Segmentacija slika primjenom dubokih neuronskih mreža

undergraduate thesis

• Title (english): Image Segmentation Using Deep Neural Networks
• Author: Jurica Uglešić
• Mentor: Goran Delač (mentor)
• Committee member: Goran Delač (predsjednik povjerenstva)
• Committee member: Marin Kačan (član povjerenstva)
• Committee member: Helena Čeović (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2023-09-13, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Uvođenjem dubokih i konvolucijskih neuronskih mreža znanstvenici polako su usavršavali model međusobno povezanih neurona koji imitiraju rad ljudskog mozga. Takva otkrića potaknula su razvoj metoda i algoritama koji ubrzavaju poslove i olakšavaju život ljudi. Najpoznatija metoda računalnog vida je detekcija objekta koja je postavila osnovu za segmentaciju slika, makar to bila semantička, panoptička ili segmentacija instanci. Zbog segmentacije i njezine preciznosti, automatizirana je provjera kvalitete hrane, brže se provode dijagnoze i počinju se proizvoditi automobili koji neće zahtjevati prisutnost vozača za volanom.

Abstract (english)

With the introduction of deep and convolutional neural networks scientists have been slowly perfecting the model of mutually connected neurons which imitate the function of a human brain. Those discoveries encouraged the development of methods and algorithms which speed up tasks and make people's lives easier. Most known computer vision method is object detection which laid the foundation for image segmentation, whether it was semantic, panoptic or instance segmentation. Because of segmentation and its precision, food quality check has been automated, diagnoses are carried out faster and automobiles which do not require the presence of a driver behind the wheel are starting to be produced.

• Keywords: Umjetna inteligencija; računalni vid; konvolucijske neuronske mreže; neuron; algoritam; segmentacija; semantička; instanca; model; treniranje; validacija; preciznost
• Keywords (english): Artificial intelligence; computer vision; convolutional neural networks; neuron; algorithm; segmentation; semantic; instance; model; training; validation; precision
• Language: croatian
• URN:NBN: urn:nbn:hr:168:403877
• 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: 2023-10-11 10:50:33