Shooting sound analysis using convolutional neural networks and long short-term memory

Se Hyeok Kang; Ji Woong Cho

doi:10.7776/ASK.2022.41.3.312

All Issue

2022 Vol.41, Issue 3 Preview Page Next Page

Research Article

Shooting sound analysis using convolutional neural networks and long short-term memory 합성곱 신경망과 장단기 메모리를 이용한 사격음 분석 기법

31 May 2022. pp. 312-318

PDF XML

Abstract

This paper proposes a model which classifies the type of guns and information about sound source location using deep neural network. The proposed classification model is composed of convolutional neural networks (CNN) and long short-term memory (LSTM). For training and test the model, we use the Gunshot Audio Forensic Dataset generated by the project supported by the National Institute of Justice (NIJ). The acoustic signals are transformed to Mel-Spectrogram and they are provided as learning and test data for the proposed model. The model is compared with the control model consisting of convolutional neural networks only. The proposed model shows high accuracy more than 90 %.

Keywords

Gunshot classification

Convolutional Neural Network (CNN)

Long Short-term Memory (LSTM)

Audio feature extraction

Audio preprocessing

본 논문은 딥러닝기법 중 하나인 합성곱 신경망과 순환 신경망 중 하나인 장단기 메모리를 이용하여 사격시 발생하는 소음(이하 사격음)만으로 화기의 종류, 사격음 발생지점에 관한 정보(거리와 방향)을 추정하는 모델을 다루었다. 이를 위해 미국 법무부 산하 연구소의 지원하에 생성된 Gunshot Audio Forensic Dataset을 이용하였으며, 음향 신호를 멜 스펙트로그램(Mel-Spectrogram)으로 변환한 후, 4종의 합성곱 신경망과 1종의 장단기 메모리 레이어로 구성된 딥러닝 모델에 학습 및 검증 데이터로 제공하였다. 제안 모델의 성능을 확인하기 위해 합성곱 신경망으로만 구성된 대조 모델과 비교·분석하였으며, 제안 모델의 정확도가 90 % 이상으로 대조모델보다 우수한 성능을 보였다.

키워드

사격음 분류

합성곱 신경망

장단기 메모리

오디오 특징 추출

오디오 전처리

References

W. Kim, K. Shin, S. Cho, S. Baek, and Y. Kim, "Development of korean warrior platform architecture" (in Korean), J. of Convergence for Information Technology, 11, 111-117 (2021).

S. Khan, A. Divakaran, and H. S. Sawhney, "Weapon identification using hierarchical classification of acoustic signatures," Proc. Sensors, and Command, Control, Communications, and Intelligence (C3I) Technologies for Homeland Security and Homeland Defense VIII, 730510 (2009). 10.1117/12.818375

M. Djeddou and T. Touhami, "Classification and modeling of acoustic gunshot signatures," AJSE, 38, 3399- 3406 (2013). 10.1007/s13369-013-0655-5

K. D. Morton Jr, P. A. Torrione, and L. Collins, "Classification of acoustic gunshot signatures using a nonparametric bayesian signal model," Proc. Sensors, and Command, Control, Communications, and Intelligence (C3I) Technologies for Homeland Security and Homeland Defense X, 80190T (2011). 10.1117/12.885114

E. Kiktova, M. Lojka, M. Pleva, J. Juhar, and A. Cizmar, "Gun type recognition from gunshot audio recordings," Proc. IEEE 3rd IWBF, 1-6 (2015). 10.1109/IWBF.2015.7110240

S. Raponi, I. Ali, and G. Oligeri, "Sound of guns: digital forensics of gun audio samples meets artificial intelligence," arXiv preprint arXiv:2004.07948 (2020).

B. McFee, C. Raffel, D. Liang, D. P. W. Ellis, M. McVicar, E. Battenberg, and O. Nieto, "librosa: Audio and music signal analysis in python," Proc. 14th Python in Science Conf. 18-24 (2015). 10.25080/Majora-7b98e3ed-003

Information

Publisher :The Acoustical Society of Korea
Publisher(Ko) :한국음향학회
Journal Title :The Journal of the Acoustical Society of Korea
Journal Title(Ko) :한국음향학회지
Volume : 41
No :3
Pages :312-318
Received Date : 2022-02-16
Accepted Date : 2022-03-28
DOI :https://doi.org/10.7776/ASK.2022.41.3.312

The Journal of the Acoustical Society of KoreaISSN:1225-4428(Print) 2287-3775(Online)한국음향학회

All Issue