This document presents an automatic vehicle identification system using license plate detection. It proposes a new algorithm for license plate detection that uses edge detection and thresholding techniques. The algorithm has three main steps: license plate detection through edge extraction and thresholding, character segmentation, and character recognition. It was tested on images captured from a web camera and achieved an accuracy of 91.4% for real-time license plate detection, outperforming existing algorithms. The computation time of the new algorithm is 47.7 ms, which is 5-9 times faster than other existing methods.