prikaz prve stranice dokumenta Kamera za očitavanje registarske tablice
Access restricted to authenticated users
master's thesis
Kamera za očitavanje registarske tablice
Split: University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, 2022. urn:nbn:hr:179:141471
Radić, Ante
University of Split
Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture
Department of Electronics and Computing

Institutional repository: Repository of the Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture

Cite this document

Radić, A. (2022). Kamera za očitavanje registarske tablice (Master's thesis). Split: University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture. Retrieved from https://urn.nsk.hr/urn:nbn:hr:179:141471

Radić, Ante. "Kamera za očitavanje registarske tablice." Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, 2022. https://urn.nsk.hr/urn:nbn:hr:179:141471

Radić, Ante. "Kamera za očitavanje registarske tablice." Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, 2022. https://urn.nsk.hr/urn:nbn:hr:179:141471

Radić, A. (2022). 'Kamera za očitavanje registarske tablice', Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, accessed 29 June 2024, https://urn.nsk.hr/urn:nbn:hr:179:141471

Radić A. Kamera za očitavanje registarske tablice [Master's thesis]. Split: University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture; 2022 [cited 2024 June 29] Available at: https://urn.nsk.hr/urn:nbn:hr:179:141471

A. Radić, "Kamera za očitavanje registarske tablice", Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, Split, 2022. Available at: https://urn.nsk.hr/urn:nbn:hr:179:141471

Metadata
TitleKamera za očitavanje registarske tablice
Title (english)License plate recognition camera
AuthorAnte Radić
MentorIvo Stančić https://orcid.org/0000-0001-9607-1791 (mentor)
GranterUniversity of Split
Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture
(Department of Electronics and Computing)
Split
Defense date and country2022-02-23, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Electrical Engineering
Electronics
Abstract
Detekcija tablica na automobilima ima široku primjenu u različitim industrijama i područjima svijeta. Stoga, postoji već mnoštvo sustava i rješenja za ovu primjenu. U ovom radu, realiziran je detektor tablica za vozila u mirovanju koristeći OpenMV H7 Plus kameru. Tehnike računalnog vida koje su korištene su OpenCV, Keras, TensorFlow Lite, Haar-ove kaskade, MicroPython.. Također su korišteni brojni alati za trening klasifikatora i neuronskih mreža, kao i za obradu slike i implementaciju sustava. Neki od njih su Cascade Trainer GUI-a, korišten za trening Haar-ovog klasifikatora i Edge Impulse, korišten za trening neuronskih mreža. Mikrokontroler je programiran u OpenMV IDE-u, koji je napravljen specifično za OpenMV kameru. Datoteke koje sadrže klasifikator i .tflite datoteke neuronskih mreža su učitane direktno u editor. OpenMV IDE radi s kamerom u realnom vremenu tako da je kompletna implementacija testirana direktno na kameri. Krajnji cilj je bio dokazati da se ova konkretna primjena može realizirati s OpenMV H7 Plus kamerom, tako da su krajnji rezultati pozitivni. Dodatna poboljšanja su moguća u vidu treninga s puno većim i kvalitetnijim setom podataka, automatiziranja detekcije i klasifikacije znakova s okvira tablice i omogućavanju rada noću.
Abstract (english)
License plate recognition is widely used in various industries and areas of the world. Therefore, there are already many systems and solutions for this application. In this paper, a plate detector for stationary vehicles was implemented using an OpenMV H7 Plus camera. The computer vision techniques used are OpenCV, Keras, TensorFlow Lite, Haar Cascades, MicroPython, etc. Numerous software tools were also used for training the classifiers and neural networks, as well as for image processing and system implementation. Some of them are the Cascade Trainer GUI, used to train Haar’s classifier, and Edge Impulse, used to train neural networks. The microcontroller is programmed using OpenMV IDE, which is made specifically for the OpenMV camera. Files containing the classifier and .tflite files of neural networks are loaded directly into the editor. The OpenMV IDE works with the camera in real time so the complete implementation is tested directly on the camera. The ultimate goal was to prove that this particular application can be done with the OpenMV H7 Plus camera, so the end results are positive. Additional improvements are possible in the form of training with a much larger and better data set, automating the detection and classification of characters from the table frame and enabling night work.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:179:141471
Study programmeTitle: Electronics and Computer Engineering
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra inženjer/inženjerka elektronike i računalnog inženjerstva (magistar/magistra inženjer/inženjerka elektronike i računalnog inženjerstva)
Type of resourceText
File originBorn digital
Access conditionsAccess restricted to authenticated users
Terms of useLicence In copyright icon
Created on2023-01-24 08:34:15