“Feature Extraction in Vehicle ID & Classification using Deep Learning & Image Processing”

Abstract

Abstract- Intelligent transportation systems have acknowledged the ration of attention in the last decades. In this area vehicle classification and localization is the important task. In this task the important challenge is to discriminate the features of different vehicles. Further, vehicle classification and detection is a bigger problem to identify and locate because different variety of vehicles don’t follow the lane discipline. In this project, to identify and locate, we have created a convolution neural network from scratch to classify and detect objects using a modern convolution neural network based on fast regions. In this project we have considered three types of vehicles like bus, car and bike for classification and detection. Our approach is to use the entire image as input and create a bounding box with a probability estimations of the feature classes as  a output. The results of the project have shown that the projected system can considerably improved the accuracy of the detection.