Μελέτη και υλοποίηση Android εφαρμογής διαχείρισης προσωπικών εξόδων με δυνατότητα OCR αποδείξεων

Χριστοδουλόπουλος, Διονύσιος Παναγιώτης

Study and implementation of Android expense management application with receipt OCR capabilities

Bachelor Dissertation

Author

Χριστοδουλόπουλος, Διονύσιος Παναγιώτης

Date

2026-07

Abstract

This thesis presents the design, development, and evaluation of an Android application for personal expense management that incorporates Optical Character Recognition (OCR) technology to automate the recording of financial transactions. The primary objective of the application is to simplify the expense tracking process by allowing users to scan purchase receipts using their mobile device's camera and automatically extract essential information, such as the store name, transaction amount, and expense category. The application was developed using modern Android technologies and architectural principles, including Kotlin, Jetpack Compose, Clean Architecture, MVVM, Room Database, CameraX, and Dagger Hilt. For text recognition, Google's Cloud Vision API was integrated to provide accurate OCR capabilities, particularly for Greek-language receipts. The extracted data are stored locally, enabling users to organize expenses, monitor budgets, visualize spending patterns through statistical charts, and export financial records in CSV format. The implementation emphasizes maintainability, scalability, usability, and data privacy by adopting contemporary software engineering practices and local data storage. The evaluation of the application demonstrates that OCR significantly reduces manual data entry while maintaining satisfactory recognition accuracy under typical usage conditions. The resulting system offers a practical and efficient solution for personal financial management while providing a solid foundation for future enhancements, including AI-assisted categorization, cloud synchronization, and advanced financial analytics.

Department

Σχολή Τεχνολογιών Πληροφορικής και Επικοινωνιών. Τμήμα Πληροφορικής

Number of pages

Language

Greek

URI

https://dione.lib.unipi.gr/xmlui/handle/unipi/19660

Collections

Τμήμα Πληροφορικής

Show full item record

Except where otherwise noted, this item's license is described as
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα