Morphology Based Approach To Recognize Number Plates in India

Abstract

Automatic number plate recognition is a real-time embedded system which automatically recognizes the license number of vehicles. Such systems require the localization of number plate area in order to identify the characters present on it. This paper presents an approach based on simple but efficient morphological opening and closing operations for localization of Indian number plates. After localization the skew correction of the number plate is also done for effective segmentation of characters. The number plate skew correction greatly affects the accuracy of the character extraction. The character extraction is based on template matching approach. Our proposed algorithm has been tested on 100 samples and is found to be robust to detect vehicle license plates as well as extracting the numbers with an overall accuracy of 90%