Through the experiment, if using the traditional watershed algorithm of cashmere image edge extraction, the results will appear serious ‘over-segmentation’ phenomenon. This paper adopts multi-resolution analysis theory, using wavelet transformation theory establishment multi-resolution analysis framework on image segmentation. Multi-wavelet transform has some unique advantages and good performance on edge detection and texture analysis. This research uses multi-wavelet transform to improve the over-segmentation watershed transformation, to get meaningful image segmentation result. It facilitates the preparation for the extraction of characteristic such as cashmere image diameter, flake highly, and etc.