Multiresolution Analysis Driven Corrosion Detection on Metal Surface

Corrosion is an important defect on the surface of a container. In order to improve the efficiency of defectdetection, this paper presents a multiresolution analysismethod. We analyze the texture and color characteristics. The image is divided into non-overlapping blocks of size 8×8 pixels. For each block, db6 wavelet is applied to decompose it into subimages in various scale and wavelet spaces. Then, feature vector is constructed to describe the texture and hue information according to the energy in each block. Finally, k-mean classifier is trained using prior knowledge, and blocks are classified into corrosion and corrosion free classes based on the nearest neighbor classifier. The tests performed on real experiments confirm the validity of our approach.