Open AccessDetection and Identification Indonesia License Plate Using Background Subtraction Based on Area
Author(s) -
Fitri Damayanti,
W. K. Dewi,
Eza Rahmanita,
Aeri Rachmad
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1569/2/022064
Subject(s) - background subtraction , license , segmentation , computer science , artificial intelligence , character (mathematics) , process (computing) , queue , identification (biology) , pattern recognition (psychology) , toll , feature extraction , computer vision , subtraction , mathematics , pixel , botany , geometry , genetics , biology , programming language , operating system , arithmetic
The increasing number of vehicles, especially four-wheeled vehicles in big cities, raises traffic problems, including toll roads. With a fast transaction process at the entrance of the toll road, it can reduce the queue of traffic congestion. For this reason, it is necessary to use a vehicle number character recognition image processing application. In this study, an application for license plate character recognition was made on vehicle images starting from the plate detection process using background subtraction. Next is the segmentation process of each character and feature extraction process using area-based features extraction. For the character recognition process, use the K-Nearest Neighbor method. The data used in this study were 50 vehicle images with each of them having 3 distances namely 50 cm, 75 cm and 100 cm. The results of this research trial showed the best level of accuracy for the detection of license plate locations is 94%, the best accuracy of segmentation per character is 71.79% and 82.59% is the best accuracy of character recognition.