Intelligent computerized system can analyze and process vast amounts of data instantaneously in the civil engineering domain recently. Most previously developed image segmentation methods for steel bridge rust defect assessment could not deal with colour images directly. Their original colour images were first converted to gray scale images and then further processed as gray-scale images. The system was often encountered with difficulties while acquiring digital images under environmental conditions such as non-uniform illuminations.

A rust defect recognition method for bridge surface was proposed by using fourier transform and support vector machine method in order to overcome those influences from environmental conditions. This system uses fourier transform to determine whether rust defects exist in a given digital colour image. If the image is judged as defective, he image is sent to the rust segmentation system to extract the colour of rust and background and make an accurate rust defect assessment. The rust segmentation system was developed by applying support vector machine method.

The proposed methods shows better performance than previous methods such as K-means and NFRA.