Implementation of Proposed of Examplar Based Image with Noise Removal Filters

Abstract

Restoration means filling of corrupted/edited areas/regions of an image or video in such a way that the
modified region(s) is visually agreeable to human eye. Restoration technique has many applications such as, object
removal in digital photos, removal of occlusions (date ,stamps ,logo etc.), such as large unwanted regions, red eye
correction, super resolution, restoration of old films and paintings etc. Inpainting can also be use to create special
effects by removing unwanted object or text from the image. A digital image is essentially a matrix of numerical
values, where each value represents the color or grayscale component Cracks and holes within the image are
denoted by unknown values within the matrix. Image Restoration fills in these unknown values by the support of the
values of known nearby pixels. To achieve our objectives I have used noise removal method which will help us to
complete the restoration task. Here my base method is exampler based Texture Synthesis image restoration
approach but I have used image noise removal filters also to improve the quality of resulted image and later the
results are compared with previous exampler method and my results shows that my proposed technique is better
than the previous approach in terms of MSE, PSNR and time consumption