Multi-class Classification of Impulse and Non-impulse Sounds Using Deep Convolutional Neural Network (DCNN)

Differentiating between military sounds can be quite tasking with high false detection rate. These sounds can either be impulse sounds (sounds released from the military weapons) or non-impulse sounds (sound released from other sources) thus causing public disturbance and unnecessary panic. This paper utilizes Deep Convolutional Neural Network (DCNN) classifier to detect military impulse and non-impulse sounds and also incorporates Adam algorithm for optimal classification. DCNN was utilized in this study based on its network embedded multiple hidden layers (non-linear) which can learn the very complicated relationship between the input data and require output. The dataset used in this study consist of six sound types with a total number of 37,464 datasets which was partitioned into training (67%) and testing (33%). The performance of the proposed classifier was evaluated based on the following metrics: True Positive (TP), True Negative (TN), False Positive (FP), False Negative (FN), Precision, Matthews Correlation Coefficient (MCC), and Accuracy. The experimental result shows that DCNN classifier gave an optimal accuracy for the Machine gun, Wind, Thunder, Blast, Vehicle, and Aircraft sounds types as 97.43%, 96.98%, 95.16%, 95.13%, 88.83%, and 87% respectively. The average classification error rate for the six sound types was 6.57% which signifies that DCNN is a promising classifier.

[1]  P D Schomer,et al.  Community Reaction to Impulsive Noise. A Final 10-Year Research Summary. Revised. , 1985 .

[2]  Tuomo Kauranne,et al.  Aircraft classification based on radar cross section of long-range trajectories , 2015, IEEE Transactions on Aerospace and Electronic Systems.

[3]  Karol J. Piczak.  Environmental sound classification with convolutional neural networks , 2015, 2015 IEEE 25th International Workshop on Machine Learning for Signal Processing (MLSP).

[4]  Yi-Hsuan Yang,et al.  Machine Recognition of Music Emotion: A Review , 2012, TIST.

[5]  Sumit Das,et al.  Applications of Artificial Intelligence in Machine Learning: Review and Prospect , 2015 .

[6]  Martin Lojka,et al.  Efficient acoustic detector of gunshots and glass breaking , 2015, Multimedia Tools and Applications.

[7]  Matthew Brandon Rhudy,et al.  Real Time Implementation of a Military Impulse Classifier , 2010 .

[8]  Karol J. Piczak.  ESC: Dataset for Environmental Sound Classification , 2015, ACM Multimedia.

[9]  Bhiksha Raj,et al.  Deep CNN Framework for Audio Event Recognition using Weakly Labeled Web Data , 2017, ArXiv.

[10]  Chenn-Jung Huang,et al.  Frog classification using machine learning techniques , 2009, Expert Syst. Appl..

[11]  Luis Pastor Sánchez Fernández,et al.  Aircraft Classification and Acoustic Impact Estimation Based on Real-Time Take-off Noise Measurements , 2013, Neural Processing Letters.

[12]  Shashidhar G. Koolagudi,et al.  Sound event detection in urban soundscape using two-level classification , 2016, 2016 IEEE Distributed Computing, VLSI, Electrical Circuits and Robotics (DISCOVER).

[13]  Justin Salamon,et al.  Unsupervised feature learning for urban sound classification , 2015, 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[14]  Colin Raffel,et al.  Lasagne: First release. , 2015 .

[15]  Nikos Fakotakis,et al.  Automatic Recognition of Urban Soundscenes , 2008, New Directions in Intelligent Interactive Multimedia.

[16]  José,et al.  Gunshot detection in noisy environments , 2010 .

[17]  Brian A. Bucci,et al.  Performance of artificial neural network-based classifiers to identify military impulse noise. , 2007, The Journal of the Acoustical Society of America.

[18]  Jeffrey S. Vipperman,et al.  Development of artificial neural network classifier to identify military impulse noise , 2006 .

[19]  Nikos Fakotakis,et al.  Automatic Recognition of Urban Environmental Sound Events , 2008 .

[20]  Jeffrey S. Vipperman,et al.  SIX NOISE TYPE MILITARY SOUND CLASSIFIER , 2012 .

[21]  Amr Ahmed,et al.  Military aircrafts’ classification based on their sound signature , 2016 .

[22]  Ann Nakashima,et al.  Review of weapon noise measurement and damage risk criteria: considerations for auditory protection and performance. , 2015, Military medicine.