ENGENALAN KARAKTER PLAT KENDARAAN BERMOTOR MENGGUNAKAN PROYEKSI PROFILE

Abstract

Recognizing vehicle’s license plate is required in a security system, parking lot management, tracking of lost vehicles and traffic monitoring. Therefore, there arises the need for characters recognition system of vehicle number plates which can facilitate controlling the security system so as to minimize the occurrence of cases of missing vehicles. Theaim of this research is to recognize the characters on the motorcycle license plate by using distance measurement method. The image was captured using a digital camera. Having obtained the image of the vehicle license plate number, then the image is processed through several processing such as image quality improvement to obtain a license plate image which indicates a motorcycle license plate. The results of license plate images are then segmented to separate each character. Afterwards, the characters which have been segmented is extracted to obtain the characteristic of each character using the profile projection and invariant moment. Then the character recognition is done using city block distance measurement method. The measurement of the value of the minimum distance is used as matching results by displaying the character recognition results. This research trial used 120 image data. License plate characters recognition can be recognized properly with an accuracy rate of 55%.