A Novel Deep Learning Approach for Number Plate Detection

The number of people in the world now is estimated to be 1.38 billion. As the population of India grows, there is a chance that there will be twice as many automobiles on the road. Numerous methods have been put out for finding and identifying the license plates of automobiles using various technologies and procedures. The license plate detection and identification accomplished in the paper is done by OpenCV and Tensor flow for the detection of the license plate. The second set makes use of CNN (Convolutional Neural Network). First, the input image has been taken and pre-processed. Next, the number plate has been detected and contours are extracted. Finally, the characters have been recognized and the model has been trained using CNN for predicting the characters in the number plate. Comparing to other TensorFlow projects, this system provides the highest accuracy.