Comparison of Real and Complex-valued Versions of Wavelet Transform, Curvelet Transform and Ridgelet Transform for Medical Image Denoising

In this study; medical images were denoising with multiresolution analyses using real-valued wavelet transform (RVWT), complex-valued wavelet transform (CVWT), ridgelet transform (RT), real-valued first-generation curvelet transform (RVFG CT), real-valued second-generation curvelet transform (RVSG CT), complex-valued second-generation curvelet transform (CVSG CT) and results are compared. First and second-generation curvelet transformations are used for realvalued curvelet transform as two techniques. For the evaluation of the proposed system, we used 32 lung CT images. These images include 10 images with benign nodules and 22 images with malign nodules. Different types of noise like the Random noise, Gaussian noise and Salt & Pepper noise were added to these images and they are removed separately. The performances of used transforms are compared using Peak Signal to Noise Ratio (PSNR) parameter. Obtained results showed that complex-valued wavelet transform are suited for removal of random noise and Gaussian noise. In case of Gaussian noise in images, PSNRs of first generation curvelet transform and complex-valued wavelet transform are around 33 dB. The ridgelet transform provides high PSNR value (30.4dB) for denoising of salt & pepper noise in images.