Open AccessLicense plate recognition system based on principal component analysis and one-against-one multi-class support vector machine
Author(s) -
Kevin Kristofer Kosasih,
Winda Astuti,
Endra Oey
Publication year - 2020
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.179
H-Index - 26
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/426/1/012038
Subject(s) - artificial intelligence , support vector machine , license , computer science , feature extraction , computer vision , image processing , pattern recognition (psychology) , feature vector , principal component analysis , fast fourier transform , process (computing) , identification (biology) , feature (linguistics) , image (mathematics) , algorithm , linguistics , philosophy , botany , biology , operating system
The license plate number which identifies the original registered place of the vehicle, as well as the usage of each vehicle which belongs to a private entity, a government vehicle, and a commercial vehicle, is identified in this work. Vehicle plate detection by a camera that mounted on parking gate. Detection process starts with collecting data using the camera. This work consists two important parts, namely image processing and classification system, respectively. In the image processing, the plate image capture by the camera is extracted to the feature space which will use as the input to the classification system. The classification system is then will identify the extracted input to the plat recognition identification. The Fast Fourier Transform (FFT) and Support Vector Machine (SVM) are applied as feature extraction technique and classification methods, respectively. The results of the training and testing phase are 80% and 66%, respectively.