Support Vector Machine Technique for License Plate Recognition System

Abstract

Artificial intelligence play very important role in our life these day’s.  Machine learning is subset and prime part of AI and further deep learning is subset of ML. In machine learning there are various types of algorithm exist which help us to train our electronic devices to execute various task as per requirement. These algorithms are divided into three categories supervised, unsupervised and reinforced learning.  In our research work support vector machine technique is used which is a supervised type learning algorithm. SVM technique has good performance and has high level accuracy of classification. In SVM technique algorithm initiates from a cluster of samples of numbers from number plates. Every character is identified by support vector machine and well in advance SVM technique is edified by known samples. In order to recognize a number plate correctly, all numbers are tested one by one using the trained model. Our experimental results impart that implemented technique worked with high level accuracy and average accuracy of SVM technique is found above 98 %. Besides this computational time is also less.