DEEP LEARNING-BASED OCR FRAMEWORK FOR RECEIPTS: PERFORMANCE EVALUATION OF EAST AND CRNN INTEGRATION
DOI:
https://doi.org/10.33480/jitk.v11i3.7073Keywords:
CRNN, Optimising, Text Detection, Text ExtractionAbstract
Existing OCR systems often struggle with shopping receipts due to irregular layouts, diverse fonts, and image noise. We propose a domain-specific OCR framework that combines the EAST detector for robust text localisation and the CRNN model for sequence-based recognition. Trained on 320 annotated receipts and tested on 84 images, our system achieved 92.6% character-level and 86.4% word-level accuracy, surpassing Tesseract (+15.2%) and standalone CRNN (+9.7%). These results demonstrate the framework’s effectiveness for receipt-specific OCR, supporting applications such as automated expense tracking and financial record digitisation
Downloads
References
[1] L. P. International et al., “AI-Driven Automation of Financial Document Processing : Enhancing Accuracy , Efficiency , and Fraud Detection with OCR , NLP , and Deep Learning,” vol. 44, no. 6, pp. 922–928, 2024.
[2] M. Ganesamoorthi and S. S. Ravikumar, “Unified Deep Learning Pipeline for License Plate Recognition: From Detection to Ocr With Faster Rcnn, Fcn, and Crnn,” in 2025 5th International Conference on Intelligent Communications and Computing (ICICC), Aug. 2025, pp. 235–240. doi: 10.1109/ICICC66840.2025.11199678.
[3] L. Xiang, H. Wen, and M. Zhao, “Pill Box Text Identification Using DBNet-CRNN,” Int. J. Environ. Res. Public Health, vol. 20, no. 5, 2023, doi: 10.3390/ijerph20053881.
[4] Y. Liu, Y. Wang, and H. Shi, “A Convolutional Recurrent Neural-Network-Based Machine Learning for Scene Text Recognition Application,” Symmetry (Basel)., vol. 15, no. 4, 2023, doi: 10.3390/sym15040849.
[5] R. Krishna and R. Samantapudi, “Table Extraction from Financial and Transactional Documents,” Am. Acad. Publ., vol. 05, no. 01, pp. 95–125, 2025.
[6] K. Banu, D. Andreas, W. Anggoro, and A. Setiawan, “OCR: Masa Depan Pengenalan Karakter Optik dan Dampaknya pada Kehidupan Modern,” J. Teknol. Inf., vol. 9, no. 2, pp. 147–156, 2023, doi: 10.52643/jti.v9i2.3798.
[7] S. Subbagari, “Leveraging Optical Character Recognition Technology for Enhanced Anti-Money Laundering (AML) Compliance,” Int. J. Comput. Sci. Eng., vol. 10, no. 5, pp. 1–7, 2023, doi: 10.14445/23488387/ijcse-v10i5p102.
[8] S. Lin and G. Hoendarto, “Aplikasi Mobile Money Management Dengan Fitur Optical Character Recognition Menggunakan Framework React Native,” Metik J., vol. 5, no. 2, pp. 19–27, 2021, doi: 10.47002/metik.v5i2.291.
[9] R. I. Indrakusuma, A. S. Ahmadiyah, and N. F. Ariyani, “Pengenalan dan Klasifikasi Tulisan pada Nota Pembelian Material (Studi Kasus Proyek Konstruksi),” J. Tek. ITS, vol. 10, no. 2, 2021, doi: 10.12962/j23373539.v10i2.77109.
[10] P. Prakash, S. Hanumanthaiah, and S. Mayigowda, “CRNN model for text detection and classification from natural scenes,” IAES Int. J. Artif. Intell., vol. 13, p. 839, 2024, doi: 10.11591/ijai.v13.i1.pp839-849.
[11] M. Kumar and S. Ajay, “OCR ‑ CRNN ( WBS ): an optical character recognition system based on convolutional recurrent neural network embedded with word beam search decoder for extraction of text,” Int. J. Inf. Technol., 2025, doi: 10.1007/s41870-025-02540-x.
[12] M. Azelicha Ayana, “Mobile Based Application Design of Management System ‘Travel Budget’ Using Flutter,” vol. 4, no. 1, p. 169, 2024, [Online]. Available: https://journal.uib.ac.id/index.php/combines
[13] X. Mifsud, L. Grech, A. Baldacchino, L. Keller, and G. Valentino, “Receipt Information Extraction with Joint Multi-Modal Transformer and Rule-Based Model,” Mach. Learn. Knowl. Extr., vol. 4, no. 167, pp. 1–21, 2025, doi: 10.3390/make7040167.
[14] Y. Jiang and J. Guan, “License Plate Recognition System Combining RISC-V Technology and Improved Faster R-CNN Algorithm,” Int. J. Intell. Transp. Syst. Res., vol. 23, no. 2, pp. 761–773, 2025, doi: 10.1007/s13177-025-00481-0.
