Gun identification from gunshot audios for secure public places using transformer learning
Abstract Increased mass shootings and terrorist activities severely impact society mentally and physically. Development of real-time and cost-effective automated weapon detection systems increases a sense of safety in public. Most of the previously proposed methods were vision-based. They visually a...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2022-08-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-022-17497-1 |
| _version_ | 1811315451676852224 |
|---|---|
| author | Rahul Nijhawan Sharik Ali Ansari Sunil Kumar Fawaz Alassery Sayed M. El-kenawy |
| author_facet | Rahul Nijhawan Sharik Ali Ansari Sunil Kumar Fawaz Alassery Sayed M. El-kenawy |
| author_sort | Rahul Nijhawan |
| collection | DOAJ |
| description | Abstract Increased mass shootings and terrorist activities severely impact society mentally and physically. Development of real-time and cost-effective automated weapon detection systems increases a sense of safety in public. Most of the previously proposed methods were vision-based. They visually analyze the presence of a gun in a camera frame. This research focuses on gun-type (rifle, handgun, none) detection based on the audio of its shot. Mel-frequency-based audio features have been used. We compared both convolution-based and fully self-attention-based (transformers) architectures. We found transformer architecture generalizes better on audio features. Experimental results using the proposed transformer methodology on audio clips of gunshots show classification accuracy of 93.87%, with training loss and validation loss of 0.2509 and 0.1991, respectively. Based on experiments, we are convinced that our model can effectively be used as both a standalone system and in association with visual gun-detection systems for better security. |
| first_indexed | 2024-04-13T11:30:28Z |
| format | Article |
| id | doaj.art-180aa21cab3e4e8eac82999fbb7337c9 |
| institution | Directory Open Access Journal |
| issn | 2045-2322 |
| language | English |
| last_indexed | 2024-04-13T11:30:28Z |
| publishDate | 2022-08-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj.art-180aa21cab3e4e8eac82999fbb7337c92022-12-22T02:48:35ZengNature PortfolioScientific Reports2045-23222022-08-0112111010.1038/s41598-022-17497-1Gun identification from gunshot audios for secure public places using transformer learningRahul Nijhawan0Sharik Ali Ansari1Sunil Kumar2Fawaz Alassery3Sayed M. El-kenawy4School of Computer Science, University of Petroleum and Energy StudiesCalifornia State University Dominguez HillsSchool of Computer Science, University of Petroleum and Energy StudiesDepartment of Computer Engineering, College of Computers and Information Technology, Taif UniversityDepartment of Communications and Electronics, Delta Higher Institute of Engineering and TechnologyAbstract Increased mass shootings and terrorist activities severely impact society mentally and physically. Development of real-time and cost-effective automated weapon detection systems increases a sense of safety in public. Most of the previously proposed methods were vision-based. They visually analyze the presence of a gun in a camera frame. This research focuses on gun-type (rifle, handgun, none) detection based on the audio of its shot. Mel-frequency-based audio features have been used. We compared both convolution-based and fully self-attention-based (transformers) architectures. We found transformer architecture generalizes better on audio features. Experimental results using the proposed transformer methodology on audio clips of gunshots show classification accuracy of 93.87%, with training loss and validation loss of 0.2509 and 0.1991, respectively. Based on experiments, we are convinced that our model can effectively be used as both a standalone system and in association with visual gun-detection systems for better security.https://doi.org/10.1038/s41598-022-17497-1 |
| spellingShingle | Rahul Nijhawan Sharik Ali Ansari Sunil Kumar Fawaz Alassery Sayed M. El-kenawy Gun identification from gunshot audios for secure public places using transformer learning Scientific Reports |
| title | Gun identification from gunshot audios for secure public places using transformer learning |
| title_full | Gun identification from gunshot audios for secure public places using transformer learning |
| title_fullStr | Gun identification from gunshot audios for secure public places using transformer learning |
| title_full_unstemmed | Gun identification from gunshot audios for secure public places using transformer learning |
| title_short | Gun identification from gunshot audios for secure public places using transformer learning |
| title_sort | gun identification from gunshot audios for secure public places using transformer learning |
| url | https://doi.org/10.1038/s41598-022-17497-1 |
| work_keys_str_mv | AT rahulnijhawan gunidentificationfromgunshotaudiosforsecurepublicplacesusingtransformerlearning AT sharikaliansari gunidentificationfromgunshotaudiosforsecurepublicplacesusingtransformerlearning AT sunilkumar gunidentificationfromgunshotaudiosforsecurepublicplacesusingtransformerlearning AT fawazalassery gunidentificationfromgunshotaudiosforsecurepublicplacesusingtransformerlearning AT sayedmelkenawy gunidentificationfromgunshotaudiosforsecurepublicplacesusingtransformerlearning |