Historical Document Image Denoising by Ising Model

In this paper, we propose a historical document image denoising method based on Ising-like model and Anisotropic filter. Physical statistics and morphological smooth process are combined together to improve image particulars. We first apply an Ising model to the 8-th bit plane of the original historical document image, then the image energy field and density is through Ising dynamic evolution. After the image is processed, it has been smoothed by Anisotropic morphological for several times. At last, we get the restored image document. We compare the visual quality changes before and after Ising evolution in different Anisotropic morphological condition respectively. Experimental results demonstrate that the Ising evolved image can better restore the local area details of the degraded historical document image.

[1]  Guangtao Zhai,et al.  The impact of phase transition on quality assessment of natural images , 2016, 2016 IEEE Global Conference on Signal and Information Processing (GlobalSIP).

[2]  Fei Peng,et al.  Identifying photographic images and photorealistic computer graphics using multifractal spectrum features of PRNU , 2014, 2014 IEEE International Conference on Multimedia and Expo (ICME).

[3]  Andrei Dmitri Gavrilov,et al.  Convolutional Neural Networks: Estimating Relations in the Ising Model on Overfitting , 2018, 2018 IEEE 17th International Conference on Cognitive Informatics & Cognitive Computing (ICCI*CC).

[4]  K. Shirai,et al.  Character Shape Restoration of Binarized Historical Documents by Smoothing via Geodesic Morphology , 2013, 2013 12th International Conference on Document Analysis and Recognition.

[5]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[6]  Ofer Hadar,et al.  When physics meets signal processing: Image and video denoising based on Ising theory , 2015, Signal Process. Image Commun..

[7]  Kazuyuki Tanaka,et al.  Probabilistic image processing based on the Q-Ising model by means of the mean-field method and loopy belief propagation , 2004, Proceedings of the 17th International Conference on Pattern Recognition, 2004. ICPR 2004..

[8]  Lei Zhang,et al.  Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising , 2016, IEEE Transactions on Image Processing.

[9]  D. Titterington,et al.  Probabilistic image processing based on the Q-Ising model by means of the mean-field method and loopy belief propagation , 2004, Proceedings of the 17th International Conference on Pattern Recognition, 2004. ICPR 2004..

[10]  D.K.T. Lam.  Fractal dimension and multifractal spectra of the surface shape of textile patterns , 2010, 2010 International Conference on Image Analysis and Signal Processing.

[11]  Boualem Haddad,et al.  Multifractal analysis by the large deviation spectrum to detect osteoporosis , 2013, 2013 8th International Workshop on Systems, Signal Processing and their Applications (WoSSPA).

[12]  V. Amaral,et al.  Thermodynamics of the 2-D Ising Model From a Random Path Sampling Method , 2014, IEEE Transactions on Magnetics.

[13]  Lu Xiankai,et al.  Ising-like model for neural representation of natural images , 2014, 2014 12th International Conference on Signal Processing (ICSP).

[14]  Jieyu Zhao,et al.  A polynomial-time algorithm for image segmentation using Ising models , 2011, 2011 Seventh International Conference on Natural Computation.

[15]  Kai Sun,et al.  Texture analysis for deep seabed type classification based on multifractal spectrum , 2014, OCEANS 2014 - TAIPEI.

[16]  Stanley H. Chan,et al.  Plug-and-Play ADMM for Image Restoration: Fixed-Point Convergence and Applications , 2016, IEEE Transactions on Computational Imaging.

[17]  Yong Xu,et al.  Scale-space texture description on SIFT-like textons , 2012, Comput. Vis. Image Underst..

[18]  Dawei Qi,et al.  Hölder exponent and multifractal spectrum analysis in the pathological changes recognition of medical CT image , 2011, 2011 Chinese Control and Decision Conference (CCDC).

[19]  Laurent Risser,et al.  Fast bilinear extrapolation of 3D ising field partition function. application to fMRI image analysis. , 2009, 2009 16th IEEE International Conference on Image Processing (ICIP).

[20]  W. David Pan,et al.  Evaluation of the biased run-length coding method on binary images generated by a modified Ising model , 2016, SoutheastCon 2016.