Asphalt cores are routinely drilled from existing roadways and manually tested to determine the thickness of individual layers and classify the gradation of the aggregate mixture within each layer. This process is time-consuming, hazardous, and destroys the sample core. This study presents a non-destructive, close-range photogrammetry-based 3D scanning method for determining the layer divisions and aggregate gradation within asphalt cores. The proposed method uses structure-from-motion techniques to produce distortion-free images of the cylindrical surface of the core exposed during drilling. From these images, the asphalt mix gradation is determined from the exposed cross sections of aggregate within the core. Our method achieved a 75% classification accuracy and did not damage the sample, leaving the core intact for other uses. Additionally, we also find that surface image-based methods for gradation curve generation tend to underestimate the amount of smaller aggregate within a mix and show signs of higher variability in detecting the largest sizes of aggregate. This study demonstrates that the close-range photogrammetry-based 3D scanning technology can easily be developed into an automatic and non-destructive tool for asphalt core analysis.
Photogrammetric 3D Scanning of Asphalt Cores for Automatic Layer Detection and Gradation Classification
KSCE J Civ Eng
KSCE Journal of Civil Engineering ; 27 , 8 ; 3542-3554
2023-08-01
13 pages
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
Effect of Gradation on Asphalt Mixture Performance
Transportation Research Record | 1999
|Effect of Gradation on Asphalt Mixture Performance
Online Contents | 1999
|Gradation Effects on Hot-Mix Asphalt Performance
Transportation Research Record | 2001
|Gradation Effects an Hot-Mix Asphalt Performance
Online Contents | 2001
|Effect of Aggregate Gradation on Measured Asphalt Content
Online Contents | 1993
|