International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 05 | May 2024
p-ISSN: 2395-0072
www.irjet.net
AI Based Smart CCTV video Surveillance Application Kuntal Pal1, Manas Kumar Roy2 and Aashutosh Kumar3 1Kuntal Pal Department of Information Technology, Dr. B C Roy Engineering College, Durgapur 713206, India
2Professor Manas Kumar Roy Department of Information Technology, Dr. B C Roy Engineering College, Durgapur
713206, India
3Aashutosh Kumar Department of Information Technology, Dr. B C Roy Engineering College, Durgapur 713206,
India ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Video surveillance nowadays is getting popular
With this app, you can connect to any Wi-Fi-enabled camera to safeguard the area it covers. Our goal is to enhance a regular camera through this app, making it accessible to everyone at a very affordable price.
day by day. Schools, hospitals, homes, market places, roads require costly high end video cameras require mostly RFIDS features which are mostly expensive hence security and all becomes costly. In this paper, AI based surveillance camera application has been proposed named “ smart CCTV video surveillance “to identify and report suspicious events. The proposed model for the Smart CCTV will involve using a combination of different computer vision and machine learning techniques to detect and recognize different objects and events in the surveillance area. This paper mainly focuses on object detection algorithm (YOLO) to detect and track weapon within the video frames, applying non maximum suppression (NMS) to remove redundant boundary boxes and finally triggers an alarm or alert indicating if a weapon is detected within the surveillance area. In this research paper an algorithm has been also developed successfully that uses the HSV color space for fire detection and Simple Mail Transfer Protocol (SMTP) can be used to notify relevant parties when a fire is detected by sending an email with relevant information.
Simply purchase a basic, low-cost Wi-Fi-Bluetooth camera from the market, connect it to our app, and open the app. You'll find four main buttons on the home screen. When you activate these buttons, four screen tabs will appear on your mobile screen, each for a different function. These tabs serve distinct purposes: one for recording regular footage, another for verifying individuals entering the camera's vicinity, a third for detecting fire and issuing immediate alerts, and the last for identifying weapons and notifying accordingly. So, we have taken up these challenges on our project to bring all those facilities to smart CCTV.
2. Literature Review
Key Words: Surveillance system, RFIDS, AI, CCTV,YOLO,NMS
G. Sreenu, et al. had described in 2019, their paper is a applications of deep learning and the smart security-based surveillance system. According to this paper all the information about how to improve the CCTV system with deep learning and its use are provided.[2]
1.INTRODUCTION The research is based on a smart CCTV camera surveillance system. A security system based on CCTV cameras currently exist in different parts of our city, state or country. The system is designed using wireless CCTV cameras Technology. Using surveillance systems, image recognition is becoming more important in today's life [1]. An embedded surveillance systems are often used for home, office or factory for image processing and traffic monitoring. Here in our case, we build our research with many important features, like a smart guard, which gets all the information like entering the area/room, fire detection etc. The current CCTV security system provides very basic security, which need to be modified or modernize rise with time. Today’s modern CCTV will easily be able to take all decisions by itself and this multifeatured CCTV system will be able to recognize a person, detect whether a person is carrying a firearm or weapon. In our opinion, modern day CCTV can't make our life so much secure, we feel a bit secure with the current CCTV system. Let's dive into our project details. First and foremost, it's a smart AI-based security camera mobile app.
© 2024, IRJET
|
Impact Factor value: 8.226
Prof S.B. Kothari et al. had published in Dec 2021, a paper which was based on survey on smart security surveillance system. This paper is based on some idea of motion detection and clear image detection using an image processing method.[3] Smart CCTV security service in IOT Environment ,in this paper the author Cho, Jeorg, Rae ,published his research paper provide how we can prevent crime in any CCTV monitored area by using IOT system. What is IOT and its usage rules also mentioned in this paper. [4] Automated Detection of Firearms and Knives in a CCTV Image ,in their paper Micha Grega et al. in 2016 .In this paper, there are some idea of detection the weapon inside the surveillance area and know about the danger notification alert process from SURF feature detection algorithms.[5]
|
ISO 9001:2008 Certified Journal
|
Page 81