Application of Multiresolution Pattern Classification to Steel Bridge Coating Assessment

Abstract

Digital image recognition has been broadly utilized in different engineering disciplines; in civil engineering, it has been used experimentally in the inspection of steel bridge coatings and underground sewer systems. The multiresolution pattern classification (MPC) method is an image-recognition technique that extracts features from an image and clusters the image into several featured areas. In this paper, the MPC method will be introduced for steel bridge coating assessment with a detailed description of the theoretical background of MPC. An application combining the MPC method and a double-sampling plan for steel bridge coating assessment will also be presented, followed by the conclusions of this paper.