prikaz prve stranice dokumenta Beskontaktni sustav za prepoznavanje otisaka prstiju
  Download
PDF 2.04 MB
master's thesis
Beskontaktni sustav za prepoznavanje otisaka prstiju
2023. urn:nbn:hr:200:390866
Gojić, Andreja
Josip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
Department of Computer Engineering and Automation
Chair of Automation and Robotics

Cite this document

Gojić, A. (2023). Beskontaktni sustav za prepoznavanje otisaka prstiju (Master's thesis). Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek. Retrieved from https://urn.nsk.hr/urn:nbn:hr:200:390866

Gojić, Andreja. "Beskontaktni sustav za prepoznavanje otisaka prstiju." Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2023. https://urn.nsk.hr/urn:nbn:hr:200:390866

Gojić, Andreja. "Beskontaktni sustav za prepoznavanje otisaka prstiju." Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2023. https://urn.nsk.hr/urn:nbn:hr:200:390866

Gojić, A. (2023). 'Beskontaktni sustav za prepoznavanje otisaka prstiju', Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, accessed 31 January 2025, https://urn.nsk.hr/urn:nbn:hr:200:390866

Gojić A. Beskontaktni sustav za prepoznavanje otisaka prstiju [Master's thesis]. Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek; 2023 [cited 2025 January 31] Available at: https://urn.nsk.hr/urn:nbn:hr:200:390866

A. Gojić, "Beskontaktni sustav za prepoznavanje otisaka prstiju", Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, Osijek, 2023. Available at: https://urn.nsk.hr/urn:nbn:hr:200:390866

Please login to the repository to save this object to your list.
Metadata
TitleBeskontaktni sustav za prepoznavanje otisaka prstiju
Title (english)Contactless fingerprint recognition system
AuthorAndreja Gojić
MentorRatko Grbić (mentor)
Committee memberRatko Grbić (predsjednik povjerenstva)
Committee memberMarijan Herceg (član povjerenstva)
Committee memberMario Vranješ (član povjerenstva)
GranterJosip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
(Department of Computer Engineering and Automation)
(Chair of Automation and Robotics)
Osijek
Defense date and country2023-09-26, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Artificial Intelligence
Abstract
U ovom diplomskom radu izrađen je beskontaktni sustav za prepoznavanje otisaka prstiju. Prije svega dan je pregled dosadašnjih rješenja koji se temelje na tri metode, metoda za izdvajanje detalja, dubokim učenjem te kombinacije tih dviju metoda. Zatim je opisano predloženo rješenje u ovom radu. Budući da nije pronađen odgovarajući skup podataka za učenje neuronske mreže, izrađen je vlastiti skup podataka koji je služio za treniranje i testiranje neuronske mreže. Sustav se temelji na sijamskoj neuronskoj mreži koja se sastoji od dvije neuronske mreže. Sijamska neuronska mreža na ulazu prima dvije RGB slike na kojima su prikazani otisci prstiju, slike se predaju neuronskoj mreži čiji su izlazi vektori značajki. Zatim se računa euklidska udaljenost između dobivenih vektora koja predstavlja mjeru sličnosti u intervalu između 0 i 1. Izlaz iz mreže će biti bliže 1 ukoliko su otisci prstiju različiti ili bliže 0 ukoliko su otisci prstiju sličniji. Istrenirani model je implementiran na NVIDIA Jetson Nano te je izrađeno fizičko okruženje s ugrađenom kamerom i LED trakom koja služi za slikanje otisaka prstiju. Izrađena sijamska neuronska mreža evaluirana je na skupu podataka za testiranje i iz rezultata testiranja je vidljivo kako ima mjesta za napredak.
Abstract (english)
In this master thesis, a non-contact system for recognizing fingerprints was developed. Firstly, an overview of the previous solutions based on three methods, methods for extracting details, deep learning and a combination of these two methods is given. Then the solution proposed in this paper is described. Since no suitable data set was found for learning the neural network, an own data set was created that was used for training and testing the neural network. The system is based on a Siamese neural network consisting of two neural networks. Siamese neural network at the input receives two RGB images on which fingerprints are displayed, the images are given to the neural network whose outputs are feature vectors. Then the Euclidean distance between the obtained vectors is calculated, which represents a measure of similarity in the interval between 0 and 1. The output from the network will be closer to 1 if the fingerprints are different or closer to 0 if the fingerprints are more similar. The trained model was implemented on NVIDIA Jetson Nano and a physical environment was created with a built-in camera and an LED strip used for fingerprint imaging. The constructed Siamese neural network was evaluated on the test data set and the test results show that there is room for improvement.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:200:390866
Study programmeTitle: Graduate University Study Programme in Computer Engineering
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Created on2023-10-09 09:49:23