Header menu link for other important links
An efficient image interpolation algorithm based upon the switching and self learned characteristics for natural images
S.P. Jaiswal, V. Jakhetiya,
Published in IEEE
Pages: 861 - 864
This paper presents a new image interpolation technique for enhancement of spatial resolution of images. The proposed algorithm uses the switching of existing Soft-decision Adaptive Interpolation (SAI) algorithm and Single Pass Interpolation Algorithm (SPIA) methods. We learn the error pattern in the interpolation process of SAI method and SPIA Method after interpolating downsampled version of LR image. Then we deviced a mechanism to correct the error pattern. Emperically we found that SAI methods works better on smooth images (variation among the pixels is less) while SPIA method works better on detailed images (more variation among the pixels), because of the type of pixels used in the interpolation. So, a hybrid scheme of combining SAI method and SPIA method is proposed for best prediction of high resolution (HR) image. The proposed algorithm produces the best results in different varieties of images in terms of both PSNR measurement and subjective visual quality. © 2011 IEEE.
About the journal
JournalData powered by TypesetProceedings - IEEE International Symposium on Circuits and Systems
PublisherData powered by TypesetIEEE
Open AccessNo