1. Design and analysis of wavelet based sharpening filter for interpolated images
Authors : F. Taher, A. Al-maazmi, Alavi Kunhu, H. Al-ahmad
Pages : 57-72
DOI : http://dx.doi.org/10.21172/1.131.11
Keywords : wavelet, satellite, images, resolution, enhancement. Abstract :Satellite images have been used in many applications and fields of research. One of the major limitation of satellite image is their resolution. This paper proposes a new satellite image resolution enhancement algorithm, based on the design and analysis design of new sharpening filters, by obtaining their discrete wavelet transform to improve interpolated images. The sharpening filters are intended for a particular image by means of reducing the image resolution, and then enlarging it. The enlarged interpolated image and the ground truth image are decomposed into four wavelet components LL, LH, HL, and HH. These four components are compared peer to peer and used to design and obtain the sub-optimum sharpening filter kernels that maximize the peak signal to noise ratio, and ultimately, create a super resolution image. Proposed algorithm is based on cross shaped sharpening filters and the performance of the algorithm evaluated using the peak signal to noise ratio (PSNR). The evaluation process compared the wavelet sharpened interpolated images to the ground truth images. The resulted output images showed in a significant improvement compared with interpolation methodology only, without losing any information content present in the satellite image.
Citing this Journal Article :F. Taher, A. Al-maazmi, Alavi Kunhu, H. Al-ahmad, "Design and analysis of wavelet based sharpening filter for interpolated images", Volume 13 Issue 1 - March 2019, 57-72
Click here to Submit Copyright Takedown Notice for this article.