Scribd
Upload
11 views14 pages

Projectk 3

Uploaded by

neerajboggavarapu098
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
11 views14 pages

Projectk 3

Uploaded by

neerajboggavarapu098
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 14

AI-Based Violence Detection System for

Street and Road Surveillance


Real-time detection and alert system
Team details

Name: C. RAHUL
ID: R200125
Tech Stack: Python, Machine Learning, openCV, Data analysis, EDA, Feature Engineering
CGPA: 8.67
Contact: 8096740366

Name: G. RAMESWARA REDDY


ID: R200355
Tech Stack: Data analysis, EDA, Machine Learning, Basic deep learning, Statistics, Feature
Engineering.
CGPA: 9.02
Contact: 6303830152
Team details
Name: B. N V S K NEERAJ
ID: R200137
Tech Stack: Python, openCV, Machine Learning, Basic deep learning, Data analysis, EDA, Feature
Engineering, Flask-python.
CGPA: 8.56
Contact: 7032370979

Name: M. GOVARDHAN REDDY


ID: R200313
Tech Stack: Deep learning, Machine Learning, Data analysis, Feature Engineering,
fastAPI
CGPA: 8.71
Contact: 9392648201
Introduction
Overview
1)Importance of public safety on streets and roads….

2)Role of technology in enhancing security….

Objective
-Develop a system to detect violence and notify authorities
in real-time

Problem Statement
The current manual monitoring systems for street and road
surveillance face delays, inefficiencies, and high costs,
necessitating an automated solution for real-time violence
detection and faster response times.
Proposed Solution

AI-Based System:

The proposed solution involves developing an AI-based system that


leverages advanced computer vision and machine learning
technologies to enhance street and road surveillance. The system will
analyze CCTV footage in real-time to detect violent incidents such as
fights and other aggressive behaviors. Upon detecting such incidents,
the system will automatically send alerts to the nearest authorities,
enabling them to respond promptly and effectively.
Key features

Real-time analysis of CCTV footage


Detection of fights and violent incidents


Automatic alerts to authorities


Snapshot and video clip generation


Location-based alert system


Integration with existing CCTV infrastructure
System Architecture
Workflow
Components

Data capture from CCTV

CCTV Cameras

Real-time processing and

AI Detection Module detection

Notification System 
Alert generation and
transmission

Police Station
Interface
Sample Images:
Implementation Plan
01 Requirement Analysis:
Activities: documentation, and finalizing
requirements. 02 System Design:
Activities: Architectural planning,
interface design, and review.

03 Development and
Integration:
Activities: Coding, algorithm development,
system integration.
04 Testing and Validation:
Activities: Unit testing, integration testing,
user acceptance testing. 05 Deployment:

Activities: Implementation,
monitoring, and initial support.
Technologies Used

Machine Learning Frameworks: Numpy, Pandas, sklearn


Visualization tools: Matplotlib, seaborn, plotly, tableau


Deep Learning Frameworks: TensorFlow, PyTorch


Object Detection Models: YOLO (You Only Look Once), LSTM (Long Short


-Term Memory), CNN (Convolutional Neural Network)


Video Processing: OpenCV


Backend Frameworks: Flask, Django


Messaging Queues: RabbitMQ, Kafka


Notification Services: Twilio, Email API
Future Work

Expansion to other public areas


Integration with emergency services


Continuous improvement with more data


Using more Advanced AI techniques


Collaborative efforts with law enforcement
agencies
Expected Outcomes and Conclusion

1)Benefits:

Improved response times

Enhanced public safety

Efficient use of surveillance resources


Impact:

Reduction in street and road violence

Better resource allocation for law enforcement
Thank You

You might also like