Pembentukan Model Recurrent Neural Network dan Connectionist Temporal Classification Pada Pengenalan Kata Tulisan Tangan Offline

Diana Tri Susetianingtias; Rini Arianty; Rodiah; Eka Patriya

doi:10.51454/decode.v3i2.151

Authors

Diana Tri Susetianingtias Sistem Komputer Universitas Gunadarma
Rini Arianty Sistem Informasi Universitas Gunadarma
Rodiah Informatika Universitas Gunadarma
Eka Patriya Manajemen Universitas Gunadarma

DOI:

https://doi.org/10.51454/decode.v3i2.151

Keywords:

bobot, CNN, global performance metric, parameter, tulisan tangan

Abstract

Pengolahan informasi sebagai bentuk dari pengembangan teknologi yang semakin pesat adalah pengenalan tulisan tangan. Pengenalan tulisan tangan saat ini banyak diimplementasikan untuk melakukan identifikasi dokumen-dokumen penting berbentuk digital. Pengenalan tulisan tangan cetak dengan benar dan akurat dapat digunakan untuk menunjang kegiatan manusia baik dalam kegiatan sehari – hari, sekolah, maupun pekerjaan. Pada penelitian ini, peneliti akan melakukan pengenalan tulisan tangan cetak menggunakan Convolutional Neural Network pada perangkat mobile. Penelitian juga akan mengimplementasikan model CNN yang dihasilkan pada android dan mengidentifikasi tulisan tangan yang diambil secara langsung menggunakan kamera. Proses pelatihan menggunakan library tensorflow pada bahasa pemrograman python, dilakukan secara online. Model CNN berhasil dibentuk dengan jumlah 11 hidden layer, terdiri dari 5 lapisan konvolusi, 3 lapisan max pooling, 1 lapisan flatten, dan 2 lapisan fully connected. Model mengambil masukan berupa citra berukuran 28 x 28 piksel dan menghasilkan keluaran berupa pengenalan tulisan tangan cetak. Total parameter bobot yang dimiliki oleh model berjumlah 1.556.495 variabel. Model CNN yang sudah dilatih berhasil diimplementasikan dan dijalankan pada aplikasi android. Berdasarkan perhitungan Global Performance Metric, Aplikasi berhasil mengenali citra tulisan tangan 1 kata dengan akurasi 82,12%. Hasil penelitian diharapkan dapat mengenali tulisan tangan cetak dengan akurasi tinggi pada perangkat android.

References

Aqab, S., & Tariq, M. U. (2020). Handwriting recognition using artificial intelligence neural network and image processing. International Journal of Advanced Computer Science and Applications, 11(7), 137-146. https://doi.org/10.14569/IJACSA.2020.0110719

Bhunia, A. K., Das, A., Bhunia, A. K., Kishore, P. S. R., & Roy, P. P. (2019). Handwriting recognition in low-resource scripts using adversarial learning. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2019-June(November), 4762-4771. https://doi.org/10.1109/CVPR.2019.00490

Chammas, E., Mokbel, C., & Likforman-Sulem, L. (2018). Handwriting recognition of historical documents with few labeled data. Proceedings - 13th IAPR International Workshop on Document Analysis Systems, DAS 2018, 43-48. https://doi.org/10.1109/DAS.2018.15

Darmatasia, & Fanany, M. I. (2017). Handwriting recognition on form document using convolutional neural network and support vector machines (CNN-SVM). 2017 5th International Conference on Information and Communication Technology, ICoIC7 2017, April. https://doi.org/10.1109/ICoICT.2017.8074699

Das, M., Panda, M., & Dash, S. (2022). Enhancing the Power of CNN Using Data Augmentation Techniques for Odia Handwritten Character Recognition. Advances in Multimedia, 2022. https://doi.org/10.1155/2022/6180701

Dwivedi, U., Rajput, P., Sharma, M. K., & Noida, G. (2017). Cursive handwriting recognition system using feature extraction and artificial neural network. International Research Journal of Engineering and Technology, 4(03), 2202-2206.

