Automatizirano prepoznavanje proljetnica Pulsatilla grandis na fotografijama koje su snimane s dronom u prirodnom staništu

undergraduate thesis

• Title (english): Automated recognition of Pulsatilla grandis spring flowers in photographs taken with a drone in its natural habitat
• Author: Bruno Jerić
• Mentor: Mirko Randić (mentor)
• Committee member: Mirko Randić (predsjednik povjerenstva)
• Committee member: Bruno Blašković (član povjerenstva)
• Committee member: Frano Škopljanac-Mačina (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2021-07-13, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovom radu ukratko su opisani neki algoritmi koji su potakli razvoj računalnog vida.
Opisana je implementacija računalnog vida koja koristi tehnike dubokog učenja i
konvolucijske neuronske mreže. Objašnjena je ideja i koncept neuronskih mreža te je
detaljnije opisan Faster R-CNN model. U praktičnom dijelu rada opisan je proces pripreme
podataka u formi slika koje prikazuju predmet interesa – cvijet velecvjetna sasa te proces
treniranja mreže. Problem detekcije riješen je Faster R-CNN-om, mrežom za klasifikaciju i
lokalizaciju. Trenirani model implementiran Faster R-CNN mrežom ostvaruje mAP od
85.3%.

Abstract (english)

This paper briefly describes some algorithms that have sparked the development of computer
vision. The implementation of computer vision using deep learning techniques and
convolutional neural networks is described. The idea and concept of neural networks are
explained and the Faster R-CNN model is described in more detail. The practical part of the
paper describes the process of data preparation in the form of images showing the object of
interest - velecvjetna sasa flower and training process of the network. The detection problem
was solved by Faster R-CNN, a classification and localization network. The trained model
implemented by the Faster R-CNN network achieves an mAP of 85.3%.

• Keywords: računalni vid; neuronske mreže; konvolucijske neuronske mreže; velecvjetna sasa; Faster R-CNN
• Keywords (english): computer vision; neural networks; convolutional neural networks; velecvjetna sasa; Faster R-CNN
• Language: croatian
• URN:NBN: urn:nbn:hr:168:099898
• 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: 2023-01-14 20:40:50