Simultaneous Image Denoising and Curve Extraction by Tensor Voting

In this paper we propose a method that simultaneously performs image denoising and salient curve extraction among random dot patterns using tensor voting. Given an image containing random dots, the pixels are first converted into a set of tokens to be preprocessed by tensor voting, then the voting results are binarized and thinned by a morphological filter to extract the salient curves. At last the method is tested on several real images.