![REFERENCES
[1] D. Trottier, ‘‘Crowdsourcing CCTV surveillance on the internet,’’ Inf.,
Commun. Soc., vol. 17, no. 5, pp. 609–626, May 2014.
[2] S. Ren, K.He,R.Girshick, and J. Sun, ‘‘Faster R-CNN: Towards real-time
object detection with region proposal networks,’’ in Proc. Adv. Neural Inf.
Process. Syst., 2015, pp. 91–99.
[3] S. Targ, D. Almeida, and K. Lyman, ‘‘ResNet in ResNet: Generalizing
residual architectures,’’ 2016, arXiv:1603.08029.
[4] K. He, X. Zhang, S. Ren, and J. Sun, ‘‘Deep residual learning for image
recognition,’’ in Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR),
Jun. 2016, pp. 770–778.
[5] Z. Wu, C. Shen, and A. Van Den Hengel, ‘‘Wider or deeper: Revisiting the
ResNet model for visual recognition,’’ Pattern Recognit., vol. 90, pp. 119–
133, Jun. 2019.](https://image.slidesharecdn.com/batch13-240320033859-5324a46c/85/reducing-firearms-based-on-deep-learning-13-320.jpg)
![[6] M. Tan, R. Pang, and Q. V. Le, ‘‘EfficientDet: Scalable and efficient
object detection,’’ in Proc. IEEE/CVFConf.
Comput.Vis.PatternRecognit. (CVPR), Jun. 2020, pp. 10781–10790.
[7] M. Tan and Q. V. Le, ‘‘EfficientNet : Rethinking model scaling for
convolutional neural networks,’’ 2019, arXiv:1905.11946.
[8] A.Egiazarov, V. Mavroeidis, F. M. Zennaro, and K. Vishi , ‘‘Firearm
detection and segmentation using an ensemble of semantic neural
networks,’’ in Proc. Eur. Intell. Secur. Informat. Conf. (EISIC), Nov.
2019, pp. 70–77.
[9] S. Akçay, M. E. Kundegorski, M. Devereux, and T. P. Breckon,
‘‘Transfer learning using convolutional neural networks for object
classification within X-ray baggage security imagery,’’ in Proc. IEEE
Int. Conf. Image Process. (ICIP), Sep. 2016, pp. 1057–1061.
[10] R. Olmos, S. Tabik, and F. Herrera, ‘‘Automatic handgun detection
alarm in videos using deep learning,’’ Neurocomputing, vol. 275, pp.
66–72, Jan. 2018.](https://image.slidesharecdn.com/batch13-240320033859-5324a46c/85/reducing-firearms-based-on-deep-learning-14-320.jpg)
reducing firearms based on deep learning
- 1. Department of Computer Science & Engineering B. Tech IV-II project A Deep Learning-Based Efficient Firearms Monitoring Technique for Building Secure Smart Cities Student Names Roll Number Guide K. Silpa 20W51A0535 Mrs. Y. JYOTHSNA M. Tech M. Mounika 20W51A0540 K. Mohammad Sharif 20W51A0533 K. Pallavi 20W51A0532
- 2. A Deep Learning-Based Efficient Firearms Monitoring Technique for Building Secure Smart Cities
- 3. AGENDA 1. Abstract 2. Existing system 3. Proposed system 4. Literature survey 5. Hardware requirements 6. Software requirements 7. References
- 4. ABSTRACT Firearm-related deaths are currently a global phenomenon. It is a threat to society and a challenge to law enforcement agencies. A significant portion of such crimes happen in semi-urban areas or cities. Governments and private organizations use CCTV-based surveillance extensively today for prevention and monitoring. However, human-based monitoring requires a significant amount of person-hours as a resource and is prone to mistakes . The project main focus is to showcase that deep learning-based techniques can be used in combination to detect firearms (particularly guns). This project uses different detection techniques, such as Faster Region-Based Convolutional Neural Networks(Faster RCNN)and the latest EfficientDet-based architectures for detecting guns and human faces.
