Frontiers of Mechanical Engineering
>> 2009,
Volume 4,
Issue 2
doi:
10.1007/s11465-009-0017-x
Detection for transverse corner cracks of steel plates’ surface using wavelet
School of Mechanical and Power Engineering of East China University of Science and Technology, Shanghai 200237, China
Available online: 2009-06-05
Next
Previous
Abstract
An algorithm is presented for detecting transverse corner cracks at a steel plate surface using wavelet transform. According to characteristics of transverse corner crack images, the wavelet transform is used for the multi-scale analysis of detecting the image edges and disintegrating the image into four directions at the same time. The proper threshold value is chosen to segment the image into vertical components to obtain the final detection result. The experiment shows that transverse corner cracks of steel plates can be more effectively extracted by the proposed method than the other two common methods.