Abstract
The convenience of search, both on the personal computer hard disk as well as on the web, is still limited mainly to machine printed text documents and images because of the poor accuracy of handwriting recognizers. The focus of research in this paper is the segmentation of handwritten text and machine printed text from annotated documents sometimes referred to as the task of “ink separation” to advance the state-of-art in realizing search of hand-annotated documents. We propose a method which contains two main steps—patch level separation and pixel level separation. In the patch level separation step, the entire document is modeled as a Markov Random Field (MRF). Three different classes (machine printed text, handwritten text and overlapped text) are initially identified using G-means based classification followed by a MRF based relabeling procedure. A MRF based classification approach is then used to separate overlapped text into machine printed text and handwritten text using pixel level features forming the second step of the method. Experimental results on a set of machine-printed documents which have been annotated by multiple writers in an office/collaborative environment show that our method is robust and provides good text separation performance.
Article PDF
Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.Avoid common mistakes on your manuscript.
References
FANN. http://leenissen.dk/fann/ (2007)
Alpert, S., Galun, M., Basri, R., Brandt A.: Image segmentation by probabilistic bottom-up aggregation and cue integration. IEEE Conf. Comput. Vis. Pattern Recognit. 1–8 (2007)
Banerjee, J., Namboodiri, A., Jawahar, C.: Contextual restoration of severely degraded document images. IEEE Conf. Comput. Vis. Pattern Recognit. 517–524 (2009)
Belongie S., Malik J., Puzicha J.: Shape matching and object recognition using shape contexts. IEEE Trans. Pattern Anal. Mach. Intell. 24, 509–522 (2002)
Burns, T.J., Corso, J.J.: Robust unsupervised segmentation of degraded document images with topic models. IEEE Conf. Comput. Vis. Pattern Recognit. 1287–1294 (2009)
Cao, H., Govindaraju, V.: Handwritten carbon form preprocessing based on Markov Random Field. IEEE Conf. Comput. Vis. Pattern Recognit. 1–7 (2007)
Cao H., Govindaraju V.: Preprocessing of low-quality handwritten documents using Markov Random Fields. IEEE Trans. Pattern Anal. Mach. Intell. 31, 1184–1194 (2009)
Corso, J.: Discriminative modeling by boosting on multilevel aggregates. IEEE Conf. Comput. Vis. Pattern Recognit. 1–8 (2008)
Eduardo, J.B.D.S., Dubuisson, B., Bortolozzi, F.: Characterizing and distinguishing text in bank cheque images. In: Proceedings of the XV Brazilian Symposium on Computer Graphics and Image Processing, pp. 203–209 (2002)
Eran B., Shimon U.: Combined top-down/bottom-up segmentation. IEEE Trans. Pattern Anal. Mach. Intell. 30, 2109–2125 (2008)
Fan K.-C., Wang L.-S., Tu Y.-T.: Classification of machine printed and handwritten texts using character block layout variance. Pattern Recognit. 31, 1275–1284 (1998)
Farooq F., Sridharan K., Govindaraju V.: Identifying handwritten text in mixed documents. IEEE Int. Conf. Pattern Recognit. 2, 1142–1145 (2006)
Freeman W., Carmichael O., Pasztor E.: Learning low-level vision. Int. J. Comput. Vis. 40, 25–47 (2000)
Freeman W., Jones T., Pasztor E.: Example-based super-resolution. IEEE Trans. Comput. Graph. Appl. 22, 56–65 (2002)
Gorman L.O.: The document spectrum for page layout analysis. IEEE Trans. Pattern Anal. Mach. Intell. 15, 1162–1173 (1993)
Grigorescu S., Petkov N., Kruizinga P.: Comparison of texture features based on Gabor filters. IEEE Trans. Image Process. 11, 1160–1167 (2002)
Guo, J., Ma, M.: Separating handwritten material from machine printed text using hidden Markov models. IEEE Int. Conf. Document Anal. Recognit. 439–443 (2001)
