Published: 2026-01-01

Sistem Pengenalan Citra Dokumen Teks Terdistorsi menjadi Teks Menggunakan Metode Deep Learning

DOI: 10.35870/jtik.v10i1.4700

Talenta Teholi Zalukhu, Agustinus Rudatyo Himamunanto, Haeni Budiati
  • Talenta Teholi Zalukhu: Universitas Kristen Immanuel
  • Agustinus Rudatyo Himamunanto: Universitas Kristen Immanuel
  • Haeni Budiati: Universitas Kristen Immanuel
Cover VOLUME 10 NO 1 JANUARI 2026

Downloads

Article Metrics
Share:

Abstract

A common issue in document image processing is the inability of OCR systems to accurately read text from blurred images. This study aims to develop a deep learning-based OCR pipeline capable of recognizing text in blurred document images. The process begins with image enhancement using the DnCNN model for deblurring, followed by character segmentation and classification of A–Z characters using a CNN trained on the EMNIST Letters dataset. The recognized characters are then reconstructed into complete text. Experiments were conducted on 300 blurred images with varying levels of blur (low, medium, and high). Evaluation using PSNR and SSIM metrics showed improvements in image quality, with an average PSNR of 29,56 dB and SSIM of 0.89. Furthermore, the character classification accuracy reached 95.64%. Compared to the baseline (direct Tesseract OCR without deblurring), the proposed system showed a significant improvement in text readability. These results demonstrate the effectiveness of CNN-based approaches in enhancing OCR performance on blurred document images.

Keywords

OCR ; DnCNN ; Tesseract ; Deblurring ; Blurred Image ; Deep Learning ; Image Enhancement

Peer Review Process

This article has undergone a double-blind peer review process to ensure quality and impartiality.

Indexing Information

Discover where this journal is indexed at our indexing page to understand its reach and credibility.

Open Science Badges

This journal supports transparency in research and encourages authors to meet criteria for Open Science Badges by sharing data, materials, or preregistered studies.

How to Cite

Zalukhu, T. T., Himamunanto, A. R., & Budiati, H. (2026). Sistem Pengenalan Citra Dokumen Teks Terdistorsi menjadi Teks Menggunakan Metode Deep Learning. Jurnal JTIK (Jurnal Teknologi Informasi Dan Komunikasi), 10(1), 1-14. https://doi.org/10.35870/jtik.v10i1.4700
Article Information

This article has been peer-reviewed and published in the Jurnal JTIK (Jurnal Teknologi Informasi dan Komunikasi). The content is available under the terms of the Creative Commons Attribution 4.0 International License.

  • Issue: Vol. 10 No. 1 (2026)

  • Section: Computer & Communication Science

  • Published: %750 %e, %2026

AI Research Hub

This article is indexed and available through various AI-powered research tools and citation platforms. Our AI Research Hub ensures that scholarly work is discoverable, accessible, and easily integrated into the global research ecosystem. By leveraging artificial intelligence for indexing, recommendation, and citation analysis, we enhance the visibility and impact of published research.

Semantic Scholar Scite Dimensions Connected Papers
Scholarly Connection Platforms
Dimensions Dimensions Connected Papers Connected Papers Scite Scite Google Scholar Google Scholar Semantic Scholar Semantic Scholar Garuda Garuda Scilit Scilit Crossref Crossref BASE BASE Zenodo Zenodo Unpaywall Unpaywall OpenCitations OpenCitations
Author Biographies
Talenta Teholi Zalukhu
Google Scholar ResearchGate ORCID

Program Studi Informatika, Fakultas Sains & Komputer, Universitas Kristen Immanuel, Kabupaten Sleman, Daerah Istimewa Yogyakarta, Indonesia.

Agustinus Rudatyo Himamunanto
Google Scholar ResearchGate ORCID

Program Studi Informatika, Fakultas Sains & Komputer, Universitas Kristen Immanuel, Kabupaten Sleman, Daerah Istimewa Yogyakarta, Indonesia.

Haeni Budiati
Google Scholar ResearchGate ORCID

Program Studi Informatika, Fakultas Sains & Komputer, Universitas Kristen Immanuel, Kabupaten Sleman, Daerah Istimewa Yogyakarta, Indonesia.

