ntvНаучно-технический вестник информационных технологий, механики и оптикиScientific and Technical Journal of Information Technologies, Mechanics and Optics2226-14942500-0373Университет ИТМО10.17586/2226-1494-2024-24-1-51-61ntv-125Research ArticleКОМПЬЮТЕРНЫЕ СИСТЕМЫ И ИНФОРМАЦИОННЫЕ ТЕХНОЛОГИИCOMPUTER SCIENCEУлучшенная производительность модели RetinaNet для обнаружения огнестрельного оружия в пользовательском наборе данных и видеонаблюдения в реальном времениAn improved performance of RetinaNet model for hand-gun detection in custom dataset and real time surveillance videohttps://orcid.org/0009-0002-0512-6414КхинП. П.KhinP. P.

Кхин Пьоне Пьоне — магистр, аспирант, преподаватель

 Мандалай, 05072

 

Pyone Pyone Khin — Master, PhD Student, Lecturer

Mandalay, 05072

pyonekhin.ppk@gmail.comhttps://orcid.org/0009-0009-8295-6914ХтайкН. М.HtaikN. M.

Хтайк Най Мин — PhD, профессор

 Мандалай, 0507

Nay Min Htaik — PhD, Professor

Mandalay, 05072

nayminhtaik@gmail.comМандалайский технологический университетМьянмаMandalay Technological UniversityMyanmar2024131220242415161Copyright © Кхин П.П., Хтайк Н.М., 20242024Кхин П.П., Хтайк Н.М.Khin P.P., Htaik N.M.Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://ntv.elpub.ru/jour/article/view/125

Распространенность вооруженных ограблений стала серьезной проблемой в современном мире, что требует разработки эффективных систем обнаружения. Существующие разнообразные устройства обнаружения не обладают способностью автоматически выявлять и предупреждать о наличии оружия во время осуществления вооруженных ограблений. Для решения этой проблемы предлагается подход, основанный на глубоком обучении, с использованием модели RetinaNet. В результате его применения возможно точное обнаружение оружия и дальнейшее предупреждение об ограблении полицейского участка или владельца банка. Ядро модели RetinaNet состоит из трех основных компонентов: остаточной сети (Residual Neural Network, ResNet), функциональной пирамидальной сети (Feature Pyramid Net, FPN) и полностью сверточной сети (Fully Convolutional Networks, FCN). Эти компоненты работают вместе, обеспечивая обнаружение оружия в режиме реального времени без вмешательства человека. Предлагаемая реализация использует специальный набор данных для обнаружения грабежей, который состоит из классов активности с применением огнестрельного оружия, без оружия и грабежей. Оценка производительности предлагаемой модели на разработанном специальном наборе данных показал, что магистральная архитектура ResNet50 превосходит точность обнаружения ограблений, достигая меры оценки качества ранжирования (Mean Average Precision, mAP) 0,92. Эффективность модели заключается в ее способности точно определять наличие оружия во время ограбления.

The prevalence of armed robberies has become a significant concern in today’s world, necessitating the development of effective detection systems. While various detection devices exist in the market, they do not possess the capability to automatically detect and alarm the presence of guns during robbery activities. In order to address this issue, a deep learning-based approach using gun detection using RetinaNet model is proposed. The objective is to accurately detect guns and subsequently alert either the police station or the bank owner. RetinaNet, the core of the system, comprises three main components: the Residual Neural Network (ResNet), the Feature Pyramid Network (FPN), and the Fully Convolutional Networks (FCN). These components work together to enable real-time detection of guns without the need for human intervention. Proposed implementation uses a custom robbery detection dataset that consists of gun, no-gun and robbery activity classes. By evaluating the performance of the proposed model on our custom dataset, it is evident that the ResNet50 backbone architecture yields outperforms for the accuracy in robbery detection that reached in 0.92 of Mean Average Precision (mAP). The model effectiveness lies in its ability to accurately identify the presence of guns during robbery activities.

