License plate character recognition is an integral part in Intelligent Transportation Systems. Traditional methods used for number plate recognition were OCR, Optical Character Recognition and Formula Based Recognition. As Neural Network is an intelligence engine, it ensures greater accuracy rate along with better recognition speed. In this system, recognition is done with the help of Neural Network. License plate is automatically located using principal visual word (PVW), discovery and local feature matching. Given a new image, the license plate is extracted by matching local features with PVW. Then image preprocessing operations are performed for improving the quality of plate images. This license plate image is then passed for Character Segmentation and these segmented characters are recognized using Neural Network.