Learning 2D Morphological Network for Old Document Image Binarization

Document image binarization, especially old handwritten documents, is a very important yet challenging task. There are various bottlenecks for binarizing historical documents due to different types of degradation present imultaneously such as back impression, ink bleed through, faded colours, and wear and tear of the writing media. We consider these degradation as various types of noise in the document image. Here we have proposed a 2D morphological network which consists of basic morphological operation like dilation and erosion to perform our targeted task. The network also includes linear combination of output from dilation and erosion operations. The aforementioned 2D morphological network is applied for image binarization, where the structuring elements (SEs) and the weights of the linear combination layer are learned through back-propagation. The proposed network has been evaluated on DIBCO 2017 and H-DIBCO 2018 and ISI-Letter dataset. Our results show more convincing as compared to the results of other state-of-the-art methods. Though the network is developed for old handwritten documents, it may be tuned to work for image processing task. The source code can be found here https://github.com/ranjanZ/ICDAR_Binarization

[1]  Konstantinos Zagoris,et al.  ICFHR 2018 Competition on Handwritten Document Image Binarization (H-DIBCO 2018) , 2018, 2018 16th International Conference on Frontiers in Handwriting Recognition (ICFHR).

[2]  Jennifer L. Davidson,et al.  Morphology neural networks: An introduction with applications , 1993 .

[3]  Bhabatosh Chanda,et al.  Morphological Networks for Image De-raining , 2019, DGCI.

[4]  Eric Dubois,et al.  Reduction of Bleed-through in Scanned Manuscript Documents , 2001, PICS.

[5]  Lazaros T. Tsochatzidis,et al.  ICDAR 2019 Competition on Document Image Binarization (DIBCO 2019) , 2017, 2019 International Conference on Document Analysis and Recognition (ICDAR).

[6]  N. Otsu.  A threshold selection method from gray level histograms , 1979 .

[7]  Ioannis Pratikakis,et al.  ICFHR 2012 Competition on Handwritten Document Image Binarization (H-DIBCO 2012) , 2012, 2012 International Conference on Frontiers in Handwriting Recognition.

[8]  Peter Sussner,et al.  An introduction to morphological neural networks , 1996, Proceedings of 13th International Conference on Pattern Recognition.

[9]  Matti Pietikäinen,et al.  Adaptive document image binarization , 2000, Pattern Recognit..

[10]  Chunheng Wang,et al.  Degraded document image binarization using structural symmetry of strokes , 2018, Pattern Recognit..

[11]  Eero P. Simoncelli,et al.  Image quality assessment: from error visibility to structural similarity , 2004, IEEE Transactions on Image Processing.

[12]  Mohamed Cheriet,et al.  RSLDI: Restoration of single-sided low-quality document images , 2009, Pattern Recognit..

[13]  Mohamed Cheriet,et al.  A local linear level set method for the binarization of degraded historical document images , 2012, International Journal on Document Analysis and Recognition (IJDAR).

[14]  Bhabatosh Chanda,et al.  Dense Morphological Network: An Universal Function Approximator , 2018, ArXiv.

[15]  Konstantinos Zagoris,et al.  ICDAR2017 Competition on Document Image Binarization (DIBCO 2017) , 2017, 2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR).