Color Photo Denoising Via Hue, Saturation and Intensity Diffusion

  • Authors:
  • Lei He;Chenyang Xu

  • Affiliations:
  • Information Technology Department, Armstrong Atlantics State University, Savannah, USA 31419;Imaging and Visualization Department, Siemens Corporate Research, Princeton, New Jersey, USA 08540

  • Venue:
  • ICIAR '08 Proceedings of the 5th international conference on Image Analysis and Recognition
  • Year:
  • 2008

Quantified Score

Hi-index 0.01

Visualization

Abstract

This paper presents a new image denoising model for color photo noise removal. Unlike previous approaches, our algorithm is based on separating a color photo into hue, saturation, and intensity (HSI) components, and then processing each component with partial differential equations (PDE)-based diffusion flows. The hue and saturation denoising are implemented by a new weighted orientation diffusion and a modified curvature flow respectively. The intensity denoising is implemented with a PDE that is a combination of a gradient vector flow (GVF)-based filter and a fourth-order PDE filter. This combined technique provides a robust and accurate denoising process, i.e., it preserves edges well and at the same time overcomes the staircase effect in smooth regions. Experiments on color photos demonstrate the improved performance of the proposed model when compared with other recognized approaches and commercial software.