References
  • Baldominos, A., Saez, Y., & Isasi, P. (2019). A survey of handwritten character recognition with mnist and emnist. Applied sciences, 9(15), 3169.
    Google Scholar Scite Semantic Scholar Scilit Crossref Connected Papers
  • Fadjeri, A., Asroriyah, A. M., & Rahmawati, A. (2022). Analisis Teks Bahasa Indonesia Dan Inggris Dari Sebuah Citra Menggunakan Pengolahan Citra Digital. Jurnal Teknologi Informasi Dan Komunikasi (TIKomSiN), 10(2), 42-46.
    Google Scholar Scite Semantic Scholar Scilit Crossref Connected Papers
  • Giamiko, E. S., & Tjiong, E. (2024). Pengembangan Aplikasi Pengenalan Tulisan Tangan Abjad dan Angka Berbasis Convolutional Neural Network. KALBISCIENTIA Jurnal Sains dan Teknologi, 11(02), 22-30. https://doi.org/10.53008/kalbiscientia.v11i02.3626.
    Google Scholar Scite Semantic Scholar Scilit Crossref Connected Papers
  • Hengaju, U., & Bal, B. K. (2023). Improving the Recognition Accuracy of Tesseract-OCR Engine on Nepali Text Images via Preprocessing. Advancement in Image Processing and Pattern Recognition, 3(2), 3.
    Google Scholar Scite Semantic Scholar Scilit Crossref Connected Papers
  • Manurung, I. D. P., Hidayatno, A., & Setiyono, B. (2011). Pengenalan Teks Cetak Pada Citra Teks Biner. Universitas Diponegoro, Semarang.
    Google Scholar Scite Semantic Scholar Scilit Crossref Connected Papers
  • Mohsenzadegan, K., Tavakkoli, V., & Kyamakya, K. (2022). Deep neural network concept for a blind enhancement of document-images in the presence of multiple distortions. Applied Sciences, 12(19), 9601.
    Google Scholar Scite Semantic Scholar Scilit Crossref Connected Papers
  • Nugroho, A. W. (2024). PENGGUNAAN MACHINE LEARNING DALAM PENGENALAN TEKS DARI GAMBAR. Jurnal Dunia Data, 1(4).
    Google Scholar Scite Semantic Scholar Scilit Crossref Connected Papers
  • Nugroho, B., & Puspaningrum, E. Y. (2021). Kinerja Metode CNN untuk Klasifikasi Pneumonia dengan Variasi Ukuran Citra Input. Jurnal Teknologi Informasi dan Ilmu Komputer (JTIIK), 8(3), 533-538.
    Google Scholar Scite Semantic Scholar Scilit Crossref Connected Papers
  • Peryanto, A., Yudhana, A., & Umar, R. (2020). Rancang bangun klasifikasi citra dengan teknologi deep learning berbasis metode convolutional neural network. Format J. Ilm. Tek. Inform, 8(2), 138.
    Google Scholar Scite Semantic Scholar Scilit Crossref Connected Papers
  • Pratiwi, A., Lestari, Y. D., & Lubis, Y. F. A. (2021, October). Analisis Kombinasi Vertical Projection Profile (VPP) Dan Top Down Profile (TDP) Dalam Segmentasi Karakter Pada Aplikasi OCR. In SEMINAR NASIONAL TEKNOLOGI INFORMASI & KOMUNIKASI (Vol. 1, No. 1, pp. 470-478).
    Google Scholar Scite Semantic Scholar Scilit Crossref Connected Papers
  • Rizqi, A., & Aziz, I. S. (2019). Rancang Bangun Aplikasi Penerjemah Bahasa Jepang–Indonesia Menggunakan OCR Berbasis Android. SinarFe7, 2(1), 276-280.
    Google Scholar Scite Semantic Scholar Scilit Crossref Connected Papers
  • Septiarini, A. (2016). Pengenalan Pola Pada Citra Digital Dengan Fitur Momen Invariant. Informatika Mulawarman: Jurnal Ilmiah Ilmu Komputer, 7(1), 8-11.
    Google Scholar Scite Semantic Scholar Scilit Crossref Connected Papers
  • Shaliniswetha, S., & Mahaboob, S. T. (2022). RESIDUAL LEARNING BASED IMAGE DENOISING AND COMPRESSION USING DNCNN. ICTACT Journal on Image & Video Processing, 13(2).
    Google Scholar Scite Semantic Scholar Scilit Crossref Connected Papers
  • Siliwangi, A. K., & Prabowo, Y. D. (2022). Pencarian Informasi Berbasis Teks dalam Komik Digital Menggunakan OCR. KALBISIANA Jurnal Sains, Bisnis dan Teknologi, 8(2), 1886-1894.
    Google Scholar Scite Semantic Scholar Scilit Crossref Connected Papers
  • Sulistiyo, M. F. (2022). Penerjemah Bahasa Inggris-Indonesia Berbasis Mobile Menggunakan Optical Character Recognition & Text To Speech (Doctoral dissertation, UBP Karawang).
    Google Scholar Scite Semantic Scholar Scilit Crossref Connected Papers
  • Wijaya, I., & Lubis, C. (2022). Pengimplementasian Ocr Menggunakan Cnn Untuk Ekstraksi Teks Pada Gambar. Jurnal Ilmu Komputer dan Sistem Informasi, 10(1). https://doi.org/10.24912/jiksi.v10i1.17836.
    Google Scholar Scite Semantic Scholar Scilit Crossref Connected Papers
License & Copyright
Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

Authors who publish with this journal agree to the following terms:

1. Copyright Retention and Open Access License

Authors retain copyright of their work and grant the journal non-exclusive right of first publication under the Creative Commons Attribution 4.0 International License (CC BY 4.0).

This license allows unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

2. Rights Granted Under CC BY 4.0

Under this license, readers are free to:

  • Share — copy and redistribute the material in any medium or format
  • Adapt — remix, transform, and build upon the material for any purpose, including commercial use
  • No additional restrictions — the licensor cannot revoke these freedoms as long as license terms are followed

3. Attribution Requirements

All uses must include:

  • Proper citation of the original work
  • Link to the Creative Commons license
  • Indication if changes were made to the original work
  • No suggestion that the licensor endorses the user or their use

4. Additional Distribution Rights

Authors may:

  • Deposit the published version in institutional repositories
  • Share through academic social networks
  • Include in books, monographs, or other publications
  • Post on personal or institutional websites

Requirement: All additional distributions must maintain the CC BY 4.0 license and proper attribution.

5. Self-Archiving and Pre-Print Sharing

Authors are encouraged to:

  • Share pre-prints and post-prints online
  • Deposit in subject-specific repositories (e.g., arXiv, bioRxiv)
  • Engage in scholarly communication throughout the publication process

6. Open Access Commitment

This journal provides immediate open access to all content, supporting the global exchange of knowledge without financial, legal, or technical barriers.

Downloads

Similar Articles

You may also start an advanced similarity search for this article.