Mobilna aplikacija za klasifikaciju vrste cvijeća sa slike korištenjem metoda dubokog učenja

undergraduate thesis

• Title (english): Mobile application for flower species classification from an image using deep learning methods
• Author: Luka Raić
• Mentor: Alan Jović (mentor)
• Committee member: Alan Jović (predsjednik povjerenstva)
• Committee member: Hrvoje Nuić (član povjerenstva)
• Committee member: Eda Jovičić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2024-07-05, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovom završnom radu predstavljena je mobilna aplikacija koja omogućuje klasifikaciju slika cvijeća pomoću ugrađenog modela dubokog učenja. U sklopu učenja korišten je podatkovni skup Oxford 102 Flower koji sadrži 102 kategorije različitih cvjetova koji se često mogu naći u Ujedinjenom Kraljevstvu. Za razvoj modela dubokog učenja korištena su dvije arhitekture: višeslojni perceptron i konvolucijska neuronska mreža. Model zasnovan na arhitekturi višeslojnog perceptrona je na skupu za testiranje imao točnost od 39,15%, dok je model zasnovan na arhitekturi konvolucijske neuronske mreže imao točnost od 72,10% pa je on ugrađen u aplikaciju. Sama mobilna aplikacija je napisana u programskom jeziku Kotlin te je namijenjena za operacijski sustav Android.

Abstract (english)

In this bachelor thesis, a mobile application that enables flower image classification using an embedded deep learning model is presented. The Oxford 102 Flower dataset, which contains 102 categories of different flowers commonly found in the United Kingdom, was used for training. Two architectures were used for the development of the deep learning model: a multilayer perceptron and a convolutional neural network. The model based on the multilayer perceptron architecture achieved an accuracy of 39.15% on the test set, while the model based on the convolutional neural network architecture achieved an accuracy of 72.10%, and thus it was integrated into the application. The mobile application itself is written in the Kotlin programming language and is intended for the Android operating system.

• Keywords: klasifikacija; duboko učenje; konvolucijska neuronska mreža; višeslojni perceptron; Keras; TensorFlow; Android; Kotlin
• Keywords (english): classification; deep learning; convolutional neural network; multilayer perceptron; Keras; TensorFlow; Android; Kotlin
• Language: croatian
• URN:NBN: urn:nbn:hr:168:511502
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2024-10-12 21:58:56