Intelligent Camera Decision-Making System for Enhanced
Surveillance
Problem Statement
Design and implement an AI-based intelligent camera decision-making system to address challenges
in surveillance and security, with a focus on accurate event detection, real-time decision-making, and
seamless integration with existing systems.
Solution
Key Features
• Accurate event detection in real-time.
• Swift decision-making capabilities.
• Seamless integration with existing surveillance systems.
• Adaptive learning for continuous improvement.
• Resource optimization.
Technical Approach
• Leveraging the machine learning, deep learning, computer vision, IoT, devops, and image/real-time
video data analytics and processing.
• Integration of real-time data from cameras .
• Development of algorithms for event detection and decision-making.
• Continuous refinement of decision-making based on feedback.
Challenges
• Differentiating between suspicious and normal activities.
• Lack of publicly available datasets for training.
• Complexity introduced by multiple individuals in monitored environments.
• Utilization of existing cameras for intelligent decision-making.
• Addressing issues related to tracking gang wars in jails.
Software and Technology
• Programming Languages: Python, c, Mojo
• Frameworks: TensorFlow, PyTorch
• Tools: OpenCV for computer vision
• Cloud Services: AWS Cloud9
1
Team Members & Responsibilities
• Akash Dogra - Integrating mobile & sensor for data collection:
– Implementing the integration of sensors into the mobile device or standalone system.
– Developing functionalities for real-time data collection.
• Sarthak Bhatt - Deployment:
– Focusing on deploying the developed system onto the selected platform.
– Ensuring seamless functionality in real-world scenarios.
• Tejas Thapliyal - Deployment:
– Focusing on deploying the developed system onto the selected platform.
– Ensuring seamless functionality in real-world scenarios.
• Srishti Sharma - Deployment:
– Focusing on deploying the developed system onto the selected platform.
– Ensuring seamless functionality in real-world scenarios.
2
Flow Chart / Graphical Representation
Start
Data
Collection
Sensor
Integration
Position
Detection
Obstacle
Detection
Audio-
based
Instruc-
tions
Real-time
Feedback
End
Figure 1: System Flow Chart
Schedule
• Week 1-2: Planning and Research
• Week 3-4: Sensor Integration and Data Collection
• Week 5-6: Decision-Making Algorithms
• Week 7-8: Continuous Improvement and Feedback
• Week 9-10: Integration and Testing
• Week 11-12: Deployment and Finalization
3
Conclusion
(Conclude with a brief statement reaffirming your team’s commitment to delivering a cutting-edge solu-
tion that significantly improves surveillance and security systems for the benefit of society.)
