Abstract
The printed character recognition is an efficient and automatic method for inputting information to a computer nowadays that is used to translate the printed or handwritten images into an editable and readable text file. This paper aims to recognize a multifont and multisize of the English language printed word for a smart pharmacy purpose. The recognition system has been based on a convolution neural network (CNN) approach where line, word, and character are separately corrected, and then each of the separated characters is fed into the CNN algorithm for recognition purposes. The OpenCV open-source library has been used for preprocessing, which can segment English characters accurately and efficiently, and for recognition, the Keras library with the backend of TensorFlow has been used. The training and testing data sets have been designed to include 23 different fonts with six different sizes. The CNN algorithm achieves the highest accuracy of 96.6% comparing to the other state-of-the-art machine learning methods. The higher classification accuracy of the CNN approach shows that this type of algorithm is ideal for the English language printed word recognition. The highest error rate after testing the system using English electronic prescribing written with all proposed font-types is 0.23% in Georgia font.
Research funding: There is no research funding for this study.
Author contributions: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission.
Competing interests: The funding organization(s) played no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report or in the decision to submit the report for publication.
Informed consent: Informed consent was obtained from all individuals included in this study.
Ethical approval: The conducted research is not related to either human or animal use.
References
1. Beso, A, Franklin, BD, Barber, N. The frequency and potential causes of dispensing errors in a hospital pharmacy. Pharm World Sci 2005;27:182–90. https://doi.org/10.1007/s11096-004-2270-8.10.1007/s11096-004-2270-8Search in Google Scholar PubMed
2. James, KL, Barlow, D, McArtney, R, Hiom, S, Roberts, D, Whittlesea, C. Incidence, type and causes of dispensing errors: a review of the literature. Int J Pharm Pract 2009;17:9–30. https://doi.org/10.1211/ijpp/17.1.0004.10.1211/ijpp.17.1.0004Search in Google Scholar
3. Ashcroft, DM, Quinlan, P, Blenkinsopp, A. Prospective study of the incidence, nature and causes of dispensing errors in community pharmacies. Pharmacoepidemiol Drug Saf 2005;14:327–32. https://doi.org/10.1002/pds.1012.10.1002/pds.1012Search in Google Scholar PubMed
4. Boyd, AM, Chaffee, BW. Critical evaluation of pharmacy automation and robotic systems: a call to action. Hosp Pharm 2019;54:4–11. https://doi.org/10.1177/0018578718786942.10.1177/0018578718786942Search in Google Scholar PubMed PubMed Central
5. Parameshwarppa, S. Handwritten Kannada characters recognition using curvelet transform. Int J Comput Appl 2015:12–16. https://doi.org/10.1109/erect.2015.7499039.10.1109/ERECT.2015.7499039Search in Google Scholar
6. Patil, JM, Mane, AP. Multi font and size optical character recognition using template matching. Int J Emerg Technol Adv Eng 2013;3:504–6. https://doi.org/10.1109/icpr.1996.546816.10.1109/ICPR.1996.546816Search in Google Scholar
7. Zanwar, SR, Narote, AS, Narote, SP. English character recognition using robust back propagation neural network. In: Communications in Computer and Information Science. Singapore: Springer; 2019, vol 1037:216–27 pp. https://doi.org/10.1007/978-981-13-9187-3_20.10.1007/978-981-13-9187-3_20Search in Google Scholar
