A color image super resolution approach based on alpha channel

doi:10.3969/j.issn.1000-1565.2018.03.014

Abstract

Abstract: There is a close correlation between the three channels of a color image. The existing edge preserving super-resolution method for grayscale image can not be extended directly to RGB space, because it is easy to break the relevance and cause color artifacts. A new method of color image super resolution based on alpha channel is proposed to solve this problem. Firstly, the image is segmented by region, and the foreground, background and alpha channel of each region are extracted. Then, the alpha channel, foreground and background images are super-resoluted respectively. Finally, the high resolution image is synthesized again. The alpha channel is used to coordinate the edge information of three channels. The experimental results show that the method can keep the color edge of the image better and has better visual effects.

Key words: color image super resolution, alpha channel, edge preserving, image segmentation

CLC Number:

TP391

ZHAO Hong, WEI Yonggang,YANG Gang. A color image super resolution approach based on alpha channel[J]. Journal of Hebei University (Natural Science Edition), 2018, 38(3): 321-326.

References

[1] 杨文峰, 郑洁莹, 干宗良,等, 基于邻域嵌入的彩色图像超分辨率重建[J].计算机技术与发展,2015,25(6):25-28. DOI:10.3969/j.issn.1673-629X.2015.06.006. YANG W F,ZHENG J Y,GAN Z L,et al. A super-resolution reconstruction algorithm for color images based on neighbor embedding[J]. Computer Technology and Development, 2015,25(6):25-28. DOI:10.3969/j.issn.1673-629X.2015.06.006.
[2] LI J M, QU Y, LI C H, et al. Image super-resolution base on multi-kernel regression[J]. International Journal of Multimedia Tools and Applications,2016,75:4115-4128. DOI:10.1007/s11042-015-3016-4.
[3] TIMOFTE R D V, GOOL L V. Anchored neighborhood regression for fast example-based superresolution[Z]. IEEE International Conference on Computer Vision, Sydney, Australia, 2013. DOI:10.1109/ICCV.2013.241.
[4] SHEN M F, ZHANG L S, FU H Z. Color image super-resolution reconstruction based on sparse representation[J]. International Journal of Applied Mechanics and Materials, 2013, 278: 1221-1227. DOI:10.4028/www.scientific.net/AMM.278-280.1221
[5] YEGANLI F, NAZZAL M, UNAL M, et al. Image super-resolution via sparse representation over multiple learned dictionaries based on edge sharpness[J]. International Journal of Signal, Image and Video Processing,2016,10:535-542. DOI:10.1007/s11760-015-0771-7.
[6] CHEN H Y, LEOU J J. Saliency-directed color image interpolation using artificial neural network and particle swarm optimization[J]. Journal of Visual Communication and Image Representation, 2012, 23(2): 343-358. DOI:10.1016/j.jvcir.2011.11.006.
[7] KIKU D, MONNO Y, TANAKA M. Minimized-Laplacian residual interpolation for color image demosaicking[J].Proceedings of SPIE, 2014, 9023(1):2304-2308. DOI: 10.1117/12.2038425.
[8] CHEN X,JEON G,JEONG J. Voting-based directional interpolation method and its application to still color image demosaicking[J]. IEEE Transactions on Circuits and Systems for Video Technology,2014,24(2):255-262. DOI: 10.1109/TCSVT.2013.2255421.
[9] 周登文, 申晓留. 边导向的双三次彩色图像插值[J].自动化学报,2012,38(4):525-530. DOI 10.3724/SP.J.1004.2012.00525. ZHOU D W,SHEN X L. Edge-directed bicubic color image interpolation[J]. Acta Automatica Sinica, 2012,38(4):525-530. DOI 10.3724/SP.J.1004.2012.00525.
[10] 沈红斌, 杨杰, 刘小军,等. 基于模糊信息增益的图像相关性度量[J].上海交通大学学报,2006,40(3):466-470. DOI:10.3321/j.issn:1006-2467.2006.03.023. SHEN H B, YANG J, LIU X J,et al.A new image correlation analysis based on fuzzy information gain[J]. Journal of Shanghai Jiaotong University, 2006,40(3):466-470. DOI:10.3321/j.issn:1006-2467.2006.03.023.
[11] 曹文伦, 彭国华, 秦洪元,等. 利用色彩分量相关性的彩色图像分形编码方法[J].计算机工程与应用,2004,40(22):51-55. DOI:10.3321/j.issn:1002-8331.2004.22.017. CAO W L,PENG G H,QIN H Y,et al. Color image fractal compression coding using the correlation of color components[J]. Computer Engineering and Applications, 2004,40(22):51-55. DOI:10.3321/j.issn:1002-8331.2004.22.017.
[12] PORTER T, DUFF T. Compositing digital images[Z]. The 11th Annual Conference on Computer Graphics and Interactive Techniques, New York, USA:ACM, 1984:253-259. DOI: 10.1145/964965.808606.
[13] APOSTOLOFF N E, FITZGIBBON A W. Bayesian video matting using learnt image priors[Z]. 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Washington, DC, USA, 2004. DOI:10.1109/CVPR.2004.50.
[14] CHUANG Y, AGARWALA A, CURLESS B, et al. Video matting of complex scenes[J]. Proceedings of ACM SIGGRAPH, New York, NY, USA:ACM, 2002,21(3):243-248. DOI:10.1145/566654.566572.
[15] CHUANG Y, CURLESS B, SALESIN D,et al. A bayesian approach to digital matting[Z]. The 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Kauai, HI, USA, 2001. DOI:10.1109/CVPR.2001.990970.