- 5. EXISTING SYSTEM Multiple boxes are generated due to multiple detectors for the same image. The novelty of our work is to empirically demonstrate that the ensemble of Faster RCNN and the latest EfficientDet architectures provide an improved object detection scheme using the same trained models as the performance of the individual model. In the project title, the term efficient indicates that the detection results can be improved with the existing trained models without further training in the proposed scheme. The project has aimed to contribute as follows: ⋆ Automated detection of human faces and different types of guns together in the wild. ⋆ Introducing a deep learning-based framework to improve the performance of object detection through ensemble. ⋆ Thus, securing smart cities using intelligent surveillance.
- 6. PROPOSED SYSTEM The ensemble of FRCNN and EffcientDet object detection architectures with NMW and WBF combining techniques is the novelty of this paper, specifically in this domain. Some test case images before and after the implementation of ensemble combining techniques A. Faster region-based convolutional neural networks B. EfficientDet C. Mean average precision D. Non-maximum suppression E. Non-maximum weighted F. Weighted boxes fusion G. Ensemble detection scheme
- 7. PARAMETERS • Id • Accused name • Date of Incident • Manner of Killed • Age • Lien ace Name • City • Detected • Gender • Race
- 8. ARCHITECTURE
- 9. SOFTWARE REQUIREMENTS • PYTHON 3.7.0 • WAMP SERVER
- 10. HARDWARE REQUIREMENTS • HARD DISK • PROCESSOR • 4 GB RAM
- 11. ALGORITHMS • Algorithm 1 : Greedy Ensemble Object Detection (ENSD) • Algorithm 2: compute_mAP(mod_set,ImgSet)
- 12. LITERATURE SURVEY TITLE AUTHORS DISCRIPTION RESULT Efficient Firearms Monitoring Technique for Building Secure Smart Cities by Deep Learning TANUPRIYA CHOUDHURY, BISWARANJAN ACHARYA, ANKITA CHATTERJEE and RAJDEEP CHATTERJEE An ensemble (stacked) scheme has improved the detection performance to identify human faces and guns. The comparative results of various detection techniques and their ensembles. It helps the police gather quick intelligence about the incident and take preventive measures at the earliest.
- 13. REFERENCES [1] D. Trottier, ‘‘Crowdsourcing CCTV surveillance on the internet,’’ Inf., Commun. Soc., vol. 17, no. 5, pp. 609–626, May 2014. [2] S. Ren, K.He,R.Girshick, and J. Sun, ‘‘Faster R-CNN: Towards real-time object detection with region proposal networks,’’ in Proc. Adv. Neural Inf. Process. Syst., 2015, pp. 91–99. [3] S. Targ, D. Almeida, and K. Lyman, ‘‘ResNet in ResNet: Generalizing residual architectures,’’ 2016, arXiv:1603.08029. [4] K. He, X. Zhang, S. Ren, and J. Sun, ‘‘Deep residual learning for image recognition,’’ in Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR), Jun. 2016, pp. 770–778. [5] Z. Wu, C. Shen, and A. Van Den Hengel, ‘‘Wider or deeper: Revisiting the ResNet model for visual recognition,’’ Pattern Recognit., vol. 90, pp. 119– 133, Jun. 2019.
- 14. [6] M. Tan, R. Pang, and Q. V. Le, ‘‘EfficientDet: Scalable and efficient object detection,’’ in Proc. IEEE/CVFConf. Comput.Vis.PatternRecognit. (CVPR), Jun. 2020, pp. 10781–10790. [7] M. Tan and Q. V. Le, ‘‘EfficientNet : Rethinking model scaling for convolutional neural networks,’’ 2019, arXiv:1905.11946. [8] A.Egiazarov, V. Mavroeidis, F. M. Zennaro, and K. Vishi , ‘‘Firearm detection and segmentation using an ensemble of semantic neural networks,’’ in Proc. Eur. Intell. Secur. Informat. Conf. (EISIC), Nov. 2019, pp. 70–77. [9] S. Akçay, M. E. Kundegorski, M. Devereux, and T. P. Breckon, ‘‘Transfer learning using convolutional neural networks for object classification within X-ray baggage security imagery,’’ in Proc. IEEE Int. Conf. Image Process. (ICIP), Sep. 2016, pp. 1057–1061. [10] R. Olmos, S. Tabik, and F. Herrera, ‘‘Automatic handgun detection alarm in videos using deep learning,’’ Neurocomputing, vol. 275, pp. 66–72, Jan. 2018.