TY - JOUR
T1 - Image resolution enhancement using edge extraction, sparse representation and interpolation in wavelet domain
AU - Chavez-Roman, H.
AU - Duchen-Sanchez, G.
AU - Kravchenko, V.
AU - Ponomaryov, V.
PY - 2013
Y1 - 2013
N2 - Over the past few years, high-resolutions (HR) are desirable or essential, e.g., in online video systems, and therefore much has been done to achieve an image of higher resolution from the corresponding low resolution (LR) images. This procedure of recover or rebuild is called single image super-resolution (SR). This study addresses the problem of generating a SR image from a LR input image in the wavelet domain. In order to achieve a sharper image, an intermediate stage for estimating the high-frequency (HF) sub-bands has been proposed. It includes an edge preservation procedure and mutual interpolation between the input LR image and the HF sub-band images performed via the discrete wavelet transform (DWT). Sparse mixing weights are calculated over blocks of coefficients in an image, providing a sparse signal representation in the LR image. All sub-band images are used to generate the new HR image employing the inverse DWT. Experimental results have shown that the proposed approach outperforms existing methods in terms of objective criteria and subjective perception via human visual system, improving the image resolution.
AB - Over the past few years, high-resolutions (HR) are desirable or essential, e.g., in online video systems, and therefore much has been done to achieve an image of higher resolution from the corresponding low resolution (LR) images. This procedure of recover or rebuild is called single image super-resolution (SR). This study addresses the problem of generating a SR image from a LR input image in the wavelet domain. In order to achieve a sharper image, an intermediate stage for estimating the high-frequency (HF) sub-bands has been proposed. It includes an edge preservation procedure and mutual interpolation between the input LR image and the HF sub-band images performed via the discrete wavelet transform (DWT). Sparse mixing weights are calculated over blocks of coefficients in an image, providing a sparse signal representation in the LR image. All sub-band images are used to generate the new HR image employing the inverse DWT. Experimental results have shown that the proposed approach outperforms existing methods in terms of objective criteria and subjective perception via human visual system, improving the image resolution.
KW - Edge extraction
KW - Interpolation
KW - Sparse mixing estimators
KW - Super resolution
KW - Wavelet domain
UR - http://www.scopus.com/inward/record.url?scp=84887740265&partnerID=8YFLogxK
U2 - 10.1615/TelecomRadEng.v72.i19.80
DO - 10.1615/TelecomRadEng.v72.i19.80
M3 - Artículo
SN - 0040-2508
VL - 72
SP - 1803
EP - 1820
JO - Telecommunications and Radio Engineering (English translation of Elektrosvyaz and Radiotekhnika)
JF - Telecommunications and Radio Engineering (English translation of Elektrosvyaz and Radiotekhnika)
IS - 19
ER -