prikaz prve stranice dokumenta Poboljšanje performansi detektora objekata dodavanjem podataka o dubini slike koristeći oblak točaka
  Download
PDF 2.89 MB
master's thesis
Poboljšanje performansi detektora objekata dodavanjem podataka o dubini slike koristeći oblak točaka
Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2022. urn:nbn:hr:200:786921
Zmeškal, Ivan
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

Institutional repository: Faculty of Electrical Engineering, Computer Science and Information Technology Osijek

Cite this document

Zmeškal, I. (2022). Poboljšanje performansi detektora objekata dodavanjem podataka o dubini slike koristeći oblak točaka (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:200:786921

Zmeškal, Ivan. "Poboljšanje performansi detektora objekata dodavanjem podataka o dubini slike koristeći oblak točaka." Master's 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:200:786921

Zmeškal, Ivan. "Poboljšanje performansi detektora objekata dodavanjem podataka o dubini slike koristeći oblak točaka." Master's 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:200:786921

Zmeškal, I. (2022). 'Poboljšanje performansi detektora objekata dodavanjem podataka o dubini slike koristeći oblak točaka', Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, accessed 19 May 2024, https://urn.nsk.hr/urn:nbn:hr:200:786921

Zmeškal I. Poboljšanje performansi detektora objekata dodavanjem podataka o dubini slike koristeći oblak točaka [Master's thesis]. Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek; 2022 [cited 2024 May 19] Available at: https://urn.nsk.hr/urn:nbn:hr:200:786921

I. Zmeškal, "Poboljšanje performansi detektora objekata dodavanjem podataka o dubini slike koristeći oblak točaka", Master's 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:200:786921

Metadata
TitlePoboljšanje performansi detektora objekata dodavanjem podataka o dubini slike koristeći oblak točaka
Title (english)Improving object detector performance by adding depth information using a point cloud
AuthorIvan Zmeškal
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 country2022-12-21, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Electrical Engineering
Abstract
U okviru ovog diplomskog rada opisane su neke od postojećih metoda detekcije objekata koje se primjenjuju u autonomnim vozilima. Za potrebe izrade rada korišteni su podatci dobiveni sa kamera koje se nalaze na prednjoj strani vozila i podatci dobiveni LiDAR senzorom, dok su korišteni detektori temeljeni na YOLOv3 i YOLOv4 algoritmima. Pomoću RGB slika i LiDAR podataka koji su naknadno vizualizirani, kreiran je novi skup podataka s RGBD slikama. RGBD slike su nastale na način da je RGB slikama kao četvrti kanal dodana informacija o dubini. Testiranjem detektora nad sva tri skupa podataka zaključeno je da informacije o dubini mogu rezultirati poboljšanjem performansi detektora. TakoĎer, rezultati testiranja su pokazali i odreĎene nedostatke detektora, kao najveći nedostatak se ističe relativno niska razina detekcije objekata što ostavlja prostor za potencijalna daljnja poboljšanja.
Abstract (english)
As part of this thesis, some of the existing object detection methods used in autonomous vehicles are described. For the purposes of creating the work, data obtained from cameras located on the front of the vehicle and data obtained from a LiDAR sensor were used, while detectors based on YOLOv3 and YOLOv4 algorithms were used. Using RGB images and LiDAR data that were subsequently visualized, a new subset of RGBD image data was created. RGBD images were created by adding depth information to RGB images as fourth channel. By testing the detector on all three subsets of data, it was concluded that depth information can result in improved detector performance. Also, the test results showed certain shortcomings of the detector, the biggest drawback being the relatively low level of object detection, which leaves room for potential further improvements.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:200:786921
Study programmeTitle: UNIVERSITY STUDY AUTOMOBILE COMPUTER AND COMMUNICATION
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra inženjer/inženjerka automobilskog računarstva i komunikacija (magistar/magistra inženjer/inženjerka automobilskog računarstva i komunikacija)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Created on2023-02-17 10:27:52