вооруженное ограблениеRetinaNetResNet50FPNFCNrobbery activitiesRetinaNetResNet50FPNFCNReferencesDever J., da Vitoria Lobo N., Shah M. Automatic visual recognition of armed robbery. Proc. of the 16th International Conference on Pattern Recognition. V. 1, 2002, pp. 451–455. https://doi.org/10.1109/ICPR.2002.1044755Dever J., da Vitoria Lobo N., Shah M. Automatic visual recognition of armed robbery. Proc. of the 16th International Conference on Pattern Recognition. V. 1, 2002, pp. 451–455. https://doi.org/10.1109/ICPR.2002.1044755Ahmed S., Bhatti M.T., Khan M.G., Lövström B., Shahid M. Development and optimization of deep learning models for weapon detection in surveillance videos. Applied Sciences, 2022, vol. 12, no. 12, pp. 5772. https://doi.org/10.3390/app12125772Ahmed S., Bhatti M.T., Khan M.G., Lövström B., Shahid M. Development and optimization of deep learning models for weapon detection in surveillance videos. Applied Sciences, 2022, vol. 12, no. 12, pp. 5772. https://doi.org/10.3390/app12125772Kakadiya R., Lemos R., Mangalan S., Pillai M. AI based automatic robbery/theft detection using smart surveillance in banks. Proc. of the 3rd International Conference on Electronics, Communication and Aerospace Technology (ICECA), 2019, pp. 201–204. https://doi.org/10.1109/iceca.2019.8822186Kakadiya R., Lemos R., Mangalan S., Pillai M. AI based automatic robbery/theft detection using smart surveillance in banks. Proc. of the 3rd International Conference on Electronics, Communication and Aerospace Technology (ICECA), 2019, pp. 201–204. https://doi.org/10.1109/iceca.2019.8822186Narejo S., Pandey B., Vargas D.E., Rodriguez C., Anjum M.R. Weapon detection using YOLO V3 for smart surveillance system. Mathematical Problems in Engineering, 2021, vol. 2021, pp. 9975700. https://doi.org/10.1155/2021/9975700Narejo S., Pandey B., Vargas D.E., Rodriguez C., Anjum M.R. Weapon detection using YOLO V3 for smart surveillance system. Mathematical Problems in Engineering, 2021, vol. 2021, pp. 9975700. https://doi.org/10.1155/2021/9975700Salido J., Lomas V., Ruiz-Santaquiteria J., Deniz O. Automatic handgun detection with deep learning in video surveillance images. Applied Sciences, 2021, vol. 11, no. 13, pp. 6085. https://doi.org/10.3390/app11136085Salido J., Lomas V., Ruiz-Santaquiteria J., Deniz O. Automatic handgun detection with deep learning in video surveillance images. Applied Sciences, 2021, vol. 11, no. 13, pp. 6085. https://doi.org/10.3390/app11136085Zahrawi M., Shaalan K. Improving video surveillance systems in banks using deep learning techniques. Scientific Reports, 2023, vol. 13, pp. 7911. https://doi.org/10.1038/s41598-023-35190-9Zahrawi M., Shaalan K. Improving video surveillance systems in banks using deep learning techniques. Scientific Reports, 2023, vol. 13, pp. 7911. https://doi.org/10.1038/s41598-023-35190-9Ineneji C., Kusaf M. Hybrid weapon detection algorithm, using material test and fuzzy logic system. Computers & Electrical Engineering, 2019, vol. 78, pp. 437–448. https://doi.org/10.1016/j.compeleceng.2019.08.005Ineneji C., Kusaf M. Hybrid weapon detection algorithm, using material test and fuzzy logic system. Computers & Electrical Engineering, 2019, vol. 78, pp. 437–448. https://doi.org/10.1016/j.compeleceng.2019.08.005Warsi A., Abdullah M., Husen M.N., Yahya M., Jawaid N. Gun detection system using YOLOv3. Proc. of the IEEE International Conference on Smart Instrumentation, Measurement and Application (ICSIMA), 2019, pp. 1–4. https://doi.org/10.1109/ICSIMA47653.2019.9057329Warsi A., Abdullah M., Husen M.N., Yahya M., Jawaid N. Gun detection system using YOLOv3. Proc. of the IEEE International Conference on Smart Instrumentation, Measurement and Application (ICSIMA), 