prikaz prve stranice dokumenta Praćenje više gibajućih objekata filtrom gustoće vjerojatnosti hipoteza
No public access
master's thesis
Praćenje više gibajućih objekata filtrom gustoće vjerojatnosti hipoteza
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2017. urn:nbn:hr:168:584002
Bićanić, Borna
University of Zagreb
Faculty of Electrical Engineering and Computing

Institutional repository: FER Repository

Cite this document

Bićanić, B. (2017). Praćenje više gibajućih objekata filtrom gustoće vjerojatnosti hipoteza (Master's thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:584002

Bićanić, Borna. "Praćenje više gibajućih objekata filtrom gustoće vjerojatnosti hipoteza." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2017. https://urn.nsk.hr/urn:nbn:hr:168:584002

Bićanić, Borna. "Praćenje više gibajućih objekata filtrom gustoće vjerojatnosti hipoteza." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2017. https://urn.nsk.hr/urn:nbn:hr:168:584002

Bićanić, B. (2017). 'Praćenje više gibajućih objekata filtrom gustoće vjerojatnosti hipoteza', Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 03 June 2024, https://urn.nsk.hr/urn:nbn:hr:168:584002

Bićanić B. Praćenje više gibajućih objekata filtrom gustoće vjerojatnosti hipoteza [Master's thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2017 [cited 2024 June 03] Available at: https://urn.nsk.hr/urn:nbn:hr:168:584002

B. Bićanić, "Praćenje više gibajućih objekata filtrom gustoće vjerojatnosti hipoteza", Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2017. Available at: https://urn.nsk.hr/urn:nbn:hr:168:584002

Metadata
TitlePraćenje više gibajućih objekata filtrom gustoće vjerojatnosti hipoteza
Title (english)Multiple Moving Object Tracking Using the Probability Hypothesis Density Filter
AuthorBorna Bićanić
MentorIvan Petrović (mentor)
Committee memberIvan Petrović (predsjednik povjerenstva)
Committee memberMarija Seder (član povjerenstva)
Committee memberIvan Marković (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2017-07-17, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Electrical Engineering
Abstract
sažetakPraćenje više gibajućih objekata kompleksan je problem gdje se osim izazova praćenja pojedinačnoga objekta javljaju i problemi poput nastajanja i nestajanja objekata te nesi-gurnosti pridjeljivanja mjerenja objektima. Rješavanjem toga problema pomoću teorije slučajnih konačnih skupova izbjegava se eksplicitno pridjeljivanje mjerenja objektima tre-tirajući ih kao slučajne skupove gdje su stanja i kardinalitet slučajne varijable. Ovaj pris-tup daje osnovu za primjenu optimalnog Bayesovog filtra za praćenje više gibajućih ob-jekata koji je, međutim, računski vrlo složen te mu se često pristupa aproksimacijama poput filtra gustoće vjerojatnosti hipoteza. U ovom radu implementiran je filter gustoće vjerojatnosti hipoteza (PHD), kao i kardinalizirani filter gustoće vjerojatnosti hipoteza (CPHD). Algoritmi su implementirani u programskom sustavu Matlab, unutar kojega su i testirani te uspoređeni na temelju metrike optimalnog pridruživanja uzoraka (OSPA).
Abstract (english)
Tracking of a multiple moving objects is a very complex problem, where besides the chal-lenges of tracking single object, problems arise from appearance and disappearance of objects and uncertainty of association between measurements and objects. By solving this problem via random finite sets theory, explicit assignment of measurements to objects is avoided treating them as random sets where both states and cardinality is random varia-ble. This approach provides the basis for the application of optimal Bayes filter to the problem of tracking multiple moving objects, which is, however, computationally very complex and, often, approximations like probability hypothesis density filter are made. In this master’s thesis, the probability hypothesis density filter for tracking multiple objects is implemented. The algorithm is realized using Gaussian distribution mixtures and tested in simulations within Matlab programming system.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:584002
Study programmeTitle: Electrical Engineering and Information Technology
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra inženjer/inženjerka elektrotehnike i informacijske tehnologije (magistar/magistra inženjer/inženjerka elektrotehnike i informacijske tehnologije)
Type of resourceText
File originBorn digital
Access conditionsClosed access
Terms of useLicence In copyright icon
Public note
Created on2019-03-13 19:04:32