Pattern recognition techniques applied to rust classification in steel structures

Zhang, Yun (2012). Pattern recognition techniques applied to rust classification in steel structures. (Unpublished Masters thesis, City University London)

[img]
Preview
PDF
Download (6MB) | Preview

Abstract

The life and performance of steel structure depends directly upon the steel surface preparation. The restoration of steel structure such as steel bridges, ships and storage tanks is due mainly to the use of manual surface inspection methods accompanied by surface preparation technologies. It requires a long project duration, high costs and hazardous practices for both worker and environment to complete surface restoration.

The developments of surface preparation technologies make it essential to develop technologies that allows patch restore of corrode steel structure in practice.

This thesis addresses the problem of classification of rust steel surfaces. Various Pattern recognition methods are studied for classifying less subjective steel surfaces from a time corrosion perspective. Our primary contribution is: with appropriate features from the steel surfaces, artificial neural network pattern recognition methods have the abilities to classify the less subjective rust steel surfaces reliably and be suitable for automation. The results provide important information about the classification methods for rust steel surface analysis.

Item Type: Thesis (Masters)
Uncontrolled Keywords: pattern recognition, rust, classification, steel images
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
T Technology > TA Engineering (General). Civil engineering (General)
Divisions: School of Informatics > Department of Computing
URI: http://openaccess.city.ac.uk/id/eprint/3007

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year

View more statistics