Image Up-sampling for Enhancing Quality Image with Deep Learning

Abstract

This paper introduces an approach to deep learning for improving quality for single high-definition (H.D) picture. The method immediately acquires an end-to-end map from low- and high definition pictures. High image resolution plays a significant role when enhancing the photo as well as frames for video in gadgets including smartphones as well as computers for Better Resolution. When upscaling the image frame, the image get distorted and blurred. A high-definition image with greater clarity is produced by using a deep convolutional neural network which accepts a low-definition image as input and using Discrete Cosine Transformation to create blurred image from low resolution image and subtracting this blurred image from the result of network model . The experimental result shows that the proposed algorithm works better than the other compared algorithms.