Gupta, M., Rajaram, S., Petrovic, N., Huang, T.: Models for patch based image restoration. IEEE Workshop Comput. Vis. Pattern Recognit. 17–22 (2006)
Hamerly, G., Elkan, C.: Learning the k in k-means. Adv. Neural Inf. Process. Syst. (2003)
Jang S., Jeong S., Nam Y.: Classification of machine-printed and handwritten addresses on Korean mail piece images using geometric features. IEEE Int. Conf. Pattern Recognit. 2, 383–386 (2004)
Lelore, T., Bouchara, F.: Document image binarisation using markov field model. IEEE Int. Conf. Document Anal. Recognit. 551–555 (2009)
Lettner, M., Sablatnig, R.: Spatial and spectral based segmentation of text in multispectral images of ancient documents. IEEE Int. Conf. Document Anal. Recognit. 813–817 (2009)
Li, S.Z.: A markov Random Field model for object matching under contextual constraints. IEEE Conf. Comput. Vis. Pattern Recognit. 866–869 (1994)
Li, S.Z.: Markov Random Field models in computer vision. Third European Conf. Comput. Vis. 2, 361–370 (1994)
Li S.Z.: Markov Random Field Modeling in Image Analysis. Springer, London (2009)
Likforman-Sulem L., Vaillant P., de Bodard de la Jacopiere A.: Automatic name extraction from degraded document images. Pattern Anal. Appl. 9, 211–227 (2006)
Nagy, G., Seth, S., Stoddard, S.: Document analysis with an expert system. Pattern Recognit. Pract. II. 2, 149–155 (1984)
Pearl J.: Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. Morgan Kaufmann, San Francisco (1988)
Peng, X., Setlur, S., Govindaraju, V., Sitaram, R.: Overlapped text segmentation using Markov Random Field and aggregation. In: 9th IAPR International Workshop on Document Analysis Systems (DAS), pp. 129–134 (2010)
Peng, X., Setlur, S., Govindaraju, V., Sitaram, R., Bhuvanagiri, K.: Markov Random Field based text identification from annotated machine printed documents. IEEE Int. Conf. Document Anal. Recognit. 431–435 (2009)
Petkov N.: Biologically motivated computationally intensive approaches to image pattern recognition. Futur. Gener. Comput. Syst. 11, 451–465 (1995)
Petkov N., Kruizinga P.: Computational models of visual neurons specialised in the detection of periodic and aperiodic oriented visual stimuli: bar and grating cells. Biol. Cybern. 76, 83–96 (1997)
Schultz R.R., Stevenson R.L.: Extraction of high-resolution frames from video sequences. IEEE Trans. Image Process. 5, 996–1011 (1996)
Sheikh Y., Shah M.: Bayesian modeling of dynamic scenes for object detection. IEEE Trans. Pattern Anal. Mach. Intell. 27, 1778–1792 (2005)
Shetty, S., Srinivasan, H., Beal, M., Srihari, S.: Segmentation and labeling of documents using conditional random fields. SPIE Int. Conf. Document Recognit. Retr. IV. 6500U-1-11 (2007)
Srihari, S.N., Govindaraju, V., Shekhawat, A.: Interpretation of handwritten addresses in U. S. mailstream. IEEE Int. Conf. Document Anal. Recognit. 291–294 (1993)
Xiong, L., Wang, F., Zhang, C.: Multilevel belief propagation for fast inference on Markov Random Fields. IEEE Int. Conf. Data Min. 371–380 (2007)
Zhao, L., Davis, L.S.: Iterative figure-ground discrimination. In: Proceedings of the 17th International, IEEE International Conference Pattern Recognition, vol. 1, pp. 67–70, Aug 2004
Zheng Y., Li H., Doermann D.: Machine printed text and handwriting identification in noisy document images. IEEE Trans. Pattern Anal. Mach. Intell. 26, 337–353 (2004)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Peng, X., Setlur, S., Govindaraju, V. et al. Handwritten text separation from annotated machine printed documents using Markov Random Fields. IJDAR 16, 1–16 (2013). https://doi.org/10.1007/s10032-011-0179-z
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10032-011-0179-z