2019, pp. 1–4. https://doi.org/10.1109/ICSIMA47653.2019.9057329Bhatti M.T., Khan M.G., Aslam M., Fiaz M.J. Weapon detection in real-time CCTV videos using deep learning. IEEE Access, 2021, vol. 9, pp. 34366–4382. https://doi.org/10.1109/ACCESS.2021.3059170Bhatti M.T., Khan M.G., Aslam M., Fiaz M.J. Weapon detection in real-time CCTV videos using deep learning. IEEE Access, 2021, vol. 9, pp. 34366–4382. https://doi.org/10.1109/ACCESS.2021.3059170Hashmi T.S.S., Haq N.U., Fraz M.M., Shahzad M. Application of deep learning for weapons detection in surveillance videos. Proc. of the 2021 International Conference on Digital Futures and Transformative Technologies (ICoDT2), 2021. https://doi.org/10.1109/ICoDT252288.2021.9441523Hashmi T.S.S., Haq N.U., Fraz M.M., Shahzad M. Application of deep learning for weapons detection in surveillance videos. Proc. of the 2021 International Conference on Digital Futures and Transformative Technologies (ICoDT2), 2021. https://doi.org/10.1109/ICoDT252288.2021.9441523Alaqil R.M., Alsuhaibani J.A., Alhumaidi B.A., Alnasser R.A., Alotaibi R.D., Benhidour H. Automatic gun detection from images using Faster R-CNN. Proc. of the IEEE International Conference of Smart Systems and Emerging Technologies (SMARTTECH), 2020, pp. 149–154. https://doi.org/10.1109/SMARTTECH49988.2020.00045Alaqil R.M., Alsuhaibani J.A., Alhumaidi B.A., Alnasser R.A., Alotaibi R.D., Benhidour H. Automatic gun detection from images using Faster R-CNN. Proc. of the IEEE International Conference of Smart Systems and Emerging Technologies (SMARTTECH), 2020, pp. 149–154. https://doi.org/10.1109/SMARTTECH49988.2020.00045Yang M., Xiao X., Liu Z., Sun L., Guo W., Cui L., Sun D., Zhang P., Yang G. Deep RetinaNet for dynamic left ventricle detection in multiview echocardiography classification. Scientific Programming, 2020, vol. 2020, pp. 7025403. https://doi.org/10.1155/2020/7025403Yang M., Xiao X., Liu Z., Sun L., Guo W., Cui L., Sun D., Zhang P., Yang G. Deep RetinaNet for dynamic left ventricle detection in multiview echocardiography classification. Scientific Programming, 2020, vol. 2020, pp. 7025403. https://doi.org/10.1155/2020/7025403Bhabad D., Kadam S., Malode T., Shinde G., Bage D. Object detection for night vision using deep learning algorithms. International Journal of Computer Trends and Technology, 2023, vol. 71, no. 2, pp. 87–92. https://doi.org/10.14445/22312803/ijcttv71i2p113Bhabad D., Kadam S., Malode T., Shinde G., Bage D. Object detection for night vision using deep learning algorithms. International Journal of Computer Trends and Technology, 2023, vol. 71, no. 2, pp. 87–92. https://doi.org/10.14445/22312803/ijcttv71i2p113Rani E.E., Baulkani S. Construction of deep learning model using ResNet 50 for schizophrenia prediction from rsFMRI images. Research Square, 2022. https://doi.org/10.21203/rs.3.rs-2106170/v1Rani E.E., Baulkani S. Construction of deep learning model using ResNet 50 for schizophrenia prediction from rsFMRI images. Research Square, 2022. https://doi.org/10.21203/rs.3.rs-2106170/v1Guan Q., Wang Y., Ping B., Li D., Du J., Qin Y., Lu H., Wan X., Xiang J. Deep convolutional neural network VGG-16 model for differential diagnosing of papillary thyroid carcinomas in cytological images: A pilot study. Journal of Cancer, 2019, vol. 10, no. 20, pp. 4876–4882. https://doi.org/10.7150/jca.28769Guan Q., Wang Y., Ping B., Li D., Du J., Qin Y., Lu H., Wan X., Xiang J. Deep convolutional neural network VGG-16 model for differential diagnosing of papillary thyroid carcinomas in cytological images: A pilot study. Journal of Cancer, 2019, vol. 10, no. 20, pp. 4876–4882. https://doi.org/10.7150/jca.28769

The authors declare that there are no conflicts of interest present.