Medical images have always been an important factor in diagnosis of disease. Poisson Noise in those images has always been a problem with the image clarity. We propose two technique which combines Multi-Scale Variance Stabilizing Transform (MS-VST), Fast Discrete Curvelet Transform (FDCT) with Thresholding and MS-VST, FDCT with Null Hypothesis testing for effectively removing the Poisson Noise from the medical images. The effectiveness of using these techniques has been analyzed using Peak Signal to Noise Ratio and Universal Image Quality Index.

/pdf/poisson-noise-removal-in-biomedical-imagesusing-non-linear-4k0pj168kc.pdf

Poisson Noise Removal in Biomedical Imagesusing Non-Linear Techniques

We analyse two methods of removing poisson noise from images using a bilateral filter and by Fast discrete Curvelet Transform (FDCT). The Variance stabilizing transform (VST) is the main feature of the noise removal as it converts the Poisson distribution to the Gaussian domain, which makes the noise removal process relatively simple. Once the Gaussian distribution is obtained, the bilateral filter (BF) can be used for removing noise. We can also use the FDCT instead of bilateral filter, as it is capable of sparse representation of image intrinsic features. We implement both the methods separately, compare them and demonstrate simulations for monitoring their effectiveness in poisson noise removal. The results show that FDCT is more efficient for preserving image features, while bilateral filter is much faster and simple to implement.

/pdf/bilateral-filter-approach-and-fast-discrete-curvelet-3c4a6q8sf9.pdf

Jilu George

Papers

Bilateral Filter Approach and Fast Discrete Curvelet Transform for Poisson Noise Removal from Images