Fitrianingsih, F., Madenda, S., Ernastuti, E., Widodo, S., & Rodiah, R. (2017). Cursive handwriting segmentation using ideal distance approach. International Journal of Electrical and Computer Engineering, 7(5), 2863-2872. https://doi.org/10.11591/ijece.v7i5.pp2863-2872

Fitrianingsih, F., Susetianingtias, D. T., Pernadi, D., Patriya, E., & Arianty, R. (2022). The Implementation of Artificial Neural Network (ANN) on Offline Cursive Handwriting Image Recognition. ILKOM Jurnal Ilmiah, 14(1), 63-73.

Fitrianingsih, F., Widodo, S., Madenda, S., & Rodiah, R. (2016). Slant Correction and Detection for Offline Cursive Handwriting using 2D Affine Transform. International Journal of Engineering Research & Technology (IJERT), 5(8), 568-572.

Gonzalez, R. C., & Woods, R. E. (n.d.). (2009). Digital Image Processing, 4e. Pearson education India.

Impedovo, D., Pirlo, G., & Plamondon, R. (2012, September). Handwritten signature verification: new advancements and open issues. In 2012 International Conference on Frontiers in Handwriting Recognition, 367-372. https://doi.org/10.1109/ICFHR.2012.211

Ismael, S. O. (2020). Deep Learning and Image Processing for Handwritten Style Recognition.

J, R., & Kumar, D. P. (2020). Gesture Recognition using CNN and RNN. International Journal of Recent Technology and Engineering (IJRTE), 9(2), 230-233. https://doi.org/10.35940/ijrte.b3417.079220

Ji, B., & Chen, T. (2019). Generative Adversarial Network for Handwritten Text. https://doi.org/10.48550/arXiv.1907.11845

Karthi, M., Priscilla, R., & Syed, K. J. (2020). A novel content detection approach for handwritten english letters. Procedia Computer Science, 172(2012), 1016-1025. https://doi.org/10.1016/j.procs.2020.05.149

Kayumov, Z., Tumakov, D., & Mosin, S. (2020). Hierarchical Convolutional Neural Network for Handwritten Digits Recognition. Procedia Computer Science, 171(2019), 1927-1934. https://doi.org/10.1016/j.procs.2020.04.206

Li, P., Wang, H., Yu, M., & Li, Y. (2021, April). Overview of image smoothing algorithms. In Journal of Physics: Conference Series, 12-24. https://doi.org/10.1088/1742-6596/1883/1/012024

Marti, U. V., & Bunke, H. (2002). The IAM-database: an English sentence database for offline handwriting recognition. International Journal on Document Analysis and Recognition, 5, 39-46.

Memon, J., Sami, M., Khan, R. A., & Uddin, M. (2020). Handwritten optical character recognition (OCR): A comprehensive systematic literature review (SLR). IEEE Access, 8, 142642-142668. https://doi.org/10.1109/ACCESS.2020.3012542

Mohsin, A., & Sadoon, M. (2020). Developing an Arabic Handwritten Recognition System by Means of Artificial Neural Network Developing an Arabic Handwritten Recognition System by Means of Artificial Neural Network. June, 1-4. https://doi.org/10.36478/jeasci.2020.1.3

Samant, Y. S. (2021). Handwritten Text Recognition Using Machine Learning. International Journal of Advanced Research in Science, Communication and Technology, 14(02), 106-108. https://doi.org/10.48175/ijarsct-2006

Samsuryadi, S., Kurniawan, R., & Susilawati Mohamad, F. (2021). Automated handwriting analysis based on pattern recognition: a survey. Indonesian Journal of Electrical Engineering and Computer Science, 22(1), 196-206. https://doi.org/10.11591/ijeecs.v22.i1.pp196-206