Artificial_intelligence_in_cctv_survelliance FINAL.pptx

1. Artificial intelligence in cctv surveillance HARIPRASAD MENSE 2MM20CS011 JAY SHAPURKAR 2MM20CS014 RAMA GAWDE 2MM20CS031 SHREEDHAR JAGTAP 2MM20CS036 UNDER THE GUIDANCE OF PROF. MAHESH M

2. contents INTRODUCTION LITERATURE SURVEY EXISTING SYSTEM PROPOSED SYSTEM ARCHITECTURE PROJECT PIPELINE SYSTEM REQUIREMENTS PROJECT FUNCTIONALITY RESULTS CONCLUSION

3. INTRODUCTION AI integration with CCTV security involves the use of machine learning algorithms to enhance the capabilities of traditional CCTV systems. AI algorithms can detect and track objects, recognize faces, and analyze human behavior in real-time, allowing for the identification of potential security threats . The intersection of artificial intelligence (AI) with Closed-Circuit Television (CCTV) marks a transformative leap in the realm of surveillance and security. As technology progresses, the integration of AI algorithms into traditional CCTV systems elevates their capabilities to unprecedented levels.

4. LITERATURE SURVEY Much research has been done in the development of this AI-based intelligent surveillance system and some of this literature is reviewed in this section. Examples of surveillance systems previously studied or developed to have intelligent or automation capabilities: VSAM (Abbreviated as video surveillance and monitoring), PRISMATICA (Abbreviated as proactive integrated systems for security management through institutional technological support and communication ). AI based smart and intelligent surveillance system solves all the problems inherent in video surveillance. It uses computer vision and sensors to detect what is happening and what should be done about it. All this happens in real time and saves money. Detect, prevent and report intruders when they are imminent.

5. EXISTING SYSTEM The existing system for Closed-Circuit Television (CCTV) typically involves traditional surveillance methods where cameras capture video footage, which is then stored or monitored in real-time by human operators. These systems may have limited capabilities for automated analysis, and the focus is primarily on recording and manual review of footage. Here are key characteristics of the existing CCTV systems : Human Monitoring: Human operators are responsible for monitoring live feeds or reviewing recorded footage to identify security incidents or events . Limited Intelligence: The existing systems may lack advanced intelligence features such as real-time object detection, facial recognition, or behaviour analysis.

6. Manual Investigation: Investigations of security incidents often involve manual review of recorded footage, which can be time-consuming and less efficient. Privacy and Ethics: Privacy concerns related to continuous monitoring and recording, as well as potential misuse of footage, are significant considerations in the existing systems. Scalability Challenges: Expanding or upgrading the existing system to accommodate a growing number of cameras or evolving security needs may present challenges. Data Storage Concerns: Storing and managing large volumes of video data can be resource-intensive, and retrieval for specific events may not be as efficient.

7. PROPOSED SYSTEM AI integration with Closed-Circuit Television (CCTV) aims to overcome the limitations of traditional surveillance methods by leveraging advanced artificial intelligence technologies. Here are key features and improvements in a proposed AI-enhanced CCTV system: Real-Time Object Detection: Implement AI algorithms for real-time object detection to identify and track specific entities, such as people, vehicles, or objects of interest. Predictive Analytics: Implement predictive analytics using AI algorithms to identify patterns and trends, enabling proactive security measures and threat prevention. Automated Alerts and Notifications: Enable the system to generate real-time alerts and notifications for security personnel based on AI analysis, facilitating quick responses to potential incidents.

8. HOW AI CCTV WORKS

9. ARCHITECTURE

10. Data acquisition Data processing Training the model Measure performance matrix Testing the model Deployment PROJECT PIPELINE

11. FLOWCHART

12. SYSTEM REQUIREMENTS Software Requirements : Arduino IDE DL/ML algorithms with Jupyter Notebook IDE  Hardware Requirements : Arduino / RasberryPi High end camera GSM module Storage system Modle U sed : CNN+LSTM ( Covlstm )  

13. Accident detection and performing action Robbery detection Real world anamolies detection like fire(burning houses,shops..etc ) PROJECT GOAL/FUNCTIONALITY

14. RESULTS The system detecting the fire and sending alert to the receiver

15. The system detecting the motion of the people

16. Analysing the positions of the equipment for detecting the robbery.

17. CONCLUSION The proposed system not only augments security measures but also acknowledges the importance of ethical considerations, privacy protection, and regulatory compliance. As technology continues to evolve, the integration of AI with CCTV systems promises to redefine the landscape of security, providing a sophisticated and intelligent approach to safeguarding environments in an increasingly interconnected world.

18. Thank You