Image analysis technique for aggregate morphology analysis with two-dimensional fourier transform method

Wenjuan Sun, Linbing Wang, Erol Tutumluer

Research output: Contribution to journalArticlepeer-review

Abstract

This paper presents some results of the NCHRP 4-34 Laser Detection and Ranging project. A high-resolution Fourier transform interferometry (FTI) system was developed for aggregate morphology characterization with the use of a fiber-optic coupler to form a Young's double-pinhole interferometer. The FTI system used a charge-coupled device camera to capture fringe images of aggregates with a size ranging from 3/4 in. to No. 50 and to reconstruct three-dimensional surfaces of aggregates on the basis of fringe distortions from the captured digital images. A program was developed in MATLAB to quantify morphologic characteristics of aggregates from a variety of sources. The output results include sphericity, flatness ratio, elongation ratio, angularity, and texture for each individual aggregate particle. In this paper, only the morphologic characteristics of seven types of coarse aggregate with sizes ranging from 3/4 in. to No. 4 are presented. These aggregates were also analyzed with the Aggregate Image Measurement System (AIMS) II and the University of Illinois aggregate image analyzer (UIAIA) for further comparison. The FTI system is easy to operate and can rapidly quantify morphologic characteristics of aggregates. The FTI results of these aggregates are generally consistent with the results of AIMS II and UIAIA with regard to roughness rankings despite minor differences for several aggregates.

Original languageEnglish (US)
Pages (from-to)3-13
Number of pages11
JournalTransportation Research Record
Issue number2267
DOIs
StatePublished - Aug 20 2012

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Mechanical Engineering

Fingerprint Dive into the research topics of 'Image analysis technique for aggregate morphology analysis with two-dimensional fourier transform method'. Together they form a unique fingerprint.

Cite this