Pengembangan Model Optical Character Recognition dalam Digitalisasi Berita Acara Bimbingan Skripsi

Afiat, Muhammad Ivan Ardianadi and Abimanyu, Ignatius Ida Bagus (2026) Pengembangan Model Optical Character Recognition dalam Digitalisasi Berita Acara Bimbingan Skripsi. Project Report. [s.n.], [s.l.]. (Unpublished)

Text 5025221178_5025221189-Project_Report.pdf - Accepted Version Restricted to Repository staff only Download (944kB) | Request a copy

Abstract

Digitalisasi Berita Acara Bimbingan Skripsi secaramanual tidak efisien dan rentan kesalahan. Proyek inimengembangkan sistem aplikasi web untuk mengotomatisasiproses tersebut menggunakan Optical Character Recognition(OCR) dengan pendekatan deep learning dua tahap. Sistempertama-tama menggunakan model deteksi objek YOLOv11 untukmelokalisasi area tulisan tangan, kemudian menerapkan duamodel OCR berbasis Transformer, TrOCR dan Donut, untukmengekstraksi teks secara akurat. Aplikasi dikembangkanmenggunakan framework Streamlit sebagai antarmuka penggunayang interaktif. Fitur utama sistem mencakup kemampuan unggahdokumen, visualisasi perbandingan hasil OCR dari kedua model,dan fungsionalitas koreksi ejaan (spell-checking) untukmeningkatkan akurasi akhir. Proyek ini menghasilkan purwarupayang mampu meningkatkan efisiensi dan akurasi digitalisasidokumen akademik bertulisan tangan.
===================================================================================================================================
The manual digitization of Undergraduate Thesis Supervision Records is inefficient and prone to errors. This project develops a web application system to automate this process using Optical Character Recognition (OCR) with a two-stage deep learning approach. The system first utilizes the YOLOv11 object detection model to localize handwritten areas, then applies two Transformer-based OCR models, TrOCR and Donut, to accurately extract text. The application is developed using the Streamlit framework to provide an interactive user interface. Key features of the system include document upload capabilities, visual comparisons of OCR results from both models, and spell-checking functionality to enhance final accuracy. This project results in a prototype capable of improving the efficiency and accuracy of digitizing handwritten academic documents.

Item Type:	Monograph (Project Report)
Uncontrolled Keywords:	deep learning, donut, optical character recognition, streamlit, transformer, trocr, yolov11.
Subjects:	Q Science > QA Mathematics > QA76 Computer software Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques. Image analysis--Data processing.
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	Muhammad Ivan Ardianadi Afiat
Date Deposited:	09 Jan 2026 06:54
Last Modified:	09 Jan 2026 06:54
URI:	http://repository.its.ac.id/id/eprint/129411

Actions (login required)

View Item