A few recent developments have demonstrated that deep-learning-based solutions can outperform traditional algorithms for automated pavement crack detection. In this paper, an efficient deep network called CrackNet-V is proposed for automated pixel-level crack detection on 3D asphalt pavement images. Compared with the original CrackNet, CrackNet-V has a deeper architecture but fewer parameters, resulting in improved accuracy and computation efficiency. Inspired by CrackNet, CrackNet-V uses invariant spatial size through all layers such that supervised learning can be conducted at pixel level. Following the VGG network, CrackNet-V uses 3x 3 size of filters for the first six convolutional layers and stacks several 3x 3 convolutional layers together for deep abstraction, resulting in reduced number of parameters and efficient feature extraction. CrackNet-V has 64,113 parameters and consists of ten layers, including one pre-process layer, eight convolutional layers, and one output layer. A new activation function leaky rectified tanh is proposed in this paper for higher accuracy in detecting shallow cracks. The training of CrackNet-V was completed after 3000 iterations, which took only one day on a GeForce GTX 1080Ti device. According to the experimental results on 500 testing images, CrackNet-V achieves a high performance with a Precision of 84.31%, Recall of 90.12%, and an F-1 score of 87.12%. It is shown that CrackNet-V yields better overall performance particularly in detecting fine cracks compared with CrackNet. The efficiency of CrackNet-V further reveals the advantages of deep learning techniques for automated pixel-level pavement crack detection.

Skid resistance depends on the macro- and micro-textural characteristics of pavement surface. Mean profile depth (MPD) is a widely used two-dimensional macro-texture indicator calculated from a single pavement surface profile, while pavement micro-texture is primarily affected by the characteristics of aggregates contained within the surface. In this study, twenty-two pavement sites, constructed using six common types of preventive treatments and eight typical sources of aggregates in Oklahoma, are selected as the field test beds. Pavement skid resistance and surface macro-texture data are collected in parallel at highway speeds using a grip tester and a high speed texture profiler, while aggregate properties under 3D are measured using a portable ultra-high resolution 3D laser imaging scanner. MPD values are calculated to represent surface macro-texture. Thirty-one 3D aggregate parameters in four categories (textural, feature, height, and material ratio & volume parameters) are investigated to characterise micro-texture attributes. The Pearson correlation coefficient matrix is computed within each category to exclude the parameters with strong correlations and remove their potential multi-collinearity during friction model development. Multivariate analysis is subsequently performed to examine the relationship between pavement skid resistance, and surface and aggregate texture properties. Eight texture parameters have showed statistical significance on pavement friction. Entropy (T_etp textural parameter) and peak curvature (S_pc feature based parameter) present the most significant influences on pavement friction, while MPD demonstrates the least impact. The developed model based on surface and aggregate texture properties can better predict pavement friction performance. The comprehensive analysis of texture properties could assist highway agencies understanding the appropriate characterisation of aggregates and developing aggregate selection guideline for optimised skid resistance.

Pavement friction and texture characteristics are important to road surface safety. Despite extensive studies conducted in the last decades, the relationship between pavement texture and surface friction has not been fully understood. This paper implements deep learning (DL) techniques to investigate the application of pavement texture data for pavement skid resistance and safety analysis. High speed texture profiles and grip tester friction data are collected in parallel on high friction surface treatment (HFST) sites including various types of lead-in and lead-out pavement sections distributed in 12 states of the United States. FrictionNet, a convolutional neural network (CNN)–based DL architecture, was developed to predict pavement friction levels directly using texture profiles. This architecture is composed of six artificial neuron layers: two convolution layers, three fully connected layers, and one output layer, with 606,409 tuned hyperparameters. There were 50,400 pairs of texture and friction data sets gathered for training, whereas another 12,600 pairs were gathered for validation and testing. The input of FrictionNet is the spectrogram of original texture profile for 1 m segments, and the output is the corresponding friction level ranging from 0.2 to 1.0. FrictionNet achieves 96.85% accuracy for training, 88.92% for validation, and 88.37% for testing in friction prediction. The result demonstrates the potential of using DL methods for highway speed noncontact texture measurements for pavement friction evaluation at the network level.

Pavement friction and texture characteristics are important aspects of road safety. Despite extensive studies conducted in the past decades, knowledge gaps still remain in understanding the relationship between pavement macrotexture and surface skid resistance. This paper implements discrete wavelet transform to decompose pavement surface macrotexture profile data into multi-scale characteristics and investigate their suitability for pavement friction prediction. Pavement macrotexture and friction data were both collected within the wheel-path from six High Friction Surface Treatment sites in Oklahoma using a high-speed profiler and a Grip Tester. The collected macrotexture profiles are decomposed into multiple wavelengths, and the total and relative energy components are calculated as indicators to represent macrotexture characteristics at various wavelengths. Correlation analysis is performed to examine the contribution of the energy indicators on pavement friction. The macrotexture energy within wavelengths from 0.97 mm to 3.86 mm contributes positively to pavement friction while that within wavelengths from 15.44 mm to 61.77 mm shows negative impacts. Subsequently, pavement friction prediction model is developed using multivariate linear regressive analysis incorporating the macrotexture energy indicators. Comparisons between predicted and monitored friction data demonstrates the robustness of the proposed friction prediction model.

Skid resistance is an important surface characteristic that influences roadway safety. Various studies have been performed to understand the interaction between pavement and tires through numerical simulation for skid resistance prediction. However, the friction parameters required for simulation inputs are generally determined by objective assumptions. This paper develops a finite element method (FEM)-based skid resistance simulation framework using in-situ 3D pavement surface texture and skid resistance data. A 3D areal pavement model is reconstructed from high resolution asphalt pavement surface texture data. The exponential decay friction model is implemented in the simulation and the interface friction parameters required for the simulation are determined using the binary search back-calculation approach based on a trial process with the desired level of differences between simulated and observed skid numbers. To understand the influence of texture characteristics on interface friction parameters, the high-resolution 3D texture data is separated into macro- and micro-scales through Butterworth filtering and various areal texture indicators are calculated at both levels. Principal component analysis (PCA) regression analysis is conducted to quantify the relationship between various texture characteristics and the interface friction parameters. The results from this study can be used to better prepare the inputs of friction parameters for FEM simulation.

Many studies have concluded that pavement micro- and macro- texture characteristics have significant contributions to pavement friction and thus roadway safety. This study explores novel three-dimensional (3D) areal texture parameters to represent pavement texture characteristics at both micro- and macroscales and their usage for friction prediction. Three rounds of pavement friction and texture data were collected from 2015 to 2017 at a testing site in Oklahoma with six different mixture designs. Pavement friction was acquired by a Dynamic Friction Tester, and the corresponding texture data were collected by a portable ultrahigh-resolution 3D laser scanner. Each 3D texture datum is decomposed into micro- and macroscales with predesigned Butterworth filters. Twenty-seven 3D texture parameters falling into five categories are calculated at both texture levels. Subsequently, the random forest algorithm is implemented to determine the most important texture parameters for friction predictive model development. The selected macro- and micro-texture parameters account for 48.8 % and 39.6 % contributions to high-speed friction and 50.0 % and 14.1 % contributions to low-speed friction. The temperature during testing also exhibits a significant impact, with 11.6 % and 35.9 % contributions to high- and low-speed friction, respectively.