Denoising of Medical Images Using Total Variational Method
V. Naga Prudhvi Raj · Signal & Image Processing An International Journal · 2012
Feature extraction and object recognition from images acquired by various imaging modalities are playing the key role in diagnosing the various diseases. These operations will become difficult if the images are corrupted with noise. So the need for developing the efficient algorithms for noise removal became an important research area today. Developing Image denoising algorithms is a difficult operation because fine details in a medical image embedding diagnostic information should not be destroyed during noise removal. In this paper the total variational method which had success in computational fluid dynamics is adopted to denoise the medical images. We are using split Bregman method from optimisation theory to find the solution to this non-linear convex optimisation problem. The present approach will outperform in denoising the medical images while compared with the traditional spatial domain filtering methods. The performance metrics we used to measure the quality of the denoised images is PSNR (Peak signal to noise ratio).The results showed that these methods are removing the noise effectively while preserving the edge information in the images.