Robust visual place recognition using deep representations and sequence-based image matching

Maltar, Jurica

prikaz prve stranice dokumenta Robust visual place recognition using deep representations and sequence-based image matching

Download
PDF 16.99 MB

doctoral thesis

Robust visual place recognition using deep representations and sequence-based image matching

Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2023. urn:nbn:hr:168:950209

Maltar, Jurica

University of Zagreb
Faculty of Electrical Engineering and Computing
Department of Control and Computer Engineering

Cite this document

APA 6th Edition

Maltar, J. (2023). Robust visual place recognition using deep representations and sequence-based image matching (Doctoral thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:950209

MLA 8th Edition

Maltar, Jurica. "Robust visual place recognition using deep representations and sequence-based image matching." Doctoral thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2023. https://urn.nsk.hr/urn:nbn:hr:168:950209

Chicago 17th Edition

Harvard

Maltar, J. (2023). 'Robust visual place recognition using deep representations and sequence-based image matching', Doctoral thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 04 October 2023, https://urn.nsk.hr/urn:nbn:hr:168:950209

Vancouver

Maltar J. Robust visual place recognition using deep representations and sequence-based image matching [Doctoral thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2023 [cited 2023 October 04] Available at: https://urn.nsk.hr/urn:nbn:hr:168:950209

IEEE

J. Maltar, "Robust visual place recognition using deep representations and sequence-based image matching", Doctoral thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2023. Available at: https://urn.nsk.hr/urn:nbn:hr:168:950209

Cite this item: https://urn.nsk.hr/urn:nbn:hr:168:950209

Please login to the repository to save this object to your list.

Metadata

Title	Robust visual place recognition using deep representations and sequence-based image matching
Title (croatian)	Robusno vizualno prepoznavanje mjesta uporabom dubokih reprezentacija i uparivanja sljedova slika
Author	Jurica Maltar MBZ: 366813
Mentor	Ivan Marković (mentor)
Mentor	Domagoj Matijevic (komentor)
Committee member	Ivan Petrović (predsjednik povjerenstva)
Committee member	Siniša Šegvić (član povjerenstva)
Committee member	Kristian Sabo (član povjerenstva)
Granter	University of Zagreb Faculty of Electrical Engineering and Computing (Department of Control and Computer Engineering) Zagreb
Defense date and country	2023-05-15, Croatia
Scientific / art field, discipline and subdiscipline	TECHNICAL SCIENCES Computing
Universal decimal classification (UDC)	004 - Computer science and technology. Computing. Data processing
Abstract	The doctoral thesis examines the problem of visual place recognition. During visual place recognition, negative impacts such as different environmental conditions, different viewpoints of the camera and perceptual aliasing, affect the system’s performances. Therefore, the thesis proposes a novel sequence-based place matching method. Such a place matching is more robust than the naïve approach because data is sequential and captured during the vehicle’s traversal around a route. Also, the thesis proposes a robust visual representation for a place image by using fine-tuned deep convolutional neural networks. Feature maps extracted with a convolutional neural network are further refined with the mutual information-based feature selection technique. Finally, to not be an end in itself, the implemented system is additionally adapted and used in the context of the famous robotic problem – simultaneous localization and mapping – for the purpose of loop closing detection where even better performances are obtained. Scientific contributions of the thesis are: Sequence-based visual place matching method that uses directed acyclic graphs and single source shortest path. Method for robust visual place recognition with deep representations that uses softmax regression and mutual information-based feature selection. Procedure for adapting sequence-based visual place recognition method for loop closing in simultaneous localization and mapping algorithms.
Abstract (croatian)	Doktorski se rad odnosi na problem vizualnog prepoznavanja mjesta. Pri vizualnom prepoznavanju mjesta, negativni čimbenici kao što su različiti okolišni uvjeti, različita gledišta kamere te vizualna sličnost različitih mjesta, utječu na performanse sustava. Stoga ovaj doktorski rad predstavlja novu metodu prepoznavanja mjesta uparivanjem sljedova slika. Takvo je uparivanje robusnije od uobičajenog pristupa jer su podaci nastali pri obilasku vozila oko nekog područja sekvencijalne prirode. Također, doktorski rad predstavlja robusnu vizualnu reprezentaciju slike mjesta korištenjem dubokih modela dobivenu prilagodbom konvolucijskih neuronskih mreža. Dobivene se globalne značajke slike mjesta dodatno pročišćuju korištenjem odabira značajki zasnovanom na uzajamnom sadržaju informacije. Naposljetku, kako vizualno prepoznavanje ne bi bilo problem sam za sebe, implementirani se sustav dodatno prilagođava i koristi u sklopu poznatog robotičkog problema – istovremene lokalizacije i kartiranja – u svrhu detekcije zatvaranja petlje čime su postignute još bolje performanse u tome problemu. Znanstveni doprinosi doktorskog rada su: Metoda vizualnog uparivanja mjesta zasnovana na sljedovima slika koja koristi usmjerene acikličke grafove i najkraći put iz jednog izvora. Metoda za robusno vizualno prepoznavanje mjesta dubokim reprezentacijama koja koristi softmax regresiju i odabir značajki zasnovan na uzajamnom sadržaju informacije. Postupak prilagodbe metode vizualnog prepoznavanja mjesta zasnovane na sljedovima slika za zatvaranje petlje u algoritmima istovremene lokalizacije i kartiranja.
Keywords
Keywords (croatian)
Language	english
URN:NBN	urn:nbn:hr:168:950209
Promotion	2023
Study programme	Title: Computer Science Study programme type: university Study level: postgraduate Academic / professional title: Doktor znanosti (Doktor znanosti)
Type of resource	Text
Extent	xi, 114 str. : graf. prikazi ; 31 cm.
File origin	Born digital
Access conditions	Open access
Terms of use
Repository	FER Repository
Created on	2023-09-18 15:32:17