Edge Model Based High Resolution Image Generation
The present paper proposes a new method for high resolution image generation from a single image. Generation of high resolution (HR) images from lower resolution image(s) is achieved by either reconstruction-based methods or by learning-based methods. Reconstruction based methods use multiple images of the same scene to gather the extra information needed for the HR. The learning-based methods rely on the learning of characteristics of a specific image set to inject the extra information for HR generation. The proposed method is a variation of this strategy. It uses a generative model for sharp edges in images as well as descriptive models for edge representation. This prior information is injected using the Symmetric Residue Pyramid scheme. The advantages of this scheme are that it generates sharp edges with no ringing artefacts in the HR and that the models are universal enough to allow usage on wide variety of images without requirement of training and/or adaptation. Results have been generated and compared to actual high resolution images.
Index terms: Super-Resolution, edge modelling, Laplacian pyramids.
KeywordsSharp Edge Ringing Artefact Bicubic Interpolation Laplacian Pyramid Edge Model
Unable to display preview. Download preview PDF.
- 2.Nguyen, N., Milanfar, P.: An efficient wavelet-based algorithm for imaqe superresolution. In: Proc. Int. Conf. Image Processing, vol. 2, pp. 351–354 (2000)Google Scholar
- 3.Tsai, R., Huang, T.: Multiple frame image restoration and registration. In: Advances in Computer and Image Processing, pp. 317–339. JAI Press Inc., CT (1984)Google Scholar
- 5.Chaudhuri, S. (ed.): Super-Resolution Imaging. Kluwer Academic, Norwell (2001)Google Scholar
- 6.Park, S.C., Park, M.K., Kang, M.G.: Super-resolution image reconstruction: A technical overview. IEEE Signal Processing Magazine, 21–36 (2003)Google Scholar
- 7.Joshi, M., Chaudhuri, S.: Super-resolution imaging: Use of zoom as a cue. In: Proc. ICVGIP, Ahmedabad, India (2002)Google Scholar
- 8.Joshi, M., Chaudhuri, S.: Zoom based super-resolution through sar model fitting. In: Proc.Intl Conf. on Image Processing (ICIP), Singapore (October 2004)Google Scholar
- 9.Jiji, C.V., Joshi, M.V., Chaudhuri, S.: Single frame image super-resolution using learnt wavelet coefficients. Intl. J. Imaging Science & Tech (special issue on high resolution image reconstruction) 14, 105–112 (2004)Google Scholar
- 10.Jiji, C.V., Joshi, M.V., Chaudhuri, S.: Single frame image super-resolution through contourlet learning. EURASIP J. Applied Signal Processing, 1–11 (2006)Google Scholar
- 11.Chaudhuri, S., Joshi, M.: Motion-Free Super-Resolution. Springer, HeidelbergGoogle Scholar
- 13.Nema, M.K., Rakshit, S.: Edge-model based representation of laplacian subbands. In: Proc. Seventh Asian Conf. on Computer Vision (ACCV 7), Hyderabad, India, January 2006, pp. 80–89 (2006)Google Scholar
- 14.Rakshit, S., Nema, M.K.: Symmetric residue pyramids: An extension to burt laplacian pyramids. In: Proc. IEEE ICASSP, Hong Kong, pp. III–317–III–320 (2003)Google Scholar