Field Evaluation of Ballast Fouling Conditions Using Machine Vision

TRB's Innovations Deserving Exploratory Analysis (IDEA) Program Safety 27: Field Evaluation of Ballast Fouling Conditions Using Machine Vision introduces a machine vision-based inspection system to assess ballast degradation levels for an in-service rail track. This system incorporates an approach to ballast sampling, procedures for image acquisition of horizontal and vertical ballast cross sections, and proposes an image-based index called “Percent Degraded Segments” (PDS) that is defined in terms of image segments. Use of a machine vision-based inspection system may reduce the cost, labor, and time used with traditional ballast sampling and sieving processes. This system may help provide more consistent readings that are not dependent on the level of track inspectors’ expertise or the locations along the track where ballast samples are obtained.