See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/367376895

AI Based Smart Surveillance System

Article in International Journal of Scientific Research in Science Engineering and Technology · January 2023
DOI: 10.32628/IJSRSET229672

CITATIONS READS

2 8,275

1 author:

Rasadurai Kumaravel
Kuppam Engineering College
36 PUBLICATIONS 45 CITATIONS

SEE PROFILE

All content following this page was uploaded by Rasadurai Kumaravel on 24 January 2023.

The user has requested enhancement of the downloaded file.

 International Journal of Scientific Research in Science, Engineering and Technology
Print ISSN: 2395-1990 | Online ISSN : 2394-4099 (www.ijsrset.com)
doi : https://doi.org/10.32628/IJSRSET229672

AI Based Smart Surveillance System

J Bhavyasri, Dr. G N Kodanda Ramaiah, Dr. K Rasadurai
Department of ECE, Kuppam Engineering College, Kuppam, Andhra Pradesh, India

ABSTRACT
The AI Based smart surveillance system is gaining huge attention because of rise
Article Info in demand for safety and security. The surveillance system is designed to analyse
the video, image, and audio or any kind of surveillance data automatically
Publication Issue : without any human involvement. The developments that happened in recent
Volume 10, Issue 1 years in the computer vision, sensor devices and Auto ML is playing a keen role
January-February-2023 in accrediting such intelligent system.
There are many kinds of surveillance and security systems present in the market,
Page Number : 10-15 but there is no live decision making and predictive analysis surveillance i.e.,
automated self-decision making system which helps the different public
Article History departments like police, health, fire and many more to track and reach the
Accepted : 01 Jan 2023 particular location where the incident is happened. In this proposed project we
Published: 15 Jan 2023 provide the AI based intelligent surveillance and security system which analyses
and takes the decision immediately by itself based on the present parametric
conditions as per the modules trained by us, It helps in providing the alerts
quicker compared to the traditional alert system. The system which is proposed
here will react automatically based on the situations occurs as it is trained in
various modes. This paper is intending to provide the generalised outline of the
AI based smart surveillance system and its functionalities. This paper also
consists of information regarding the core processing steps of AI based
surveillance system such as tracking, object detection and classification, and
behavioural analysis.
Keywords : Artificial Intelligence, Machine Learning, Computer Vision,
Predictive Analysis, Object Detection, Behavioural Analysis.

I. INTRODUCTION importance in order to protect the personal

information of each and every individual also their day
Now a days, the safety and security is becoming very to day activities and valuable things. Large amount of
challenging in modern time society to stop the issues closed circuit security cameras with different modules
facing by people in their life and also their assets which of sensors, have been designed developed and deployed
are valuable from illegal handling. As a result, the to monitor the critical infrastructure facilities such as
personal social safety and security is attaining greater airports, military bases, banking, power plants,

Copyright: © the author(s), publisher and licensee Technoscience Academy. This is an open-access article distributed under the
terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-commercial use,
10
distribution, and reproduction in any medium, provided the original work is properly cited
 J Bhavyasri et al Int J Sci Res Sci Eng Technol, January-February-2023, 10 (1) : 10-15

