El Yazısı ile Yazılmış Kimlik Numarası Tanıma Yöntemi

Year 2025, Volume: 15 Issue: 1, 41 - 59, 15.03.2025

Anıl Çelik , Zeynep Gürler , Mehmet Kıvılcım Keleş

https://doi.org/10.31466/kfbd.1448192

Abstract

El yazısı metin tanıma, çoğunlukla değişken yazı stilleri ve yazıların yer aldığı koşulsuz ortamlar nedeniyle hala zorlu bir problem olarak karşımıza çıkmaktadır. Literatürdeki çalışmalar, el yazısı örneklerin bulunduğu ortamların genellikle herhangi bir koşula sahip olmaması sebebiyle odaklandıkları el yazısı örneklerine ve ortamlarına göre özelleşmiştir. Bu yüzden farklı yazı tipleri ve ortamlar üzerinde uygulanabilirlikleri düşüktür. Önerilen çalışma, koşulsuz ortamlarda bulunan el yazısı Türkiye Cumhuriyeti Kimlik Numarası’nın (TCKN) tanınmasını hedeflemektedir. TCKN, her Türkiye vatandaşına verilen benzersiz bir kişisel kimlik numarasıdır. TCKN içerisindeki tek bir rakamı yanlış tahmin etmek bütün numaranın yanlış okunmasına yol açacağı için her rakamı doğru tahmin etmek oldukça önemlidir. Önerdiğimiz teknik, dokümanda TCKN yakalama, TCKN içerisinde rakam yakalama ve yakalanan rakamın sınıflandırılması olarak üç ana başlığa ayrılmaktadır. Rakam sınıflandırma aşamasında yakalanan rakamlar bir oto kodlayıcı yardımıyla taslak rakama dönüştürülürken oto kodlayıcıdan elde edilen özniteliklerle sınıflandırılmaktadırlar. Bu işlem, el yazısı rakamların sınıflarına ait en iyi temsile benzemesine çalışarak daha başarılı ayrışmalarını sağlamaktadır. Yapılan deneylerde oto kodlayıcı ile taslağa dönüştürme metodunun sınıflandırma başarımını önemli ölçüde artırdığı görülmüştür.

Keywords

El yazısı, Türkiye Cumhuriyeti Kimlik Numarası, Oto kodlayıcı

Handwritten ID Number Recognition Method

Abstract

Handwritten text recognition is still a challenging problem, mostly due to the variable writing styles and unconditional environments in which handwritten text appears. The studies in the literature are specialised to the handwriting samples and environments they focus on, as the environments in which handwritten samples are found are usually unconditioned. Therefore, their applicability to different handwriting types and environments is low. The proposed work aims to recognise the handwritten Turkish Republic Identity Number (TCKN) in unconditional environments. TCKN is a unique personal identification number given to every citizen of Turkey. It is very important to guess each digit correctly, since guessing a single digit in the TCKN will lead to misreading the whole number. Our proposed technique is divided into three main parts: capturing the TCKN in the document, capturing the digit in the TCKN and classifying the captured digit. In the digit classification stage, the captured digits are converted into draft digits with the help of an autoencoder and classified with the attributes obtained from the autoencoder. This process ensures more successful discrimination of handwritten digits by trying to resemble the best representation of their class. In the experiments, it has been observed that the autoencoder to draft conversion method significantly improves the classification performance.

