Prototip mobilne aplikacije za digitalno beleženje podatkov v živinoreji

Dežman, Jure (author), Rožanc, Igor (mentor)

V diplomskem delu smo razvili prototip preproste mobilne aplikacije, ki omogoča digitalno vnašanje, shranjevanje, urejanje in pregledovanje podatkov o vsakodnevnih kmetijskih opravilih, predvsem na področju mlečne pridelave. Aplikacija prikaže podatke v različnih oblikah, kot so seznami, tabele, tortni in stolpčni grafi. Poseben poudarek je na uporabniški izkušnji in možnosti digitalizacije obstoječih fizičnih dokumentov s pomočjo OCR in umetne inteligence. Za zajem podatkov iz slik smo uporabili model GPT-4.1 mini preko OpenAI API ter funkcionalnost zaznavanja robov dokumentov z OpenCV.js. Aplikacija je sestavljena iz dveh delov. Čelni del je razvit z ogrodjem Angular in dopolnjen s knjižnicama PrimeNG za prikaz uporabniškega vmesnika in chart.js za prikaz grafov. Zaledni del temelji na ASP.NET Core in uporablja Entity Framework Core za komunikacijo s SQL podatkovno bazo. Komunikacija med čelnim in zalednim delom poteka prek REST API-ja. Končni rezultat je prototip aplikacije, ki kmetom omogoča strukturiran pregled nad analizami, dostavami in prodajami mleka ter njihovimi strankami.

• Language: Slovenian
• Keywords: kmetija, mobilna aplikacija, optično prepoznavanje znakov, umetna inteligenca, ASP .NET Core, Angular
• Work type: Bachelor thesis/paper
• Typology: 2.11 - Undergraduate Thesis
• Organization: FRI - Faculty of Computer and Information Science
• Year: 2025
• PID: 20.500.12556/RUL-172537
• COBISS.SI-ID: 248859395
• Publication date in RUL: 08.09.2025

Secondary language

• Language: English
• Title: Prototype of a mobile application for digital data recording in agriculture

Abstract

This thesis presents the development of a prototype mobile application designed to support the digital recording, storage, editing, and review of daily agricultural activities, with a focus on dairy production. The application displays data in various formats, such as lists, tables, pie charts, and bar charts. An emphasis is placed on user experience and the ability to digitize physical documents using OCR and AI. For data extraction from images, the GPT-4.1 mini model was used via the OpenAI API, combined with real-time edge detection functionality implemented using OpenCV.js. The application consists of two parts. The frontend is built with the Angular framework and enhanced with libraries such as PrimeNG for UI components and Chart.js for data visualization. The backend is developed in ASP.NET Core and uses Entity Framework Core for database access. Communication between the frontend and backend is done via a REST API. The final result is a functional application prototype that enables farmers to maintain a structured overview of milk analyses, deliveries, sales, and customer records.

• Keywords: farm, mobile application, optical character recognition, artificial intelligence, ASP .NET Core, Angular