campuses, etc. The monitoring of security cameras in intelligent camera system, camera network (existing
manual mode by human operators is not an efficient CCTV), audio surveillance, transmission system,
solution or even not practical as the human resource is operator and the main server (MIFSA) also.
expensive and has limited ability [1]. AI based smart AI based smart and intelligent surveillance system
surveillance system is designed to monitor the solves all the problems inherent in video surveillance.
environment and infrastructure automatically without It uses computer vision and sensors to detect what is
human involvement. The various monitoring tasks happening and what should be done about it. All this
includes the auto detecting and object tracking (like happens in real time and saves money. Detect, prevent
vehicle or human) and analysis of further process and and report intruders when they are imminent.
taking necessary actions. The artificial intelligence, The main intention of this document is to give an
image processing, and signal processing, techniques detailed overview of the smart surveillance devices and
play a crucial role in developing such intelligent and to review the current existing systems for each of its
smart system. processing steps.
Artificial intelligence for video surveillance uses The overview of the AI-based intelligent surveillance
computer programs that analyses images and audio system
from video surveillance cameras to identify people, A large number of security cameras, the intelligent
vehicles, objects, attributes, and events. The AI surveillance (ISS) is a system that has the intelligent
program works with the help of computer vision. ability to analyse the surveillance data automatically
Computer vision is a set of mathematical methods or and take required actions which are necessary such as
algorithms that work like a series of questions or flow generating alarms or warnings. ISS is an multi-
chart to differentiate the viewed object with the disciplinary subject that includes computer vision,
thousands of reference pictures of people in angles, electronics (sensing), machine learning (artificial
poses, movements and positions which are stored. intelligence), and pattern recognition, communication,
networking and other areas. The ISS is compatible to
II. REVIEW OF LITERATURE make it implemented in various applications and
different environments.
Much research has been done in the development of
this AI-based intelligent surveillance system and some Some typical applications are listed below.
of this literature is reviewed in this section. Examples
of surveillance systems previously studied or • Home security and intrusion detection.
developed to have intelligent or automation • Home care and security.
capabilities: VSAM (Abbreviated as video surveillance • Public transportation places, like bus/train
and monitoring), PRISMATICA (Abbreviated as terminal, seaport, airport etc.
proactive integrated systems for security management • Public area, such as university, campus,
through institutional technological support and government building.
communication), ADVISOR (video digitally annotated • Traffic monitoring.
for intelligent monitoring and optimized recovery). Fig. • Crown management and analysis.
1 shows the general description of the PRISMATICA • Pedestrian detection and autonomous car.
technology proposed and implemented to improve the • Military remote surveillance,
safety of passengers in the transport system of public • border surveillance
vehicles. It consists of many main elements like • Perimeter surveillance for headquarters,
• corporations, etc.

International Journal of Scientific Research in Science, Engineering and Technology | www.ijsrset.com | Vol 10 | Issue 1 11
 J Bhavyasri et al Int J Sci Res Sci Eng Technol, January-February-2023, 10 (1) : 10-15

III. EXISTING SYSTEM IV. PROPORSED SYSTEM

Fig. 1. An overview of PRISMATICA system

• In the evaluation of PRISMATICA machine, here

we have the Another staggering surveillance Fig. 1. The schematic diagram of proposed system
machine named DARPA ARGUS-IS (The In the proposed machine system first the digital camera
Autonomous real time ground ubiquitous (digicam) captures the video and constantly transmits
surveillance imaging system). With one point the information to the microcontroller. Here the
eight (1.8) Giga pixels video machine, ARGUS-IS microcontroller is trained with several video modules
is capable to auto-tune every moving targets with which includes numerous situations like accidents,
in a range of forty kilometres (equal to the size of thefts, harassments etc… Once the controller gets the
small city) the usage of individual platform. The information from the digicam it continuously analyses
availability of such commercialised ISS products the information initially and process it based on the
are DETER (detection of events and activities for video type and pixel rate to make the video format
threat evaluation and recognition) and DETEC AS supported for the next steps. After processing the
(www.detec.no). Intelligent surveillance machine information, the data acquired by the system is
may also plays a major role in safety and security compared with the trained data modules to
in private as well as the public domains. However, differentiate the scenario according to them. After
it is highly complicated due to few sensible issues differentiation the information is transmitted to the
such as receiver stating that the particular situation such as
• Performance: which includes the machine theft accident or harassment is happened in the
accuracy specific place. The place is detected using the GPS
• Robustness: the machine ought to be sturdy, again Module integrated with the micro controller.
the regular issues includes illumination variation,
occlusion, clutter, climate change, camouflage,
etc.
• Reliability
• Real time constrain: the machine should quick
enough to respond.
• Low price (Cost effective)

International Journal of Scientific Research in Science, Engineering and Technology | www.ijsrset.com | Vol 10 | Issue 1 12
 J Bhavyasri et al Int J Sci Res Sci Eng Technol, January-February-2023, 10 (1) : 10-15

V. RESULTS machine performance. Numerous strategies were

proposed to address numerous primary processing
steps like background-foreground segmentation, item
detection and classification, monitoring and
behavioural evaluation. extraordinary aggregate of
sensor modality ought to be explored to make the
machine sturdy or to simplify the processing system.
Current research in behaviour evaluation are
nonetheless thinking about simplified scene, and
therefore extra practical and complicated scene ought
to be investigated. With reducing rate in sensor and
Fig 3. The system detecting the fire and sending alert processing devices, researchers ought to additionally
to the receiver remember investigating and growing a low value wise
surveillance machine.