[15] J. M. Yu, H. J. Ma, and J. L. Kong, “Receipt Recognition Technology Driven by Multimodal Alignment and Lightweight Sequence Modeling,” Electron., vol. 14, no. 9, pp. 1–29, 2025, doi: 10.3390/electronics14091717.
[16] M. H. Fariz and E. B. Setiawan, “the Impact of Word Embedding on Cyberbullying Detection Using Hybird Deep Learning Cnn-Bilstm,” JITK (Jurnal Ilmu Pengetah. dan Teknol. Komputer), vol. 10, no. 3, pp. 661–671, 2025, doi: 10.33480/jitk.v10i3.6270.
[17] H. Zhang, E. Whittaker, and I. Kitagishi, “Extending TrOCR for Text Localization-Free OCR of Full-Page Scanned Receipt Images,” Proc. - 2023 IEEE/CVF Int. Conf. Comput. Vis. Work. ICCVW 2023, pp. 1471–1477, 2023, doi: 10.1109/ICCVW60793.2023.00160.
[18] M. Firmansyah and I. Artikel, “Esensi Perbedaan Metode Kualitatif Dan Kuantitatif,” J. Ekon. Pembang., vol. 3, no. 2, 2021, [Online]. Available: https://elastisitas.unram.ac.id/index.php/elastisitas/article/view/46/56
[19] A. Biró, A. I. Cuesta-Vargas, J. Martín-Martín, L. Szilágyi, and S. M. Szilágyi, “Synthetized Multilanguage OCR Using CRNN and SVTR Models for Realtime Collaborative Tools,” Appl. Sci., vol. 13, no. 7, 2023, doi: 10.3390/app13074419.
[20] V. K. Soni, V. Shukla, S. R. Tandan, A. Pimpalkar, N. K. Nema, and M. Naik, “Performance Evaluation of Efficient and Accurate Text Detection and Recognition in Natural Scenes Images Using EAST and OCR Fusion,” vol. 16, no. 1, pp. 445–453, 2025.
[21] Y. Chong, K. H. Chua, M. Babrdel, L. Hau, and L. Wang, “Deep Learning and Optical Character Recognition for Digitization of Meter Reading,” 2022, pp. 7–12. doi: 10.1109/ISCAIE54458.2022.9794463.
[22] A. Yadav, S. Singh, M. Siddique, N. Mehta, and A. Kotangale, “OCR using CRNN: A Deep Learning Approach for Text Recognition,” in 2023 4th International Conference for Emerging Technology (INCET), May 2023, pp. 1–6. doi: 10.1109/INCET57972.2023.10170436.
[23] S. Wang, C. Dong, P. Yang, W. Zan, and X. Chen, “Automatic inventory system of librarian books based on a deep learning algorithm with EAST and CRNN,” in Proceedings of the 2022 5th International Conference on Software Engineering and Information Management, in ICSIM ’22. New York, NY, USA: Association for Computing Machinery, 2022, pp. 212–218. doi: 10.1145/3520084.3520119.
[24] I. N. T. Lestari and D. I. Mulyana, “Implementation of Ocr (Optical Character Recognition) Using Tesseract in Detecting Character in Quotes Text Images,” J. Appl. Eng. Technol. Sci., vol. 4, no. 1, pp. 58–63, 2022, doi: 10.37385/jaets.v4i1.905.
[25] Z. B. Alawi, “A Comparative Survey of PyTorch vs TensorFlow for Deep Learning : Usability , Performance , and Deployment Trade-offs,” arXiv Prepr., 2025.
[26] H. Ullah, M. Tanveer, and A. Jan, “Enhancing Handwritten Prescription Recognition with AI-Driven OCR,” vol. 09, no. 02, 2025.
[27] M. Al Saadi, B. Al Saadi, D. A. Farhan, and O. A. Hassen, “Optimizing Neural Network Architectures with TensorFlow and Keras for Scalable Deep Learning,” J. Intell. Syst. Internet Things, vol. 18, no. 01, pp. 114–125, 2026, doi: 10.54216/JISIoT.180108.
[28] J. Wernersbach and C. Tracy, “East,” Swim. Holes Texas, pp. 45–68, 2021, doi: 10.7560/321522-007.
[29] X.-F. Wang, Z.-H. He, K. Wang, Y.-F. Wang, L. Zou, and Z.-Z. Wu, “A survey of text detection and recognition algorithms based on deep learning technology,” Neurocomput., vol. 556, no. C, Nov. 2023, doi: 10.1016/j.neucom.2023.126702.
[30] H. Li, P. Wang, C. Shen, and G. Zhang, “Show, attend and read: A simple and strong baseline for irregular text recognition,” 33rd AAAI Conf. Artif. Intell. AAAI 2019, 31st Innov. Appl. Artif. Intell. Conf. IAAI 2019 9th AAAI Symp. Educ. Adv. Artif. Intell. EAAI 2019, pp. 8610–8617, 2019, doi: 10.1609/aaai.v33i01.33018610
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Deo Ekel Pindonta Ginting, Siti Anzani Sitorus Pane, Marlince NK Nababan
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
-a.jpg)
-b.jpg)
