prikaz prve stranice dokumenta Klasifikacija ambalaže primjenom izlučivanja značajki
Access restricted to students and staff of home institution
master's thesis
Klasifikacija ambalaže primjenom izlučivanja značajki
2020. urn:nbn:hr:190:379294
Karlović, Kristian
University of Rijeka
Faculty of Engineering
Department of Industrial Engineering and Management
Section of Production Equipment and Robotics

Cite this document

Karlović, K. (2020). Klasifikacija ambalaže primjenom izlučivanja značajki (Master's thesis). Rijeka: University of Rijeka, Faculty of Engineering. Retrieved from https://urn.nsk.hr/urn:nbn:hr:190:379294

Karlović, Kristian. "Klasifikacija ambalaže primjenom izlučivanja značajki." Master's thesis, University of Rijeka, Faculty of Engineering, 2020. https://urn.nsk.hr/urn:nbn:hr:190:379294

Karlović, Kristian. "Klasifikacija ambalaže primjenom izlučivanja značajki." Master's thesis, University of Rijeka, Faculty of Engineering, 2020. https://urn.nsk.hr/urn:nbn:hr:190:379294

Karlović, K. (2020). 'Klasifikacija ambalaže primjenom izlučivanja značajki', Master's thesis, University of Rijeka, Faculty of Engineering, accessed 24 December 2024, https://urn.nsk.hr/urn:nbn:hr:190:379294

Karlović K. Klasifikacija ambalaže primjenom izlučivanja značajki [Master's thesis]. Rijeka: University of Rijeka, Faculty of Engineering; 2020 [cited 2024 December 24] Available at: https://urn.nsk.hr/urn:nbn:hr:190:379294

K. Karlović, "Klasifikacija ambalaže primjenom izlučivanja značajki", Master's thesis, University of Rijeka, Faculty of Engineering, Rijeka, 2020. Available at: https://urn.nsk.hr/urn:nbn:hr:190:379294

Please login to the repository to save this object to your list.
Metadata
TitleKlasifikacija ambalaže primjenom izlučivanja značajki
Title (english)Packaging classification through the use of feature extraction
AuthorKristian Karlović
MentorZlatan Car (mentor)
Committee memberNeven Lovrin (predsjednik povjerenstva)
Committee memberSanjin Braut (član povjerenstva)
Committee memberZlatan Car (član povjerenstva)
GranterUniversity of Rijeka
Faculty of Engineering
(Department of Industrial Engineering and Management)
(Section of Production Equipment and Robotics)
Rijeka
Defense date and country2020-09-23, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Mechanical Engineering
Abstract
Kroz ovaj diplomski rad govorit će se o primjeni algoritama za izlučivanje značajki u svrsi treniranja SVM (Support Vector Machines) modela. Potrebno je klasificirati ambalažu primjenom izlučivanja značajki. Klasifikacija se vrši nad limenkama, plastičnim i staklenim bocama koje su slikane na Tehničkom fakultetu Sveučilišta u Rijeci. Program je pisan u programskom jeziku Python, a glavni moduli koji su se koristili za ovaj diplomski rad su OpenCV i scikit-learn. Za izlučivanja značajki na ambalažama primjenjuju se SIFT, SURF i ORB algoritmi. Performanse ORB-a u otkrivanju značajki su iste kao i kod SIFT-a što znači da je bolji od SURF-a, a brži je za gotovo dva reda veličine. Za klasificiranje ambalaža koristi se SVM algoritam. SVM algoritam pokušava pronaći hiperravninu s najvećom mogućom marginom između vektora potpore. Postoje različite jezgre koje koristi SVM algoritam, kao naprimjer linearna, RBF i polinomska jezgra.
Abstract (english)
Through this thesis, we will talk about the application of algorithms for extracting features for the purpose of training SVM (Support Vector Machines) models. Packaging needs to be classified using feature extraction. Classification is performed on cans, plastic and glass bottles that are photographed at the Technical Faculty of the University of Rijeka. The program is written in the Python programming language, and the main modules used for this thesis are OpenCV and scikitlearn. SIFT, SURF and ORB algorithms are used to extract features on packaging. The performance of ORB in feature detection is the same as with SIFT which means it is better than SURF and is faster by almost two orders of magnitude. The SVM algorithm is used to classify packaging. The SVM algorithm tries to find the hyperplane with the largest possible margin between the support vectors. There are different kernels used by SVM algorithm, such as linear, RBF, and polynomial kernel.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:190:379294
Study programmeTitle: Mechanical Engineering
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra inženjer/inženjerka strojarstva (magistar/magistra inženjer/inženjerka strojarstva)
Type of resourceText
File originBorn digital
Access conditionsAccess restricted to students and staff of home institution
Terms of useLicence In copyright icon
Created on2021-08-27 10:17:44