A wavelet-based watershed image segmentation for VOP generation

This paper presents a wavelet-based watershed image segmentation method for extracting video objects from a video sequence. The method is based on a multiresolution application of a wavelet and watershed transformation, followed by a wavelet coefficient-based region merging procedure. The procedure toward complete segmentation consists of four steps: pyramid representation, region segmentation, region merging, and region projection. First, the pyramid representation creates multiresolution images using a wavelet transform. Next, the region segmentation is used to segment the lowest-resolution image of the created multiresolution image. Then, the region merging is merged to segmented regions using the third-order central moment values of the wavelet coefficients. Finally, the region projection is used to recover a full-resolution image by inverse wavelet transform. Experimental results of the presented method can be applied to the segmentation of noise or degraded images and reduction of over-segmentation.