A content adaptive method of de-blocking and super-resolution of compressed images

摘要

In this paper, a new method of image upscaling along with de-blocking of compressed images has been presented. In the case of highly compressed images, there is a high probability that these images may contain the noise in the form of blocking artifacts. In this presented work, a spatial domain-based approach has been suggested with two roles, one of which is to process the image for reduction of compression-based blocking artifacts and other is to upscale the low-resolution image to high-resolution image. Image upscaling is one of the implementation techniques of image super-resolution (SR). It is a type of SR where only a single image-based SR is being implemented. In the proposed technique, image de-blocking along with interpolation based super resolution has been developed in the spatial domain, therefore it is a practical and realistic method. The results of the proposed method in the form of quality metrics like PSNR, MSE and MSSIM have been compared with other methods of interpolation along with de-blocking method.