Keywords

Handwritten, Turkish Republic Identity Number, Autoencoder

References

• Ahlawat S., Choudhary A., Nayyar A., Singh S., Yoon B. (2020). Improved handwritten digit recognition using convolutional neural networks (CNN), Sensors, 20(12):3344.
• An S., Lee M., Park S., Yang H., So J. (2020). An ensemble of simple convolutional neural network models for MNIST digit recognition, arXiv preprint arXiv:2008.10400.
• Basu S., Das N., Sarkar R., Kundu M., Nasipuri M., Basu DK. (2009). A hierarchical approach to recognition of handwritten Bangla characters, Pattern Recognition, 42(7):1467-84.
• Barua S., Malakar S., Bhowmik S., Sarkar R., Nasipuri M., Bangla (2017). Handwritten city name recognition using gradient-based feature, InProceedings of the 5th International Conference on Frontiers in Intelligent Computing: Theory and Applications: FICTA, Volume 1, 343-352.
• Boukharouba A., Bennia A. (2017). Novel feature extraction technique for the recognition of handwritten digits, Applied Computing and Informatics, 13(1):19-26.
• Busta M., Neumann L., Matas J. (2017). Deep textspotter: An end-to-end trainable scene text localization and recognition framework, InProceedings of the IEEE international conference on computer vision, 2204-2212.
• Byerly A., Kalganova T., Dear I. (2021). No routing needed between capsules, Neurocomputing, 463:545-53.
• Canny J. (1986). A computational approach to edge detection, IEEE Transactions on pattern analysis and machine intelligence, 679-98.
• Calonder M., Lepetit V., Strecha C., Fua P. (2010). Brief: Binary robust independent elementary features, In Computer Vision–ECCV 2010: 11th European Conference on Computer Vision, Heraklion, Crete, Greece, Proceedings, Part IV 11 2010, 778-792.
• Disc Bhowmik TK., Parui SK., Roy U. (2008). Discriminative HMM training with GA for handwritten word recognition, In 2008 19th International Conference on Pattern Recognition.
• Everingham M., Van Gool L., Williams C. K., Winn J., Zisserman A. (2010). The pascal visual object classes (voc) challenge, International journal of computer vision, 88:303-38.
• Fischler M. A., Bolles R. C. (1981). Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography, Communications of the ACM, 24(6):381-95.
• Frutiger A. (1980). Type. Sign. Symbol. ABC Verlag, Zurich, 50.
• Ghosh M. M., Maghari A. Y. (2017). A comparative study on handwriting digit recognition using neural networks, In 2017 international conference on promising electronic technologies (ICPET), pp. 77-81.
• He K., Zhang X., Ren S., Sun J. (2016). Deep residual learning for image recognition, In Proceedings of the IEEE conference on computer vision and pattern recognition, 770-778.
• Hirata D., Takahashi N. (2020). Ensemble learning in CNN augmented with fully connected subnetworks, arXiv preprint arXiv:2003.08562.
• Lin T. Y., Maire M., Belongie S., Hays J., Perona P., Ramanan D., Dollár P., Zitnick C.L. (2014). Microsoft coco: Common objects in context, In Computer Vision–ECCV 2014: 13th European Conference, Zurich, Switzerland, Proceedings, Part V 13 2014,740-755.
• Lin T. Y., Dollár P., Girshick R., He K., Hariharan B., Belongie S. (2017). Feature pyramid networks for object detection, In Proceedings of the IEEE conference on computer vision and pattern recognition, 2117-2125.
• Madhvanath S., Govindaraju V. (2001). The role of holistic paradigms in handwritten word recognition, IEEE transactions on pattern analysis and machine intelligence, 23(2):149-64.
• Mohebi E., Bagirov A. (2014). A convolutional recursive modified self organizing map for handwritten digits recognition, Neural Networks, 60, 104-118.
• Neumann L., Matas J. A. (2010). method for text localization and recognition in real-world images, InComputer Vision–ACCV 2010: 10th Asian Conference on Computer Vision, Queenstown, New Zealand, 8-12, Part III 10, 770-783.
• Neumann L., Matas J. (2012). Real-time scene text localization and recognition, In IEEE conference on computer vision and pattern recognition, 3538-3545.
• Redmon J., Farhadi A. (2017). YOLO9000: better, faster, stronger, In Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 7263-7271.
• Redmon J., Farhadi A. (2018) Yolov3: An incremental improvement, arXiv preprint arXiv:1804.02767.
• Ronneberger O., Fischer P., Brox T. (2015). U-net: Convolutional networks for biomedical image segmentation, In Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015: 18th International Conference, Munich, Germany, Proceedings, Part III 18 2015, 234-241.
• Roy PP., Bhunia AK., Das A., Dey P., Pal U. (2016). HMM-based Indic handwritten word recognition using zone segmentation, Pattern recognition, 60:1057-75.
• Rublee E., Rabaud V., Konolige K., Bradski G. (2011). ORB: An efficient alternative to SIFT or SURF, In 2011 International conference on computer vision, 2564-2571.
• Siddique F., Sakib S., Siddique M. A. (2019). Recognition of handwritten digit using convolutional neural network in python with tensorflow and comparison of performance for various hidden layers, In 2019 5th international conference on advances in electrical engineering (ICAEE), 541-546.
• Sueiras J., Ruiz V., Sanchez A., Velez JF. (2018). Offline continuous handwriting recognition using sequence to sequence neural networks, Neurocomputing, 289:119-28.
• Tamen Z., Drias H., Boughaci D. (2017). An efficient multiple classifier system for Arabic handwritten words recognition, Pattern Recognition Letters, 93:123-32.
• Viswanathan D. G. (2009). Features from accelerated segment test (fast), In Proceedings of the 10th workshop on image analysis for multimedia interactive services, London, UK, 6-8.