Otsu image segmentation algorithm based on morphology and wavelet transformation

In order to overcome the traditional Otsu algorithm weakness in the complex image segmentation, a new segmentation algorithm was introduced in this paper to detect the surface defect of parts, which applied with morphology and wavelet transform into the Otsu algorithm. There were two steps in the algorithm, one was segmentation target from the background and the other was segmentation defect from the target. As the first segmentation, the target image was acquired through separating the target from background image using the morphology method. Then, the monolayer wavelet coefficient was applied to separate the target image. Thirdly, the redundancy information in the image was removed by means of the low frequency-reconstruction. Finally, the one and two dimensional Otsu algorithm were used to segment the defect from part surface. The experiments prove that this new algorithm has higher segmentation accuracy and better noise resistance than that of the traditional one and two dimensional Otsu. Moreover, improved one-dimensional algorithm is superior to that of the two-dimensional ones.