Hybrid Neural Network based on Feature Fusion for Vehicle Type Identification

Due to the audio information of different types of vehicle models are distinct, vehicle information can be identified by the audio signal of vehicle accurately. In real life, in order to determine the type of vehicle, we do not need to obtain the visual information of vehicles and just need to obtain the audio information. In this paper, we extract the Mel frequency cepstrum coefficients in perceptual characteristics, pitch class profile in psychoacoustic characteristics and short-term energy in acoustic characteristics. In addition, we improve the performance of neural network classifier by fusing long short-term memory (LSTM) unit into the convolutional neural networks. At last, we put the fusion feature to the hybrid neural networks to recognize different vehicles. The results suggest that the fusion feature we proposed in this paper can increase the recognition rate by 7%; and LSTM has great advantages in modeling time series, adding LSTM to the networks can improve the recognition rate of 3.39%.