Author

Riaz Ahmed

Date of Award

12-2004

Degree Name

Master of Science in Engineering

Department

Electrical and Computer Engineering

First Advisor

Dr. Ikhlas Abdul-Qader

Second Advisor

Dr. R. Gejji

Third Advisor

Dr. Osama Abudayyeh

Fourth Advisor

Dr. M. Niewiadomska-Bugaj

Access Setting

Masters Thesis-Open Access

Abstract

An algorithm to extract features from digital images such as cracks in concrete bridges for automated inspection is developed. The algorithm is based on the use of Principal Component Analysis (PCA). In general, PCA is used to reduce the dimensionality of a data set that consists of a large number of interrelated variables while retaining the variation present in the original data set. This ability of PCA will be used in this project to identify clusters using two training sets of images. Linear structure modeling is used to emphasize the image features (cracks) prior to applying PCA. Whole images processing and block processing are used with different distance measures to optimize the accuracy of the results. Real concrete bridge images with cracks. of different sizes and shapes as well as non-cracked images are used. The algorithm development and experimental results are presented.

Share

COinS