Search

Raveling is a widely encountered defect found in asphalt pavements. Raveling deteriorates riding safety and accelerates the development of other pavement defects. Therefore, timely and accurate classification of raveling severity is required to establish cost-effective pavement maintenance plans. This paper compares the performances of the Histogram-Based Gradient Boosting Classification Machine (HGBCM), Random Forest (RF), and Deep Convolutional Neural Network (DCNN) in recognizing raveling severity. Texture analysis based on the Gray-Level Co-Occurrence Matrix and statistical measurements of color channels is used for feature extraction. An image dataset including 3600 samples and three categories (non-raveling, minor raveling, and severe raveling) has been collected to train and verify the proposed methods. Experimental results supported by statistical hypothesis tests point out that HGBCM has achieved an outstanding performance with an accuracy rate > 0.96, F1 score > 0.94, and Cohen's Kappa coefficient > 0.92 for all class labels. • Propose computer vision approaches for classifying pavement raveling severity. • Employ histogram-based gradient boosting machine. • Image texture analysis is used for feature extraction. • Random forest and convolutional neural network are benchmark models. • The gradient boosting machine achieves an accuracy rate > 0.96. [ABSTRACT FROM AUTHOR]

This paper presents alternative solutions for classifying concrete spall severity based on computer vision approaches. Extreme Gradient Boosting Machine (XGBoost) and Deep Convolutional Neural Network (DCNN) are employed for categorizing image samples into two classes: shallow spall and deep spall. To delineate the properties of a concrete surface subject to spall, texture descriptors including local binary pattern, center symmetric local binary pattern, local ternary pattern, and attractive repulsive center symmetric local binary pattern (ARCS-LBP) are employed as feature extraction methods. In addition, the prediction performance of XGBoost is enhanced by Aquila optimizer metaheuristic. Meanwhile, DCNN is capable of performing image classification directly without the need for texture descriptors. Experimental results with a dataset containing real-world concrete surface images and 20 independent model evaluations point out that the XGBoost optimized by the Aquila metaheuristic and used with ARCS-LBP has achieved an outstanding classification performance with a classification accuracy rate of roughly 99%. • Propose a computer vision method for classifying concrete spall severity. • Extreme Gradient Boosting Machine and deep neural networks are employed. • Use local binary pattern variants as image texture descriptors. • The boosting machine is finetuned by the Aquila optimizer. • The proposed method achieves a high classification accuracy rate = 0.99. [ABSTRACT FROM AUTHOR]