Image resolution enhancement using hybrid wavelets and bicubic interpolation

Abstract

Digital Image Processing is a rapidly evolving field with growing applications in Science and Engineering. The need of digital image processing is motivated due to the improvement of pictorial information for human interpretations, for autonomous machine applications, for efficient storage and transmission. Broadly, image processing may be classified into following categories: restoration, segmentation, and enhancement. Image enhancement techniques are used to emphasize and sharpen image features for display and analysis. A lot of techniques are available for the improving the quality of an image such as contrast stretch, density slicing, spatial filtering, Edge enhancement, and image interpolation which has been discussed by the authors [4], [30], [46]. In the last few years, Wavelet transforms have been found as most important tool for image processing, especially for the applications such as time-frequency analysis, segmentation, enhancement and data compression. The thesis presents the theory of fundamental mathematical tools (Discrete wavelet transform, and Stationary wavelet transform) for image resolution enhancement. First method describes a super resolution technique based on interpolation of the high-frequency subband images obtained by DWT and the input image. DWT is used to decompose an image into different subband images. Then the high-frequency subband images and the low-resolution input image have been interpolated, followed by combining all these images to generate a new super resolved image by using IDWT. Second method is based on interpolation of the high frequency subband images obtained by DWT and the input image. The edges are enhanced by introducing an intermediate stage by using SWT. DWT is applied in order to decompose an input image into different subbands. Then the high frequency subbands as well as the input image are interpolated. The estimated high frequency subbands are being modified by using high frequency subband obtained through SWT. Then all these subbands are combined to generate more enhanced image by using IDWT. These two approaches also turned out to be extremely efficient for preservation of edge. The performance of these image resolution enhancement techniques are quantitatively accessed using different criteria namely the MSE, PSNR and the visual appearance. It has been observed that these techniques give better PSNR results. Hybrid wavelet and bicubic interpolation provide significant improvement in texture based images. An image zooming result shows that hybrid wavelets and bicubic interpolation gives less aliasing effects as compared to DWT and bicubic interpolation, but gives less PSNR in images having pixels at same intensity level as compared to DWT and bicubic interpolation.