Efficient acoustic detector of gunshots and glass breaking

Lojka, Martin; Pleva, Matúš; Kiktová, Eva; Juhár, Jozef; Čižmár, Anton

doi:10.1007/s11042-015-2903-z

Cited by 24 publications

(13 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…sound events with abrupt changes in energy, such as door slams, claps, and recrackers [10][11][12][13][14][15]. Although it will take a very long time to capture data on a multitude of surroundings, the environments mentioned above are representative to expose the ML algorithm to a wide range of features.…”

Section: Environmentmentioning

confidence: 99%

“…The motivation of this work stems from the fact that datasets containing sounds closely related to gunshot sounds are not readily available for research purposes. Although a great amount of work has been presented in the detection of gunshot sounds [3][4][5][6][7][8][9][10][11][12][13][14][15][16], comparative analyses of similar audio events are not as detailed, though in the literature, researchers carried out comparative analyses of similar sounding events to gunshot audio events with abrupt changes in energy, such as door slams, claps, and firecrackers [10][11][12][13][14][15][16].…”

Section: Introductionmentioning

confidence: 99%

“…The work in [16] provided the comparison of impulsive sounds which includes the explosion of a balloon, clap and the sound of plosive consonants such a [p], [t], and [k]. The authors of [10][11][12][13][14][15][16] all discuss the detection of gunshot sounds in comparison to other sounds; however, they did not provide a database of the non-gunshot sounds readily available for further research.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Data Collection, Modeling, and Classification for Gunshot and Gunshot-like Audio Events: A Case Study

Singh

Zhuang

Pawani

2021

Sensors

View full text Add to dashboard Cite

Distinguishing between a dangerous audio event like a gun firing and other non-life-threatening events, such as a plastic bag bursting, can mean the difference between life and death and, therefore, the necessary and unnecessary deployment of public safety personnel. Sounds generated by plastic bag explosions are often confused with real gunshot sounds, by either humans or computer algorithms. As a case study, the research reported in this paper offers insight into sounds of plastic bag explosions and gunshots. An experimental study in this research reveals that a deep learning-based classification model trained with a popular urban sound dataset containing gunshot sounds cannot distinguish plastic bag pop sounds from gunshot sounds. This study further shows that the same deep learning model, if trained with a dataset containing plastic pop sounds, can effectively detect the non-life-threatening sounds. For this purpose, first, a collection of plastic bag-popping sounds was recorded in different environments with varying parameters, such as plastic bag size and distance from the recording microphones. The audio clips’ duration ranged from 400 ms to 600 ms. This collection of data was then used, together with a gunshot sound dataset, to train a classification model based on a convolutional neural network (CNN) to differentiate life-threatening gunshot events from non-life-threatening plastic bag explosion events. A comparison between two feature extraction methods, the Mel-frequency cepstral coefficients (MFCC) and Mel-spectrograms, was also done. Experimental studies conducted in this research show that once the plastic bag pop sounds are injected into model training, the CNN classification model performs well in distinguishing actual gunshot sounds from plastic bag sounds.

show abstract

Section: Environmentmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Data Collection, Modeling, and Classification for Gunshot and Gunshot-like Audio Events: A Case Study

Singh

Zhuang

Pawani

2021

Sensors

View full text Add to dashboard Cite

show abstract

“…In a recent publication, Martin et. al [68] proposed the a novel model called "Weighted Finite-State Transducers (WFSTs)" for the detection of acoustic events such as gunshot and glass-break in surveillance environments. The WFSTs model is in fact a combination of a set of Hidden Markov Models (HMMs), with each HMM for one type of events.…”

Section: Features Beyond Visionmentioning

confidence: 99%

Automatic Event Detection for Signal-based Surveillance

Xu,

Fookes,

Sridharan

2016

Preprint

View full text Add to dashboard Cite

Signal-based Surveillance systems such as Closed Circuits Televisions (CCTV) have been widely installed in public places. Those systems are normally used to find the events with security interest, and play a significant role in public safety. Though such systems are still heavily reliant on human labour to monitor the captured information, there have been a number of automatic techniques proposed to analysing the data. This article provides an overview of automatic surveillance event detection techniques . Despite it's popularity in research, it is still too challenging a problem to be realised in a real world deployment. The challenges come from not only the detection techniques such as signal processing and machine learning, but also the experimental design with factors such as data collection, evaluation protocols, and ground-truth annotation. Finally, this article propose that multi-disciplinary research is the path towards a solution to this problem.

