Cascading the Images based on S-R Algorithm

Teja Dhanawade, Varsha Bobade

Abstract: Image Inpainting is the art of filling in lost portions of an image using background details in a visually disingenuously smooth way. When an image fills missing region sometimes it cannot fill a coarse version of the input mage at that time it first in-painted by a non- parametric patch sampling. If we compare with an existing system, it introduce blurring, resolution decimation, and corruption by noise. We are introducing new algorithm which converts the original image into low resolution image then it stores the patches into the database and then apply to an image. The input image allows reducing the computational complexity in low version of Inpainting, in our proposed algorithm image is less sensitive to noise and it works with the dominant orientations of image structures. From the low-resolution in painted image, a single-image super-resolution is applied to recover the details of missing areas.

Keywords: Blurring, Decimation, Inpainting, Patch Sampling, Resolution