Suspicious-Activity-Detection-Application

1. Project Overview:

The Suspicious-Human-Activity-Detection-LRCN project is an upgrade to the existing Suspicious-Human-Activity-Detection-VGG16-LSTM system. It focuses on the detection of suspicious human activities in CCTV camera footage using the Long-Short Term Memory (LSTM) and Convolutional Neural Network (CNN) combination known as the LRCN Model. The model aims to classify activities such as walking, running, and fighting into suspicious or non-suspicious classes. Instead of integrating into an existing video surveillance system, the project will develop a cross-platform React Native application for real-time monitoring and alerts.

2. Objectives:

The primary objective of this project is to enhance the accuracy and efficiency of suspicious human activity detection by leveraging the capabilities of the LRCN Model. Specific goals include:

• Improve the accuracy of human activity recognition through the integration of LRCN.
• Extend the range of detectable activities to include walking, running, and fighting.
• Develop a cross-platform React Native application for real-time monitoring and alerts.

3. Project Components:

The project will comprise the following key components:

• LRCN Model Integration: Implementing the LRCN model for improved human activity recognition.
• Activity Classification: Enhancing the model to classify activities like walking, running, and fighting.
• Real-time Monitoring Application: Developing a cross-platform React Native application for real-time monitoring and alerts.
• Integration Testing: Testing the application's integration with the LRCN model and its performance in real-world scenarios.

4. Methodology:

The project will follow a systematic approach:

• Data Collection: Gather CCTV camera footage from a campus environment to train and test the model.
• Preprocessing: Clean and preprocess the data to ensure optimal model training.
• LRCN Model Training: Train the LRCN model using the preprocessed data.
• Activity Classification: Implement algorithms for classifying walking, running, and fighting activities.
• Application Development: Develop a cross-platform React Native application for real-time monitoring and alerts.
• Testing and Deployment: Test the application's functionality and deploy it for real-world usage.

5. Expected Outcomes:

Upon completion, the project is expected to deliver:

• A more accurate and efficient Suspicious-Human-Activity-Detection system.
• Improved classification of human activities, including walking, running, and fighting.
• A cross-platform React Native application for real-time monitoring and alerts.

6. Project Report:

A comprehensive project report will be compiled, documenting the methodologies, challenges faced, solutions implemented, and the overall performance of the system. The report will serve as a valuable resource for understanding and replicating the project.

7. Future Scope:

The project opens avenues for future enhancements, including:

• Integration of additional activity classes for broader recognition.
• Enhancement of the React Native application with additional features such as user authentication and notification customization.
• Collaboration with security agencies for deployment in various environments, including public spaces and commercial establishments.

In conclusion, the Suspicious-Human-Activity-Detection-LRCN project aims to significantly improve the capabilities of intelligent video surveillance systems while providing a user-friendly interface for real-time monitoring and alerts through a cross-platform React Native application.