show abstract

“…EAR-TUKE olarak akustik olay tespit sistemi sunulan çalışmada HMM ve MFCC, FBANK ve MELSPEC kullanılarak cam kırılması ve silah sesini %98.89 başarı ile ayırt etmişti. 46 silah sesi ve 13 cam kırılması sesi kullanılmıştır [19].…”

Section: Gi̇ri̇ş (Introduction)unclassified

Silah Sesleri Kullanılarak Ateşli Silahların Sınıflandırılmasında Akustik Parametrelerin Etkisi

Özseven¹,

Düğenci²

2018

Gazi Üniversitesi Fen Bilimleri Dergisi Part C: Tasarım Ve Teknoloji

View full text Add to dashboard Cite

Makale Bilgisi ÖzAkustik analiz özellikle malzeme bilimi ve insan seslerinde kullanılmakta olup son yıllarda savunma sanayiinde ses perdesi, silah ve nişancı tespiti, insansız hava araçları gibi bir çok alanda kullanılmaktadır. Ateşli silahlarla ilgili eylemler hem güvenlik güçleri hem de halk için artan bir endişe kaynağıdır. Silah seslerinin sınıflandırılması için ticari veya deneysel çeşitli çalışmalar mevcuttur. Bu çalışmanın amacı silah seslerini kullanarak ateşli silahların türünü tespit etmede akustik analizin etkilerinin incelenmesidir. Çalışmada 23 ateşli silaha ait 510 atış ses kaydı kullanılmıştır. Akustik analiz için formant frekansları, MFCC, LPCC ve enerji parametreleri incelenmiştir. Akustik parametrelerin silahları sınıflandırmadaki etkinliği istatistiksel olarak analiz edilmiş ve kullanılan tüm parametrelerin etkili olduğu görülmüştür. MLP sınıflandırıcı ile sınıflandırma performansı test edilmiş ve ateşli silah tanıma oranı %71.56 elde edilmiştir. Tanıma oranı en yüksek silah türü "Carl Gustav M45" ve "Tokarev PPSh", tanıma oranı en düşük olan silah "Tikka Model T2" olarak tespit edilmiştir. The Impact of Acoustic Parameters on the Classification of Firearms Using Gunshots AbstractAcoustic analysis is mainly used in material science and human voice. In recent years, many areas such as sound perception, gun and shooter detection, unmanned aerial vehicles have been used in the defense industry. Actions related to firearms are an increasing source of concern for both security forces and the public. There are various commercial or experimental studies for the classification of gunshots. The purpose of this study is to examine the effects of acoustic analysis on the detection of firearms by using gunshots. In the study, 511 gunshots recordings of 23 firearms were used. The formant frequencies, MFCC, LPCC and energy parameters for acoustic analysis are examined. The effectiveness of the acoustic parameters in classifying the weapons was analyzed statistically and all the parameters used were found to be effective. The classification performance was tested with the MLP classifier and the firearm recognition rate was 71.56%. The "Carl Gustav M45-b" and "Tokarev PPSh" was identified as the weapon with the highest recognition rate and the lowest "Tikka Model T3" recognition rate.

show abstract

Efficient acoustic detector of gunshots and glass breaking

Cited by 24 publications

References 17 publications

Data Collection, Modeling, and Classification for Gunshot and Gunshot-like Audio Events: A Case Study

Data Collection, Modeling, and Classification for Gunshot and Gunshot-like Audio Events: A Case Study

Automatic Event Detection for Signal-based Surveillance

Silah Sesleri Kullanılarak Ateşli Silahların Sınıflandırılmasında Akustik Parametrelerin Etkisi

Contact Info

Product

Resources

About