8. Ali, H, Ali, Y, Hossain, E, Sultana, S. Character recognition using wavelet compression. In: Proceedings 2010 13th International Conference and Computer and Information Technology ICCIT 2010; 2010:452–7 pp. https://doi.org/10.1109/ICCITECHN.2010.5723900.10.1109/ICCITECHN.2010.5723900Search in Google Scholar
9. Siriteerakul, T. Mixed Thai-English character classification based on histogram of oriented gradient feature. In: Proceedings 12th International Conference on Document Analysis Recognition, ICDAR; 2013:847–51 pp. https://doi.org/10.1109/ICDAR.2013.173.10.1109/ICDAR.2013.173Search in Google Scholar
10. Janrao, RR, Dighe, MDD. Handwritten English character recognition using lvq and knn. Int J Eng Sci Res Technol 2016;5:904–12. https://doi.org/10.5281/zenodo.60830.Search in Google Scholar
11. Dash, T, Nayak, T. English character recognition using artificial neural network. In: Proceedings of the National Conference AIRES-2012, Andhra University; 2013:7–9 pp.Search in Google Scholar
12. Rahiman, MA, Adheena, CV, Anitha, R, Deepa, N, Kumar, GM, Rajasree, MS. Bilingual OCR system for printed documents in Malayalam and English. In: ICECT 2011 – 2011 3rd International Conference on Electronics Computer Technology; 2011, vol 3:40–5 pp. https://doi.org/10.1109/ICECTECH.2011.5941797.10.1109/ICECTECH.2011.5941797Search in Google Scholar
13. Prasad, BK, Sanyal, G. A model approach to off-line English character. Int J Sci Res Publ 2012;2:2–7.Search in Google Scholar
14. Seeri, SV, Pujari, JD, Hiremath, PS. PNN based character recognition in natural scene images. Bonfring Int J Softw Eng Soft Comput 2016;6:109–13. https://doi.org/10.9756/bijsesc.8254.10.9756/BIJSESC.8254Search in Google Scholar
15. Nasser, EF, Karim, AA. English character recognition from video stream based on bag of visual words (BOVW). J Madent Alelem Coll 2018;10:12–29.Search in Google Scholar
16. Ansari, NN. License number plate recognition using template matching. Int J Comput Trends Technol 2016;35:175–8. https://doi.org/10.14445/22312803/ijctt-v35p133.10.14445/22312803/IJCTT-V35P133Search in Google Scholar
17. Kaur, I, Singh, KJ. Printed text recognition system for multi-script image. Int J Signal Process Syst 2016;4:411–16. https://doi.org/10.18178/ijsps.4.5.411-416.10.18178/ijsps.4.5.411-416Search in Google Scholar
18. Mohammed, AZ, Mohammed, EA, Aaref, AM. Real-time driver awareness detection system. IOP Conf Ser Mater Sci Eng 2020;745:1–13. https://doi.org/10.1088/1757-899x/745/1/012053.10.1088/1757-899X/745/1/012053Search in Google Scholar
19. Ram Mohan Rao, GIRK, Ramesh, B. Font and size identification in Telugu printed document. Int J Eng Res Dev 2013;6:92–110.Search in Google Scholar
20. Priyanka, N, Pal, S, Mandal, R. Line and word segmentation approach for printed documents. RTIPPR Int J Comput Appl IJCA 2010;1:30–6.Search in Google Scholar
21. Hariharan, B, Arbelaez, P, Girshick, R, Malik, J. Object instance segmentation and fine-grained localization using hypercolumns. IEEE Trans Pattern Anal Mach Intell 2017;39:1–14. https://doi.org/10.1109/tpami.2016.2578328.10.1109/TPAMI.2016.2578328Search in Google Scholar PubMed
22. Cai, Z, Vasconcelos, N. Cascade R-CNN: delving into high quality object detection. Cvpr 2018:6154–62.10.1109/CVPR.2018.00644Search in Google Scholar
23. Ramaiah, NP, Ijjine, EP, Mohan, CK. Illumination invariant thermal face recognition using convolutional neural network. In: 2015 IEEE International Conference on Signal Processing, Informatics, Communication and Energy Systems (SPICES). Kozhikode, India: IEEE; 2015. https://doi.org/10.1109/SPICES.2015.7091490.10.1109/SPICES.2015.7091490Search in Google Scholar
24. Lecun, Y, Bottou, L, Bengio, Y, Haffner, P. Gradient-based learning applied to document recognition. Proc IEEE 1998;86:2278–324. https://doi.org/10.1109/5.726791.10.1109/5.726791Search in Google Scholar
