Abstract

Ink-jet printing image denoising is a crucial step to increase image quality and to improve the performance of all the tasks needed for quantitative ink-jet imaging analysis. In this paper, we present a novel image denoising strategy based on shape prior technology for Beltrami manifold representation. The main idea of the proposed algorithm is to minimize Beltrami manifold that includes additional information on shape references. A variational formulation of Beltrami manifold with a data fidelity term helps preserving local image structures such as edges and textures. The shape prior is incorporated with a family of rotation and scale invariants from shape templates to guide the evolving curve to approaching desired features. The main advantage of the proposed algorithm is that it has both good noise reduction and edge protection capabilities. Experimental results show that the proposed algorithm can achieve better results in preserving the texture and details for ink-jet printing image denoising.

Paper Details