Accomplishments

Comparative study of Wavelet methods for Resolution Enhancement of Satellite images
- Abstract
Today various fields require satellite images for their research work, but resolution is the main drawback of satellite images. Some important transform domain methods such as stationary wavelet transform (SWT), Discrete Wavelet Transform (DWT) and proposed method i.e. Integer Wavelet Transform (IWT) are presented to solve the resolution problem of satellite images. The proposed method uses the integer coefficient of the image. Here integer coefficient of IWT is processed in proposed method. The low resolution input is image is decomposed into sub band images such as LL, LH, HL and HH. The high frequency components of IWT are interpolated with factor 2 for resizing sub band images. To estimate the sub band images by mean of the high frequency sub band images of IWT. The interpolation factor alpha/2 applied on estimated images and low resolution input image and combine both images with Inverse Lifting wavelet transform (ILWT) to get high resolution image. The proposed method is tested and compared with different categories of satellite images that are LANDSAT8, LANDSAT7 and IKONOS. The proposed method produces sharper and more detailed image. The evaluation parameters PSNR, RMSE, MAE and MSE show the superiority of the proposed method compared to other transform domain methods