VII. REFERENCES

[1]. N. Sulman, T. Sanocki, D. Goldgof and R.

Kasturi, How effective is human video
surveillance performance?, 19th International
Confer ence on Pat tern Recognition, Tampa, FL,
2008, pp.1-3R.
Fig 4. The system detecting the motion of the people [2]. R.J. Radke, S. Andra, O. Al-Kofahi and B.
Roysam, Image change detection algorithms: A
systematic survey, IEEE Trans. Image Process.
14 (2005) 294–307.
[3]. M. Cristani, M. Farenzena, D. Bloisi and V.
Murino, Background Subtraction for Automated
Multisensor Surveillance: A Compreh en sive
Review, EURASIP Journal on Advances in
Signal Processing 2010 (2010) 343057.
Fig 5. Analysing the positions of the equipment for [4]. T. Bouwmans, F. El Baf, and B. Vachon,
detecting the robbery Statistical background modeling for foreground
detection: A survey, Handbook of Pat tern
VI. CONCLUSION
Recognit ion and Computer Vision (Volume 4).
Singapore: World Scient ific, 2010.
In this paper, the general overview and evaluate of the
[5]. T. Bouwmans, Recent advanced statistical
surveillance systems has been presented. Such
backgro und mod el in g fo r foreground
intelligent devices are promising to be carried out in
detection: A systematic survey, Recent Patents
numerous environments and applications. This paper
Comput . Sci. 4 (2011) 147–176.
additionally has mentioned a few feasible sensor
[6]. S. Brutzer, B. Hoferlin, G. Heidemann,
modalities and their fusion situations to enhance the
Evaluation of background subtraction

International Journal of Scientific Research in Science, Engineering and Technology | www.ijsrset.com | Vol 10 | Issue 1 13
 J Bhavyasri et al Int J Sci Res Sci Eng Technol, January-February-2023, 10 (1) : 10-15

techniques for video surveillance, IEEE [16]. M. Cristani, R. Raghavendra, A. Del Bue and V.
Conference on Computer Vision and Pattern Murino, Human behavior analysis in video
Recognition (CVPR), Colorado Sp r in gs, USA, surveillance: A Social Signal Processing
2011, pp. 1937-1944. perspective, Neurocomput ing 100 (2013) 86-97.
[7]. A. Sobral, A.Vacavant , A comprehensive review [17]. D. Gowshikaa, S Abirami and R Baskaran,
of background subtraction algorithms evaluated Automated Human Behaviour Analysis from
with synthetic and real videos, Computer Vision Surveillance videos: A survey, Art ificial
and Image Understanding 122 (2014) 4-21. Intelligence Review 42 (2014) 747-765.
[8]. M. Paul, S. M. Haque and S. Chakraborty, [18]. R. Gade and T. B. Moeslund, Thermal cameras
Human detection in surveillance videos and its and applications: A survey, Machine Vision &
applications-a review, EURASIP Journal onar Applications 25 (2014) 245-262.
Advances in Signal Processing 2013(2013) 1-16. [19]. A. Javed, A. Ejaz, S. Liaqat , A. Ashraf and M.B.
[9]. M. Enzweiler and D. M. Gavrila, Monocular Ihsan, Automatic target classifier for a Ground
pedestrian detection: Survey and experiments, Surveillance Radar using lin ear d iscrimina nt
IEEE Trans. Pattern An aly sis an d Machine analysis and Logistic regression, Radar
Intelligence 31 (2009) 2179-2195. Conference (EuRAD), 2012 9 th European,
[10]. P. Dollar, C. Wojek, B. Schiele and P. Perona, Amsterdam, 2012, pp.302-305.
Pedestrian Detection: An Evaluation of the State [20]. D. Kocur, P. Kazimir, J. Fortes, D. Novak, M.
of the Art, IEEE Transactions on Pattern Drutarovsky, P. Galajda and R. Zet ik, Short-
Analysis and Machine Intelligence 34 (2012) range UWB radar: Surveillance robot equipment
743-761. o f the future, IEEE Internat ional Conference on
[11]. A. Yilmaz and M. Shah, Object tracking: A Systems, Man and Cybernet ics (SMC), San
survey, Journal ACM Comput ing Surveys 38 Diego, CA, USA, 2014, pp.3767-3772.
(2006) article 13 (45 pages). [21]. L. Spinello, K. O. Arras, R. Triebel and R.
[12]. A.W.M. Smeulders, Dung M. Chu, R.Cucchiara, Siegwart , A layered approach to people
S. Calderara, A. Dehghan and M. Shah, Visual detection in 3D range data, AAAI Conference on
Tracking: an Experimental Survey, IEEE Ar t i f icial Intelligence (AAAI-10), Atlanta,
Transaction on Pattern Analysis and Machine Georgia, USA, 2010.
Intel l igence 36 ( 2 014 ) 1442-1468. [22]. L. Spinello, M. Luber and K.O. Arras, Tracking
[13]. A. Bedagkar-Gala and Shishir K. Shah, A survey people in 3D u sing a bottom-up top-down
of approaches and trends in person re- detector, 2011 IEEE International Co n feren ce
identification, Image and Vision Comput ing 32 on Robot ics and Automation (ICRA), Shanghai,
(2014) 270-286. 2011, pp.1304-1310.
[14]. Y. Wu, J Lim and M Yang, Object Tracking [23]. Csaba Benedek, 3D people surveillance on range
Benchmark, IEEE Transactions on Pattern data seq uences o f a rotating Lidar, Pattern
Analysis and Machine Intelligence 37 (2 015 ) Recognition Letters 50 (2014) 149-158.
1834-1848. [24]. Benny Ping Lai Lo, Jie Sun and Sergio A.Velast
[15]. T. Ko, A survey on behavior analysis in video in, Fusing Visu a l a nd Audio Information in a
surveillance for homeland security applications, Distributed Intelligent Surveillance S ystem fo r
37th IEEE Applied Imagery Pattern Reco gn i t Public Transport Systems, Acta Automatica
ion Workshop, Washington, DC, USA, 2008, Sinica, 29 (2003) 393-407.
pp.1-8.

