Super Resolution of Medical Reconstructed Image Using Wavelet Transform
Muthana N. Abdul-Hussein Al-Hassany, Mohammed H. Ali Al-Hayani · 2018
Super resolution of the image is the process of obtaining high resolution image from low resolution imageShepp-Logan image for the medical image, has taken, of size (256×256), which has down sampled from size (512×512) using Bicubic interpolation, where the anti-aliased filter has used before down sampling process to use it as the input image to the algorithm to enlarge it to the output of the size (512×512) pixels. The results show increasing the PSNR of the magnified image with respect to the original high resolution image. Where we has used the discrete wavelet transform, lifting wavelet transform with stationary wavelet transform, the traditional interpolation (Bicubic, Bilinear...etc.) and second order derivative directional interpolation (ICBI, DCC) has used to enlarge the sub bands, where the results has stated that the proposed schemes has different reconstruction images in both the edges and smooth area. Therefore, we take the less differences with the original reconstructed image and regard it as the super resolution image and denoising the images with FIR filter to reach and note the small effects on the images with the helping of the image quality metrics (PSNR, CORR, MSE, SSIM).