Color Image Segmentation Via Wavelet Frames

Image segmentation aims at partitioning an image into its constituent parts, which plays a crucial role in practical applications. In this paper, we present a wavelet frame-based model for color images segmentation, which can be regarded as a discretization to the classical Chan-Vese (C-v) model. The advantage of the wavelet frame-based approach is that it has fast algorithm and is able to extract important features of the input images. We then apply the alternating direction method of multipliers (ADMM) algorithm to solve the model. The experiments on some color image segmentation tasks indicate that our algorithm performs favorably against several existing methods.