International Journal of Scientific Research in Science, Engineering and Technology | www.ijsrset.com | Vol 10 | Issue 1 14
 J Bhavyasri et al Int J Sci Res Sci Eng Technol, January-February-2023, 10 (1) : 10-15

[25]. D.L. Hall and J. Llinas, An introduction to Cite this article as :

multisensor data fusion, Proceedings of the IEEE
85 (1997) 6-23. J Bhavyasri, Dr. G N Kodanda Ramaiah, Dr. K
[26]. E. F. Nakamura, A. A. F. Laureiro and A. C. Rasadurai, "AI Based Smart Surveillance System ",
Frery, Informat ion fu sion for wireless sensor International Journal of Scientific Research in Science,
networks: Methods, models, and classifications, Engineering and Technology (IJSRSET), Online ISSN :
ACM Comput . Surveys 39 (2007) 1–55. 2394-4099, Print ISSN : 2395-1990, Volume 10 Issue 1,
[27]. E.I. Gokce, A.K. Shrivastava, Jung Jin Cho and pp. 10-15, January-February 2023. Available at doi :
Yu Ding, Decision Fusion from Heterogeneous https://doi.org/10.32628/IJSRSET229672
Sensors in Surveillance S en sor S ystems, IEEE Journal URL : https://ijsrset.com/IJSRSET229672
Transactions on Automation Science and
Engineerin g 8 (2 011 ) 228-233.

AUTHORS PROFILE

J Bhavyasri completed B.Tech in the stream of

Electronics and Communication Engineering from
Kuppam Engineering College, and studying M.Tech in
the stream of Embedded systems from Kuppam
Engineering College, Kuppam. She is Athirst to
exploring various technologies like Embedded devices,
Artificial intelligence and IoT.
Dr G N Kodanda Ramaiah, M. Tech, Ph.D. is working
as Head of the department of ECE Kuppam engineering
college. He had published 25+ papers in international
journals and filed 12+ patents. He has organized
multiple entrepreneurship orientation programs and
technology workshops on analog electronics ad
embedded tools. His areas of interest are
instrumentation and communications, signal
processing, Digital electronics, IoT and embedded
systems.
Dr K Rasadurai, M.E, Ph.D. is working as professor in
the department of ECE, Kuppam Engineering College.
He published 20+ papers in international journals and
filed 7 patents, He has organized multiple technology
related orientation programs and workshops on analog
electronics and matlab tools. His areas of interest are
Image processing, Analog and digital electronics,
Embedded systems and Artificial Intelligence

International Journal of Scientific Research in Science, Engineering and Technology | www.ijsrset.com | Vol 10 | Issue 1 15

View publication stats