Morphological segmentation and classification of underground pipe images

TitleMorphological segmentation and classification of underground pipe images
Publication TypeJournal Article
Year of Publication2006
AuthorsSinha, S. K., and P. Fieguth
JournalMachine Vision and Applications
Pagination21 - 31
KeywordsClassification, Inspection, Morphology, Segmentation

Visual inspection based on closed circuit television surveys is used widely in North America to assess the condition of underground pipes. Although the human eye is extremely effective at recognition and classification, it is not suitable for assessing pipe defects in thousand of miles of pipeline because of fatigue, subjectivity, and cost. In this paper, simple, robust, and efficient image segmentation and classification algorithm for the automated analysis of scanned underground pipe images is presented. The experimental results demonstrate that the proposed algorithm can precisely segment and classify pipe cracks, holes, laterals, joints and collapse surface from underground pipe images