diplomsko delo
Lara Anžur (Author), Peter Peer (Mentor), Tim Oblak (Co-mentor)

Abstract

Prstni odtisi so izjemno zanesljiv način identifikacije posameznikov v forenziki, saj so edinstveni in trajni. Klasične metode prepoznave prstnih odtisov, ki uporabljajo strojno učenje, se pogosto soočajo z izzivi pri obdelavi nizkokakovostnih primerov, kar zahteva asistenco forenzikov. Vedno bolj priljubljena postaja uporaba globokega učenja, ki odpravlja nekatere od omejitev klasičnih metod, vendar je na tem področju še vedno razvitih premalo rešitev. V tej diplomski nalogi smo razvili model, ki temelji na siamskih nevronskih mrežah (SNN) v kombinaciji z arhitekturo ResNet34, kar nam je omogočilo učinkovito primerjavo prstnih sledi v latentnem prostoru. Osnovni model smo dodatno izboljšali z integracijo prostorskih transformatorskih mrež (STN), ki zagotavljajo rotacijsko invariantnost, ter z vključitvijo domenskega znanja o minucijah, s čimer smo v proces dodali dodatne relevantne informacije. Metode smo ovrednotili na več javno dostopnih podatkovnih zbirkah, pri čemer je naš model dosegel višjo stopnjo natančnosti v primerjavi z nekaterimi klasičnimi metodami, kar potrjuje tudi perspektivnost globokega učenja na področju identifikacije prstnih sledi.

Keywords

iskanje ujemanj prstnih sledi;identifikacija;siamske nevronske mreže;prostorske transformatorske mreže;trojna izguba;univerzitetni študij;diplomske naloge;

Data

Language: Slovenian
Year of publishing:
Typology: 2.11 - Undergraduate Thesis
Organization: UL FRI - Faculty of Computer and Information Science
Publisher: [L. Anžur]
UDC: 004.93:57.087.1(043.2)
COBISS: 209149443 Link will open in a new window
Views: 104
Downloads: 54
Average score: 0 (0 votes)
Metadata: JSON JSON-RDF JSON-LD TURTLE N-TRIPLES XML RDFA MICRODATA DC-XML DC-RDF RDF

Other data

Secondary language: English
Secondary title: Fingerprint recognition using deep learning
Secondary abstract: Fingerprints are an extremely reliable method of identifying individuals in forensic science, as they are unique and permanent. Classical fingerprint recognition methods that use machine learning often face challenges when processing low-quality samples, which requires forensic experts' assistance. The use of deep learning, which overcomes some of the limitations of classical methods, is becoming increasingly popular, but there are still too few developed solutions in this field. In this thesis, we developed a model based on Siamese neural networks (SNN) combined with the ResNet34 architecture, enabling us to efficiently compare fingerprints in latent space. We further enhanced the basic model by integrating spatial transformer networks (STN), which ensure rotational invariance, and incorporating domain knowledge about minutiae, adding additional relevant information to the process. We evaluated the methods on several publicly available datasets, where our model achieved a higher level of accuracy compared to some classical methods, confirming the potential of deep learning in the field of fingerprint identification.
Secondary keywords: fingermark matching;identification;deep learning;Siamese neural networks;spatial transformer network;triplet loss;computer and information science;diploma;Biometrična identifikacija;DNK prstni odtis;Globoko učenje (strojno učenje);Računalništvo;Univerzitetna in visokošolska dela;
Type (COBISS): Bachelor thesis/paper
Study programme: 1000468
Embargo end date (OpenAIRE): 1970-01-01
Thesis comment: Univ. v Ljubljani, Fak. za računalništvo in informatiko
Pages: 1 spletni vir (1 datoteka PDF (52 str.))
ID: 24892309