Neural network classification for detecting abnormal events in a public transport vehicle

Abstract

A method to detect an abnormal situation inside a public transport bus using audio signals is presented. Mel Frequency Cepstral Coefficients (MFCC) were used as a feature vector and a multilayer backpropagation neural network as a classifier. Audio samples were taken inside the bus running along Epifanio Delos Santos Avenue (EDSA), Metro Manila, Philippines. The audio samples depict sounds under normal operation inside the bus. The abnormal situation was represented by superimposing the sound of normal operation and the sounds of gunshots, crowd in panic and screams for signal to noise ratio of 10, 20, 30, and 40dB. The sounds were divided into 3-second audio clips. The audio clips were divided into frames and each 3-second audio clip produced 594 frames. Each frame is represented by 12 MFCCs. The accuracy of the system was tested for all frames and all 3-second audio clips. The accuracy of the system when measurement was based on the number of frames that were correctly classified divided by the total number of frames tested is 99.41 %. When the measurement was based on 3-second audio clips, the proposed system correctly classified all the events in 20, 30 and 40 dB signal-to-noise ratios. Errors occurred in the classification of abnormal events at 10 dB signal-to-noise ratio the classification of normal events. The accuracy is 97% and 93%, respectively.