Machine vision based characterization of particle shape and asphalt coating in Reclaimed Asphalt Pavement

Abstract

Reclaimed Asphalt Pavement (RAP) particles are created from the impact removal and/or reprocessing of existing asphalt layers. RAP particles contain a combination of asphalt and aggregates with varying degrees of coating and morphology. Particle size and shape properties, amount of asphalt coating the RAP particles, and the binder content of the RAP are among the important engineering properties that control the performance of this material. This paper introduces an innovative machine vision-based inspection system to quantify the percentage of asphalt coating in different RAP aggregate sources. The Enhanced-University of Illinois Aggregate Image Analyzer (E-UIAIA) is used to acquire the color RGB images of RAP particles from six different sources with sizes between 1/4 in. (6.35 mm) and 1/2 in. (12.5 mm). The influence of asphalt coating percentage on the RAP particle size and shape properties are quantified in this paper. Then, using the advanced color image thresholding scheme incorporated in the E-UIAIA, the corresponding segmented binary images of RAP particles are generated. A newly defined image mean property is used as an automatic variable threshold limit to segment the bright areas in the associated grayscale version of RAP images to detect the uncoated areas on each particle. A relationship was found between the results of the proposed image processing technique in terms of asphalt coating percentages and the asphalt content of the RAP. Furthermore, the asphalt surface coating percentages could be successfully correlated to the fracture energies of concrete specimens containing these RAP particles blended with other virgin aggregates.