License Plate Recognition for High Security Registration Plates: A Review Siddhartha Choubey 1 , Anshuman Rai 2 1 Associate Professor, Computer Science and Engineering, Shree Shankaracharya Technical Campus, Bhilai, India 1 2 Pro-term Lecturer, Computer Science and Engineering, Shree Shankaracharya Technical Campus, Bhilai, India 2 Abstract— Traffic management and security issues have become all pervasive in today’s environment. In order to regulate traffic it is necessary to identify vehicles with their registration plates. Vehicle license plate recognition technology has helped in vehicle identification, theft prevention, speed regulation, and various other traffic management concerns. With the introduction of high security registration plates in India, various security measures will be improved. We present a survey of various techniques for vehicle license plate detection and recognition. Keywords— License Plate Recognition, License Plate Detection, High Security Registration Plate. I. INTRODUCTION License Plate Recognition (LPR) technology has been used to uniquely identify vehicles. Vehicle License Plate registration plates are assigned by the regional transport office to uniquely identify a vehicle after the vehicle has been registered with the regional office[1]. This helps in identifying the owner of the vehicle, insurance information and other. The new high security registration plate (HSRP) offers security features like a hot stamped chromium based hologram, ingressed IND legend, laser etched 9-digit code, snap lock, security inscript, embossed borders and alpha- numerals, and superior grade reflective sheeting. It also has a height, thickness and separation of characters. Fig1. shows the schematic of an HSRP. Fig 1. High Security Registration Plate. II. LICENSE PLATE RECOGNITION SYSTEM A License Plate Recognition System can be thought of as composed of three phases [2] 1. LP Detection: LP Detection is the detection of the license plate within the input image. Various Methods have been identified for license plate detection Binary Image Processing, Gray Level Processing, Color Processing and Classifiers [2]. 2. Character Segmentation: After license plate detection, individual characters are extracted. This phase extracts individual characters from the license plates. Various techniques like binary image processing, gray level processing, histogram processing, local adapative thesholding and transformation [2] have been developed. 3. Character Recognition: After characters have been segmented individual characters can be recognized by various techniques like classifiers, template matching [2]. 4. Fig 2. shows the schematic of a license plate recognition system. Fig2. License Plate Recognition Process. III.RELATED RESEARCH WORK A variety of research work has been done on license plate recognition system. Most can be classified as differing on the plate detection method, character segmentation methods and the character recognition methods. In [3] License Plates are localized by firstly pre-processing which includes rgb to gray,edge detection and noise reduction following which possibly license plates are identified by changes in contrast, width by height factor. Character segmentation is done by looking for similarities in characteristics of characters. Character recognition is done with the help of ANNs. In [4] the authors have the authors have presented a system that uses frequency domain filter, the Mexican hat operator filter and the spatial filter for image enhancement to improve results of an LPR system for high security registration plates. In [5] The authors have used the basic morphological operations of skeleton, end points and thinning of binary images along with a standard LPR system. For recognition Siddhartha Choubey et al, / (IJCSIT) International Journal of Computer Science and Information Technologies, Vol. 6 (2) , 2015, 1036-1039 www.ijcsit.com 1036