Licence Plate Recognition System Using Open-CV and Tesseract OCR Engine

Abstract

As the technology has taken a leap to make sure human lives get easier, it has also come with certain consequences. One of them being traffic control and vehicle owner identification has really become a serious issue in the 21st century. Due to the advancements in automobile technology, it is very easy for a person to violate traffic rules and it is practically not possible for humans to stop or have a track record of the vehicles’ number plate travelling at higher speeds. This is a major problem which is being faced by developing countries and our paper will discuss an implementable solution for this problem. Licence plate recognition (LPR) is an information processing system which performs an optical character recognition (OCR) on a digital image of the licence plate which consists of alpha-numeric characters. In this paper we put forward three staged licence plate recognition system based on open-cv and tesseract OCR engine which consists of licence plate detection, character segmentation and character recognition. The system generally uses infrared (IR) illumination to allow the camera to capture images at any point of the day. It also performs various functions such as capturing the image of the vehicle, storing the captured image along with the transcript of the licence plate. Open cv plays an important role in preparing images and videos to identify objects and tesseract OCR is used for text recognition in our prototype. The main purpose of this system is to design and develop an accurate image processing method along with successful recognition of the alphanumeric characters.