Two-stage image segmentation by adaptive thresholding and gradient watershed

Many images contain the coarse pieces and fine particles. Hence it is often important to segment the coarse pieces and fine particles using different segmentation techniques. In this paper, we combine adaptive thresholding and watershed transform in a two-stage image segmentation method, which has been used for oil-sand size analysis. We also present an image-retrieval-like approach to speed up size estimation of the fine particles.