Volume 33 #2

Contents

1. Computation of eye optics of the cetacean’s eye
2. Neurophysiological mechanisms of two sequential orientations comparison in working memory
3. Contribution of the marginal peripheral retina to color constancy: Evidence obtained due to contact lens with implanted occluder
4. Contribution of the spectral and temporal mechanisms to analysis of complex sound signals
5. Role of the donor of NO molecules in regulation of primary sensory neuron responses
6. Identity documents forgery detection with mobile devices
7. Transient processes of visual evoked potentials in the tasks of human-computer interfaces
8. Algebraic reconstruction in case of limited GPU memory in the task of computed tomography
9. New criteria for neural network encoder learning in the string segmentation problem

In this paper weconsider the problem of character segmentation. Image processing methods are widely used in various specialized algorithms, whichcannot solve the problem entirely. In some papers authors use different customizations based on the structures on the image and improve the final segmentation on the basis of knowledge about the alphabet and the font. However, such algorithms can be complicated and too non-robusttowards the alphabet and the font of the text.We suggest the neural network encoder learning approach instead of classic method for cuts estimation between characters. We demonstrate the effectiveness of this approach using strings from RF passport fields as an example and show the advantages of the suggested approach over the classic one.

Key words: convolutional neural networks, deep learning, text segmentation, cumulative sum

DOI: 10.1134/S0235009219020094

References:

• Chernov T.S., Il’in D.A., Bezmaternyh P.V., Faradzhev I.A., Karpenko S.M. Issledovanie metodov segmentacii izobrazenii tekstovuch blokov dokymentov pri pomochi algoritmov strukturnogo analiza I maschinnogo obucheniy [Research of Segmentation Methods for Images of Document Textual Blocks Based on the Structural Analysis and Machine Learning] Vestnik RFFI [RBRF Information Bulletin]. 2016. № 4 (92). P. 55–71. DOI: 10.22204/2410-4639-2016-092-04-55-71 (In Russian).
• Bulatov K., Arlazarov V.V., Chernov T., Slavin O., Nikolaev D.P. Smart IDReader: Document Recognition in Video Stream. The 14th IAPR International Conference on Document Analysis and Recognition. 2018. P. 39–44. DOI: 10.1109/ICDAR.2017.347.
• Casey R.G., Lecolinet E. A Survey of Methods and Strategies in Character Segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence. 1996. V. 18 (7). P. 690–706. DOI: 10.1109/34.506792.
• Chernyshova Y., Gayer A., Sheshkus A. Generation method of synthetic training data for mobile OCR system. Proc. SPIE 10696, Tenth International Conference on Machine Vision. 2018. P. 1–7. DOI: 10.1117/12.2310119.
• Chomphuwiset P. Printed Thai Character Segmentationand Recognition. IEEE 4th International Conference on Soft Computing & Machine Intelligence. 2017. P. 123–127. DOI: 10.1109/ISCMI.2017.8279611.
• Firdaus F.I., Khumaini A., Utaminingrum F. Arabic Letter Segmentation using Modified Connected Component Labeling. International Conference on Sustainable Information Engineering and Technology. 2017. P. 392–397. DOI: 10.1109/SIET.2017.8304170.
• Grafmuller M., Beyerer J. Performance improvement of character recognition in industrial applications using prior knowledge for more reliable segmentation. Expert Systems with applications. 2013. V. 40(17). P. 6955–6963. DOI: 10.1016/j.eswa.2013.06.004.
• Jacobs C.E., Rinker J.R., Simard P.Y., Viola P.A. Low resolution OCR for camera acquired documents. US Application. No. US20050259866A1. 2005.
• Jia F., Shi C., Wang Y., Wang C., Xiao B. Grayscale-projection based Optimal Character Segmentation for Camera-captured Faint Text Recognition. 14th IAPR International Conference on Document Analysis and Recognition. 2017. P. 1301–1306. DOI: 10.1109/ICDAR. 2017.214
• Manwatkar P.M., Yadav S.H. Text Recognition from Images. IEEE International Conference on Innovations in Information, Embedded and Communication Systems. 2015. P. 1–6.
• Mai B.Q.L., Huynh T.H., Doan A.D. An Independent Character Recognizer for Distantly Acquired Mobile Phone Text Images. International Conference on Advanced Technologies for Communications. 2016. P. 85–90. DOI: 10.1109/ATC.2016.7764836.
• Mei Y., Wang X., Wang J. An Efficient Character Segmentation Algorithm for Printed Chinese Documents. Advanced Science and Technology Letters. 2013. V. 22. P. 183–189.
• Mollah A.F., Basu S., Nasipuri M. Segmentation of Camera Captured Business Card Images for Mobile Devices. International Journal of Computer Science and Applications. 2010. V. 1 (1). P. 33–37.
• Qi W., Li X., Yang B. A Character Segmentation Method without Character Verification. International Symposium on Intelligent Information Technology Application Workshops. 2008. P. 581–584. DOI: 10.1109/IITA
• Radhiah A., Machbub C., Hidayat E.M.I., Prihatmanto A.S. Printed Arabic Letter Recognition Based on Images. International Conference on Signals and Systems. 2018. P. 86– 91. DOI: 10.1109/ICSIGSYS.2018.8373574.
• Rehman A. Offline Touched Cursive Script Segmentation Based on Pixel Intensity Analysis. Twelfth International Conference on Digital Information Management. 2017. P. 324–327. DOI: 10.1109/ICDIM.2017.8244641.
• Ryan M., Hanafiah N. An Examination of Character Recognition on ID card using Template Matching Approach. Procedia Computer Science. 2015. V. 59. P. 520–529. DOI: 10.1016/j.procs.2015.07.534.
• Sahare P., Dhok S.B. Multilingual Character Segmentation and Recognition Schemes for Indian Document Images. IEEE Access. 2018. V. 6. P. 10603–10617. DOI: 10.1109/ACCESS.2018.2795104.
• Shivakumara P., Bhowmick S., Su B., Tan C. L., Pal U. A New Gradient based Character Segmentation Method for Video Text Recognition. International Conference on Document Analysis and Recognition. 2011. P. 126–130. DOI: 10.1109/ICDAR.2011.34.
• Vuckovic V., Azinovic B. General Character Segmentation Approach for Machine-Typed Documents. Proceedings of 4th International Conference on Electrical, Electronics and Computing Engineering. 2017. P. RTI2.2.1-6.