale-weighted HOG (GHOG) feature patterns. These feature patterns were input to the CNN with a specific structure and parameters. The trained indices suggested that the performance of the GHOG-based method was significantly improved, compared with the traditional HOG-based method. Furthermore, the GHOG-feature-based CNN technique exhibited flexibility and effectiveness under the same accuracy, in comparison to those deep learning techniques that directly deal with raw data. Since the grayscale has a definite physical meaning, the present detection method possesses a potential application for the further detection of damage details in the future.

Automated summary

This study proposes a fast detection method for highway pavement damage based on CNN and grayscale-weighted HOG features. The gamma correction is used to highlight the grayscale distribution of the damaged area, and the preprocessed image is divided into unit cells for feature calculation. The experimental results show that the GHOG-based method outperformed the traditional HOG-based method, and it exhibited flexibility and effectiveness compared to deep learning techniques that directly deal with raw data. Therefore, this method has a potential application for further detection of damage details.