Sub-pixel edge detection for structural health monitoring

Laurence Pap, Ju Jia Zou, Brian Uy, Yang Xiang

    Research output: Chapter in Book / Conference PaperConference Paperpeer-review

    Abstract

    ![CDATA[In structural engineering it is essential to monitor the health of structural members under various load conditions. Traditionally, engineers are required to travel on site to retrieve various test results. However, in modern times, engineers can remotely monitor test results both statistically and visually with the help of image processing. When image processing is applied to structural health monitoring, one major problem to overcome is the offset of an actual edge location caused by the digitalization of straight lines. It is important to locate the edges of objects as accurately as possible to ensure that correct data is received and correct decisions are made. This paper presents a new edge detection algorithm at the sub-pixel level for the purpose of structural health monitoring. By using sub-pixel edge detection, we can achieve higher precision than other methods, which are limited by the resolution of imaging devices. Non-linear image interpolation is used to increase the precision of determining edges by estimating values which are at undefined points. By using non-linear interpolation, the over smoothing effect produced by linear methods can be overcome. The proposed method, when applied to structural health monitoring, can perform a more accurate and efficient determination of the edges of an object than previous edge detection methods. The proposed method also benefits in images, that are noisy compared to previous edge detection methods. Experiment results on sample objects show the superiority of the proposed method.]]
    Original languageEnglish
    Title of host publicationStructural Health Monitoring of Intelligent Infrastructure: Proceedings of the 4th International Conference on Structural Health Monitoring of Intelligent Infrastructure (SHMII-4), 22-24 July 2009, Zurich, Switzerland
    PublisherEMPA
    Number of pages9
    ISBN (Print)9783905594522
    Publication statusPublished - 2009
    EventInternational Conference on Structural Health Monitoring of Intelligent Infrastructure -
    Duration: 9 Dec 2013 → …

    Conference

    ConferenceInternational Conference on Structural Health Monitoring of Intelligent Infrastructure
    Period9/12/13 → …

    Fingerprint

    Dive into the research topics of 'Sub-pixel edge detection for structural health monitoring'. Together they form a unique fingerprint.

    Cite this