Which super-resolution algorithm is proper for Farsi text image sequences

In this paper we propose a new algorithm for super-resolution of Farsi text image sequences. Our algorithm contains three main steps as prior super-resolution algorithms; registration, reconstruction, and restoration. Due to special properties of Farsi texts such as appearance of dots in alphabet, selecting a proper super-resolution algorithm, especially in presence of noise, is more important. We propose an algorithm with an accurate sub-pixel registration and IBP reconstruction that reconstructs a high resolution image from a set of noisy low resolution observations. In restoration step we have exploited NLM algorithm to overcome image noise. We test our algorithm on synthetic and real data. Both quantitative and qualitative results show outperformance of our algorithm.

[1]  Andrew Zisserman,et al.  Super-resolution enhancement of text image sequences , 2000, Proceedings 15th International Conference on Pattern Recognition. ICPR-2000.

[2]  A. N. Rajagopalan,et al.  Superresolution of License Plates in Real Traffic Videos , 2007, IEEE Transactions on Intelligent Transportation Systems.

[3]  Sabine Süsstrunk,et al.  A Frequency Domain Approach to Registration of Aliased Images with Application to Super-resolution , 2006, EURASIP J. Adv. Signal Process..

[4]  Michael Elad,et al.  Generalizing the Nonlocal-Means to Super-Resolution Reconstruction , 2009, IEEE Transactions on Image Processing.

[5]  D. Hayner.  The Missing Cone Problem in Computer Tomography and a Model for Interpolation in Synthetic Aperture Radar. , 1983 .

[6]  Wilfried Philips,et al.  Robust reconstruction of low-resolution document images by exploiting repetitive character behaviour , 2008, International Journal of Document Analysis and Recognition (IJDAR).

[7]  Majid Mirmehdi,et al.  An Introduction to Super-Resolution Text , 2007 .

[8]  David S. Doermann,et al.  Superresolution-based enhancement of text in digital video , 2000, Proceedings 15th International Conference on Pattern Recognition. ICPR-2000.

[9]  Manuel Guizar-Sicairos,et al.  Efficient subpixel image registration algorithms. , 2008, Optics letters.

[10]  A. Bovik,et al.  A universal image quality index , 2002, IEEE Signal Processing Letters.

[11]  Katherine Donaldson,et al.  Bayesian Super-Resolution of Text in Video with a Text-Specific Bimodal Prior , 2005, CVPR.

[12]  J. Fienup,et al.  Efficient Image Registration Algorithms for Computation of Invariant Error Metrics , 2007 .

[13]  J. Morel,et al.  On image denoising methods , 2004 .

[14]  Michal Irani,et al.  Improving resolution by image registration , 1991, CVGIP Graph. Model. Image Process..

[15]  C. V. Jawahar,et al.  Super-Resolution of Text Images Using Edge-Directed Tangent Field , 2008, 2008 The Eighth IAPR International Workshop on Document Analysis Systems.