Image Super Resolution Using Direction Lifting Schemes

Abstract

Now a day, Image super resolution is in great demand as of its use in various applications such as high definition television broadcasting, video conferencing, medical imaging and satellite imaging. To achieve super resolution one has to design the appropriate equipment so that the resolution of a particular image can be enhanced to a large extent. But due to various influencing parameters such as cost, speed and storage requirements, it is not easy to achieve high resolution up to a large extent by using the digital equipment like camera. In order to meet the customers and researchers demands for the high resolution image, designing of an image super resolution algorithm is necessary. Image super resolution is the process of constructing a high resolution image of a given low-resolution image or from a set of low resolution images for compensating the losses by increasing the number of high frequency components followed by the enhancement of edges. Our main aim is to develop an image super resolution algorithm by lowering the computational complexity and decreasing the cost. In this work, various super resolution algorithms are proposed which produces better super resolved images. In this dissertation, there are total five chapters. In Chapter 1, introduction related to super image resolution is discussed. In Chapter 2, literature survey related to image super resolution and direction lifting schemes is discussed. In Chapter 3, an image super resolution approach is proposed using the interpolation and star polygon direction lifting scheme. In Chapter 4, image super resolution based on direction lifting, SWT and mid-point algorithm is proposed followed by the proposed approach using fusion of images in Chapter 5. Chapter 6 concludes the dissertation and the future scope is discussed.