Railroad track substructure consisting of the ballast, subballast, and subgrade plays an important role in the overall performance of a railway track system in response to repeated train loading. Current approaches for determining ballast condition based on visual inspection of the track are mostly qualitative and subjective in nature. This research project introduced a new degradation index, Percent Degraded Segments (PDS), based on the machine vision algorithms developed to analyze in-service ballast cut section images and effectively quantify the level of ballast degradation in the field. The performance of the PDS index was validated using a variety of ground truth data. A Graphical User Interface (GUI) was also developed to assist users in processing the acquired images and finding the best image processing and segmentation parameters. Finally, a statistical approach was used to relate PDS and the commonly used Selig’s Fouling Index (FI) through regression analysis.

The developed technology can potentially replace the current state-of-the-practice of visual inspection, sampling, and mechanical sieve analysis. It can quantify in-service ballast condition and its properties at any location, possibly identified by ground penetrating radar or other network condition monitoring devices, without the need for ballast sampling from trench cut sections. It can also be used with shoulder cleaning and undercutting equipment to automate the condition assessment using images of ballast cut sections below the ties. As such, this automated evaluation could greatly improve the quality and efficiency of ballast maintenance activities. In addition, the results of this process can be used for inspection purposes and to map out recommendations of follow-up rehabilitation strategies. Further, the proposed method has the potential to be applied for in-situ evaluation of permeability and strength properties of railroad ballast at different degradation levels. As a component of a comprehensive Ballast Management System (BMS), the developed technology would help to evaluate designs and deterioration mechanisms of ballasted track and provide predictive service life and life-cycle analysis for improving the safety and network reliability of the U.S. railroad transportation system.

The contractor's report is available. https://www.trb.org/Main/Blurbs/176825.aspx