Image Restoration using Hybrid Super Resolution Error Model

Image restoration is the method of restoring the corrupted images from the degradation. The degradation can happen due to some external and internal noise like blur, relative motion between the camera and the object whose image is to be captured and many more. Super resolution helps in increasing the resolution of the images. Therefore, image restoration used along with the super resolution not only helps in recovering the image from any kind of degradation but also increases the quality of the image to be restored. In this paper, a hybrid model has been proposed and implemented that provides the above said features by defining Gaussian and Laplacian noise models for the medical images, so that better image quality can be achieved. The image quality is determined with the help of performance parameters like PSNR, SSIM, MSE and Correlation, and then is compared with Wavelet Transform and Median Filter. The analysis shows that better results are achieved using the proposed super resolution model.