prikaz prve stranice dokumenta Raspoznavanje oštećenja prometnica uporabom postupaka strojnog učenja
Access restricted to students and staff of home institution
undergraduate thesis
Raspoznavanje oštećenja prometnica uporabom postupaka strojnog učenja
2022. urn:nbn:hr:200:774928
Kaučić, Luka
Josip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
Department of Software Engineering
Chair of Programming Languages and Systems

Cite this document

Kaučić, L. (2022). Raspoznavanje oštećenja prometnica uporabom postupaka strojnog učenja (Undergraduate thesis). Retrieved from https://urn.nsk.hr/urn:nbn:hr:200:774928

Kaučić, Luka. "Raspoznavanje oštećenja prometnica uporabom postupaka strojnog učenja." Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2022. https://urn.nsk.hr/urn:nbn:hr:200:774928

Kaučić, Luka. "Raspoznavanje oštećenja prometnica uporabom postupaka strojnog učenja." Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2022. https://urn.nsk.hr/urn:nbn:hr:200:774928

Kaučić, L. (2022). 'Raspoznavanje oštećenja prometnica uporabom postupaka strojnog učenja', Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, accessed 21 March 2023, https://urn.nsk.hr/urn:nbn:hr:200:774928

Kaučić L. Raspoznavanje oštećenja prometnica uporabom postupaka strojnog učenja [Undergraduate thesis]. Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek; 2022 [cited 2023 March 21] Available at: https://urn.nsk.hr/urn:nbn:hr:200:774928

L. Kaučić, "Raspoznavanje oštećenja prometnica uporabom postupaka strojnog učenja", Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, Osijek, 2022. Available at: https://urn.nsk.hr/urn:nbn:hr:200:774928

Please login to the repository to save this object to your list.
Metadata
TitleRaspoznavanje oštećenja prometnica uporabom postupaka strojnog učenja
Title (english)Pothole detection utilising machine learning based approaches
AuthorLuka Kaučić
MentorBruno Zorić (mentor)
Committee memberBruno Zorić (predsjednik povjerenstva)
Committee memberGoran Knežević (član povjerenstva)
Committee memberEmmanuel Karlo Nyarko (član povjerenstva)
GranterJosip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
(Department of Software Engineering)
(Chair of Programming Languages and Systems)
Osijek
Defense date and country2022-09-21, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Artificial Intelligence
Abstract
U ovom radu najprije je objašnjeno kako dolazi do nastanka oštećenja na prometnicama te koje su opasnosti koja oštećenja donoese. Predstavljeni su podaci koji daju konkretnu informaciju o najčešćim uzrocima prometnih nezogda te faktorima koji na njih utječu. Problem klasifikacije prometnice kao neoštećene ili oštećene je predstavljen kao tema ovoga rada, dok su u teorijskom dijelu pojašnjene najčešće metode i okruženja koja se korište pri rješavanju takvih problema. Na temelju spoznaja iz prvog dijela rada, provedena je eksperimentalna anliza korištenjem Inception V3, MobileNet V2, InceptionResNet V2 te Xception modeli te podatkovni skup prilagođen binarnoj klasifikaciji. Određeni su hiperparametri za svaki ode navedenih modela te je izvršavanjem eksperimenta utvrđeno da je najbolje performanse dao Xception model. Sljedeći mogući koraci pri stvaranju digitaliziranog i autonomnog sustava uključuju višeklasnu klasifikaicju temeljenu na veličini oštećenja, kao i implementaciju sustava na prijenosne uređaje
Abstract (english)
Pothole detection utilising machine learning based approaches. This paper firstly outlines the basic information on potholes and their dangers. Next, data is presented, containing concrete information on traffic accidents and their causes. The topic of this paper was to give a series of deep learning methods for binary pothole classification. Based on facts presented in the first part of the paper, for the experimental analysis Inception V3, MobileNet V2, InceptionResNet V2, and Xception models are used and changed so they could fit the input data, considering its size and number of classes. Hyperparameters of all models were decided on, and after the experiment was run, it was concluded that the Xception model had the best performance. Next possible steps in developing a digitalized and autonomous system include a multiclass classification based on the size of the pothole, along with mobile phone integration of the system
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:200:774928
Study programmeTitle: Undergraduate University Study Programme in Computer Engineering
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 resourceText
File originBorn digital
Access conditionsAccess restricted to students and staff of home institution
Terms of useLicence In copyright icon
Created on2022-10-14 11:20:42