Skeniranje dokumentov s pametnim telefonom

BRODNIK, MATEJ (author), Čehovin Zajc, Luka (mentor)

Diplomsko delo se osredotoča na razvoj mobilne aplikacije za operacijski sistem Android, ki z uporabo računalniškega vida prepozna papirne dokumente in jih na pogledu kamere obriše. Algoritem prepoznavanja dokumentov smo razvili v programskem jeziku C++ z uporabo odprtokodne knjižnice OpenCV. Vhodne parametre algoritma smo objektivno izbrali na podlagi reprezentativne zbirke slik, ki smo jih zajeli. Razvit algoritem smo integrirali v mobilno aplikacijo na operacijskemu sistemu Android. Aplikacija omogoča skeniranje in shranjevanje dokumentov preko mobilnega uporabniškega vmesnika. Služi tudi kot odprtokoden primer integracije C++ OpenCV knjižnice z operacijskim sistemom Android.

• Language: Slovenian
• Keywords: mobilna aplikacija, računalniški vid, Android, OpenCV
• Work type: Bachelor thesis/paper
• Typology: 2.11 - Undergraduate Thesis
• Organization: FRI - Faculty of Computer and Information Science
• Year: 2023
• PID: 20.500.12556/RUL-144767
• COBISS.SI-ID: 147965699
• Publication date in RUL: 10.03.2023

Secondary language

• Language: English
• Title: Scanning documents using a smartphone

Abstract

This thesis focuses on the development of a mobile application for the Android operating system, which employs computer vision to identify paper documents and outline them in the camera view. We developed the document detecting algorithm in the C++ programming language using the open source library OpenCV. We objectively determined the input parameters of the algorithm based on a representative set of images that we captured. We integrated the developed algorithm into a mobile application on the Android operating system. The application allows the scanning and saving of documents through a mobile user interface. It also serves as an open source example of integrating the C++ OpenCV library with the Android operating system.

• Keywords: mobile application, computer vision, Android, OpenCV