International Journal of Engineering Research
  • Year: 2014
  • Volume: 3
  • Issue: 9

A New Osrad Filter for Despeckling of Medical Ultrasound Images

  • Author:
  • O. Saraniya1, M. Ezhilarasi2
  • Total Page Count: 6
  • Page Number: 549 to 554

1Dept of Electronics and Communication Engineering, Government College of Technology, Coimbatore, India

2KGISL Institute of Technology, Coimbatore, India

Online published on 8 November, 2017.

Abstract

Speckle noise is an inherent nature of ultrasound images, which results in poor image interpretation and diagnosis. Oriented speckle reducing Anisotropic diffusion filter improves the speckle reduction and edge preservation in these images. In regions of weak edges, the diffusion coefficient is less sensitive. An ideal diffusion coefficient should act so as to diffuse homogeneous regions and sensitive to the edge region which could be well realized by a sigmoid function. In this paper, we propose a new OSRAD filter based on sigmoid function. The diffusion coefficient is chosen to be uniform in the homogeneous region, and decreases rapidly in the transition region which improves weak edges and details. The sigmoid function based OSRAD performs directional filtering of the image along the structures which reduces the blocking effect of diffusion coefficient and enhances the speckle suppressing ability. The proposed filter gives better performance in edge preservation and speckle noise reduction in comparison to the existing diffusion filters.

Keywords

Anisotropic diffusion, Sigmoid function, adaptive median filter