25. Goodfellow, I, Courville, A. Deep learning. MIT Press; 2016. http://www.deeplearningbook.org.Search in Google Scholar
26. Lin, G, Shen, W. Research on convolutional neural network based on improved Relu piecewise activation function. Procedia Comput Sci 2018;131:977–84. https://doi.org/10.1016/j.procs.2018.04.239.10.1016/j.procs.2018.04.239Search in Google Scholar
27. Wani, MA, Bhat, FA, Afzal, S, Khan, AI. Advances in deep learning. Kacprzyk, J, editor. Studies in Big Data. India: Springer Singapore; 2020. Chapter 2; Basics of supervised deep learning; 13–29 pp.10.1007/978-981-13-6794-6_2Search in Google Scholar
28. Cristina, M, Neri, G, Osiris, O, Villegas, V, Guadalupe, V, Sánchez, C. A convolutional neural network for handwritten digit recognition 2 materials and methods. Int J Comb Optim Probl Informatics 2020;11:97–105.Search in Google Scholar
29. Krizhevsky, A, Hinton, GE. ImageNet classification with deep convolutional neural networks. In: Advances in neural information processing systems; 2012:1097–105 pp.10.1145/3065386Search in Google Scholar
30. Chung, H, Lee, SJ, Park, JG. Deep neural network using trainable activation functions. In: Proceedings of the International Joint Conference on Neural Networks; 2016, vol 2016:348–52 pp.https://doi.org/10.1109/IJCNN.2016.7727219.10.1109/IJCNN.2016.7727219Search in Google Scholar
31. Nwankpa, CE, Ijomah, W, Gachagan, A, Marshall, S. Activation functions: comparison of trends in practice and research for deep learning [Internet]. Glasgow, UK, arXiv; 2018. Available from: http://arxiv.org/abs/1811.03378v1.Search in Google Scholar
32. Kouretas, I, Paliouras, V. Simplified hardware implementation of the softmax activation function. In: 2019 8th International Conference on Modern Circuits and Systems Technologies, MOCAST 2019. https://doi.org/10.1109/MOCAST.2019.8741677.10.1109/MOCAST.2019.8741677Search in Google Scholar
33. Benaddy, M, El Meslouhi, O, Es-saady, Y, Kardouchi, M. Handwritten tifinagh characters recognition using deep convolutional neural networks. Sens Imaging 2019;20:1–17. https://doi.org/10.1007/s11220-019-0231-5.10.1007/s11220-019-0231-5Search in Google Scholar
34. Hinton, GE, Srivastava, N, Krizhevsky, A, Sutskever, I, Salakhutdinov, RR. Improving neural networks by preventing co-adaptation of feature detectors [Internet]. Canada, arXiv; 2012. Available from: http://arxiv.org/abs/1207.0580.Search in Google Scholar
35. Alla, S, Adari, SK. Beginning anomaly detection using python-based deep learning. Apress. California: Springer; 2019:319–60 pp. Appendex A, Intro to keras.10.1007/978-1-4842-5177-5Search in Google Scholar
36. Bharath, V, Rani, NS. A font style classification system for English OCR. In: Proceedings 2017 International Conference Intelligent Computing and Control. I2C2 2017; 2017:1–5 pp. https://doi.org/10.1109/I2C2.2017.8321962.10.1109/I2C2.2017.8321962Search in Google Scholar
37. Mohammed, AJ, Al-Khaffaf, HSM. Font recognition of English letters based on distance profile features. Sci J Uni Zakho 2020;8:66–71. https://doi.org/10.25271/sjuoz.2020.8.2.694.10.25271/sjuoz.2020.8.2.694Search in Google Scholar
38. Samadiani, N, Hassanpour, H. A neural network-based approach for recognizing multi-font printed English characters. J Elect Sys Inform Technol 2015;2:207–18. https://doi.org/10.1016/j.jesit.2015.06.003.10.1016/j.jesit.2015.06.003Search in Google Scholar
39. Wei, TC, Sheikh, U, Rahman, AAHA. Improved optical character recognition with deep neural network. In: Proceedings – 2018 IEEE 14th International Colloquium on Signal Processing and its Applications, CSPA 2018; 2018:245–9 pp. https://doi.org/10.1109/CSPA.2018.8368720.10.1109/CSPA.2018.8368720Search in Google Scholar
© 2020 Walter de Gruyter GmbH, Berlin/Boston
