Three‐Dimensional Surface Texture Characterization of Portland Cement Concrete Pavements

Abstract

This article investigates the effectiveness of different mathematical methods in describing the three‐dimensional surface texture of Portland cement concrete (PCC) pavements. Ten PCC field cores of varying surface textures were included in the analysis. X‐ray Computed Tomography (CT) was used to scan the upper portion of these cores, resulting in a stack of two‐dimensional grayscale images. Image processing techniques were utilized to isolate the void pixels from the solid pixels and reconstruct the three‐dimensional surface topography. The resulting three‐dimensional surfaces were reduced to two‐dimensional “map of heights” images, whereby the grayscale intensity of each pixel within the image represented the vertical location of the surface at that point with respect to the lowest point on the surface. The “map of heights” images were analyzed using four mathematical methods, namely the Hessian model, the Fast Fourier transform (FFT), the wavelet analysis, and the power spectral density (PSD). Results obtained using these methods were compared to the mean profile depth (MPD) computed in accordance with ASTM E1845.

Pavement macro-texture analysis using wavelets

“…These include surface mixture type, aggregate source/gradation, binder type, traffic level, tyre characteristics and weather conditions (e.g. Flintsch et al 2003, Prowell and Hanson 2005, Abbas et al 2007, Cafiso and Taormina 2007, Kowalski et al 2009, Masad et al 2009, El Gendy et al 2011. Macro-texture measurements obtained at the Virginia Smart Road using the sand patch compared favourably with CTMeter measurements (Flintsch et al 2003).…”

“…The wavelet analysis was applied in quantifying aggregate particle surface texture using two-dimensional(2D) grey-scale images (Al-Rousan et al 2007). Work by Abbas et al (2007) also used wavelet analysis to characterise the 3D surface texture of Portland cement concrete (PCC) cores captured through X-ray computed tomography.…”

Pavement Friction Estimation Based on the Heinrich/Klüppel Model

“…The output of the transformation represents squared amplitude at each frequency of the original data in the frequency domain. When scaled in this manner, this transform is called a PSD function (Abbas et al, 2007;Sayers et al, 1998).…”

Impacts of Sample Size on Calculation of Pavement Texture Indicators with 1mm 3D Surface Data

“…Typically two methods are widely used to compute PSD for road profiles: one based on autocorrelation function; the other based on Fourier Transform (Abbas et al, 2007). In this research the second method is used.…”