Pengembangan Aplikasi Deteksi Tulisan Tangan Mahasiswa Teknik UNP Menggunakan Metode Computer Vision dan Deep Learning
DOI:
https://doi.org/10.29407/avzzx427Keywords:
Tulisan Tangan, CNN, ResNet-18, Deteksi Pemilik, Computer Vision.Abstract
Sistem pengenalan tulisan tangan (handwriting recognition) menjadi sangat penting karena mampu mengubah citra manual menjadi informasi digital untuk kebutuhan identifikasi penulis secara otomatis. Penelitian ini berfokus pada pengembangan aplikasi berbasis web untuk mengidentifikasi pemilik tulisan tangan dari lima individu berbeda menggunakan metode Deep Learning dengan arsitektur Convolutional Neural Network (CNN) ResNet-18. Proses inti algoritma dimulai dengan tahap pra-pemrosesan citra melalui konversi grayscale untuk pengurangan dimensi data serta normalisasi ukuran citra menjadi 224 224 piksel. Model kemudian melakukan ekstraksi fitur visual secara hierarkis melalui lapisan konvolusi untuk mengenali pola unik garis dan sudut tulisan tangan, serta memanfaatkan residual block untuk mengatasi masalah vanishing gradient selama pelatihan.. Klasifikasi akhir dilakukan pada lapisan fully connected menggunakan fungsi aktivasi softmax untuk menentukan probabilitas pemilik tulisan5. Hasil pengujian terhadap 117 data uji menunjukkan bahwa model berhasil mencapai akurasi keseluruhan sebesar 86,32%. Temuan kunci menunjukkan performa terbaik pada kelas "Afandi" dengan presisi 1.00, sementara kendala identifikasi masih ditemukan pada kelas "Adit" dengan F1-score terendah sebesar 0,74 akibat kemiripan pola tulisan antar individu. Secara fungsional, hasil pengujian alpha dan beta mengonfirmasi bahwa seluruh fungsi utama aplikasi berjalan baik dan mampu membantu pengguna mengetahui identitas pemilik tulisan secara cepat.
Downloads
References
[1] Abdelrahman, M., & Ali, H. (2021). Handwritten character recognition using deep convolutional neural networks. IEEE Access, 9, 11562–11574. https://doi.org/10.1109/ACCESS.2021.3051884
[2] Ooi, C., Lee, S., & Lim, K. (2021). Signature verification using CNN-based feature extraction for handwriting authentication. Expert Systems with Applications, 184, 115437. https://doi.org/10.1016/j.eswa.2021.115437
[3] Zhang, Y., Liu, Q., & Chen, H. (2022). Hybrid CNN-RNN architecture for Arabic handwritten text recognition. Pattern Recognition Letters, 155, 46–54. https://doi.org/10.1016/j.patrec.2022.01.012
[4] Singh, R., & Kaur, G. (2023). Writer identification using VGG16-based convolutional neural networks. Journal of Intelligent Systems, 32(4), 678–689. https://doi.org/10.1515/jisys-2023-0032
[5] Li, J., & Wang, Z. (2024). Enhancing handwriting identification through multi-writer dataset augmentation and CNN optimization. Neural Computing and Applications, 36(2), 985–998. https://doi.org/10.1007/s00521-024-08912-3
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Bagus Kusuma Wijaya, Dymas Anggara
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Copyright on any article is retained by the author(s).
- The author grants the journal, right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgment of the work’s authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal’s published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.
- The article and any associated published material is distributed under the Creative Commons Attribution-ShareAlike 4.0 International License
