Glassware crack defects detection based on wavelet transform

An algorithm for detecting and extracting crack defects in glassware using wavelet transform is proposed in this paper. Firstly, the canny image segmentation and the local adaptive dynamic threshold segmentation are carried out on the glassware image with unobvious crack defects. Then, the wavelet decomposition is applied separately on the segmented images. And finally the wavelet fusion is used to extract the crack defects. Experiments show that the proposed algorithm works well in detecting glassware crack defects.