Usage
  • 38 views
  • 29 downloads

Image processing-based framework for determining of growth in sewer pipe defects

  • Author / Creator
    Karabij, Mohamed
  • Municipal drainage systems have a key role in public health and are considered one of the main components of every modern city's infrastructure. However, as the drainage system ages, its pipes gradually deteriorate at rates that vary based on the different conditions of utilization. To prevent unexpected failures, municipalities have adopted a proactive approach that relies on regular condition assessments of their assets from which data-driven maintenance, rehabilitation, and replacement plans are developed. From a practical standpoint, this data-driven planning relies on assessment information used in conjunction with deterioration models to evaluate the risk of failure associated with structural and operational anomalies. In this respect, data-driven plans can be very useful in assisting municipalities insofar as budget and process management are concerned. The most popular deterioration models rely on statistical analysis and statistical models whose accuracy depends heavily on the quality and the quantity of data collected as part of underground pipe inspections, e.g. the number of cracks, fractures, roots, etc. However, understanding the development of defects over time will not only help improve the accuracy of the existing deterioration models but will also provide valuable information in terms of understanding the relationship between the various factors affecting defect development and pipe deterioration. This research presents an image registration framework for extracting crack development information from the CCTV videos of sewer pipes. Image processing techniques are used to estimate a relative change in a defect from images taken at two different times to determine the relative growth of the defect. The framework was implemented on two case studies and a visual validation was applied to the results. In this respect, since the camera is not expected to be identical for the image time series, the accuracy of Area Scaler (AS), which is used so as to have a similar scale between the images, is examined. This scaling procedure is illustrated using case studies, containing 49 pairs of images, leading to a 95.5% accuracy.

  • Subjects / Keywords
  • Graduation date
    Spring 2021
  • Type of Item
    Thesis
  • Degree
    Master of Science
  • DOI
    https://doi.org/10.7939/r3-9anz-1s84
  • License
    This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.