Performance evaluation of satellite image resolution enhancement techniques based on wavelet transform

The Resolution Enhancement (RE) algorithms for satellite images have been proposed to overcome the drawbacks (losing high frequency content) of interpolation. In this paper, the Dual Tree Complex Wavelet Transform (DTCWT) based algorithm has been implemented where DTCWT is used to decompose the image into its subbands. Lanczos Interpolation is performed to increase the resolution of the subband images. To remove the artifacts generated in the image due to interpolation, Non Local Means filtering (NLM) is used. The High Resolution (HR) image is obtained by using the Inverse DTCWT which reconstructs the image from the subbands. Peak Signal to Noise Ratio (PSNR), SSIM (Structural Similarity Index) and Quality Index (Q-Index) are calculated for a database of 40 grayscale images with the resolution of 256x256. These parameters are used to evaluate the performance of the implemented techniques. These results along with the visual results prove the superiority of the DTCWT based algorithm with NLM filtering.