Open Access Open Access  Restricted Access Subscription Access

Design and Implementation of Collaborative Cloud-Edge System Using Raspberry Pi for Video Surveillance System with AIoT to Analyse Effective Performance Parameters of Network

Dr. Trupti Kaushiram Wable

Abstract


The video surveillance can avoid many crimes as well as it will help to reduce crime rate in society as well we can save many lives. But currently implemented IoT system having various limitations like insufficient storage capacity and inadequate processing of information. Thus we can integrate traditional IoT system with Artificial Intelligence (AI) models to improve storage capacity & processing called as Artificial Intelligence of Things (AIoT). This system mainly focuses on performance parameter of video surveillance system the parameter consist of Response Latency Time, Network Bandwidth & Storage on server. In proposed system divided in two part, First part include Edge node implemented with Raspberry Pi as IoT system which having video input then it perform image processing & store output on edge node, second part include cloud node which is train with AI model as AI system to extract image and analyzed performance of system. So Cloud-Edge Collaborative system refers as Artificial Intelligence of Things (AIoT). In this research I conclude comparative study of traditional Cloud Computing System with Collaborative Cloud-Edge Computing system which shows that, the Response Latency Time improve by 5 times, Network Bandwidth improve by 10 times and storage capacity improve by 5 times of traditional Edge Computing System.


Full Text:

PDF

References


Hill A.R., Suhail Y., Anjum E. (2020). Enhancing Video Surveillance Using Edge Computing and Cloud Computing. International Journal On Advances in Internet Technology, 13(3&4), pp.203–217.

Kawsar F., Ahmad S., Ali M., et al. (2018). An Overview of Cloud-Edge Computing and Its Impact on Streaming Video Surveillance. ACM Computing Surveys, 51(2), 14.

Liu G., Wang Y., Zhang Z., et al. (2018). Approaches to Secure Video Surveillance Using Edge Computing and Cloud Computing. International Journal of Electrical and Computer Engineering, 8(3), 1792.

Chen X., Liu Z., Li H., et al. (2019). Research on Video Surveillance Technology Based on AIoT Based on Machine Learning. International Journal of Information Technology, 5(4), 179–183.

Alhaj F., Alkhalloufi B., Yousef Z. et al. (2019). AIoT-based Video Surveillance in Edge Computing Era. 2019 International Conference on Advanced Computer Science and Information Systems (ICACSIS), pp.1–7.

Wang Y., Yue P., Zhao X., et al. (2020). Video Surveillance Optimization with AIoT: A Survey. IEEE Internet of Things Journal, 7(3), 2440–2457.

Chen X., Zhao Y., Xu B., et al. (2018). A Survey of Smart Video Surveillance Based on Cloud Computing and Automated Machine Learning. IEEE Access, 6, 59072–59088. 8. Dong S., Wang C., Zang X., et al. (2017). A Multi-layer Cloud-Edge Computing for Video Surveillance. IEEE Access, 5, 14566– 14573.

Chauhan D.M., Chauhan R.K., Arya R. (2019). Collaborative Edge and Cloud Computing for Video Surveillance System. 2019 5th International Conference on Computing for Sustainable Global Development (INDIACom), pp.1428–1432.

Emir Yeşilot, Bilal Al-Najjar, Rıza Akçalı, Irfan Aslan, “Cloud- Based Adaptive Surveillance System Using Edge Computing”, IEEE Access, vol. 6, pp. 67990-68002, 2018.

A. Garrigues, T. Alonso, A. Rivas, „Smart camera networks and services for context-aware surveillance applications”, Electronics,vol. 7, nº 12, pp. 1-14, 2018.


Refbacks

  • There are currently no refbacks.