Application of Machine Learning Methods in Traffic Classification and Recognition

A systematic overview of machine learning methods and their applications in traffic classification and recognition. Summarize the commonly used machine learning methods, mainly including clustering algorithms, support vector machines, deep learning and ensemble learning. Based on four different classification and recognition targets, namely vehicle type, license plate, traffic sign and driving behavior recognition, this paper reviews the application of machine learning methods in traffic classification and recognition, and points out machine learning methods and improvement strategies vary with different traffic application scenarios.