日本語

Welcome to our repository for the Ishinomaki Hackathon project! In this hackathon, we aimed to create an innovative AI security app that focuses on the theme of the future. Our project is centered around the development of an Violence or Fight Detection System designed for public safety, particularly in identifying and addressing incidents involving fights using object detection and deep neural networks.

In today's world, public safety is of utmost importance, and rapid response to incidents such as fights is crucial to ensure the well-being of individuals and communities. To address this challenge, we set out to create an AI-based solution that can identify and report such incidents in real-time.

Here is gist of the outcome:

Our project focuses on the following key aspects:

• Object Detection: We have implemented object detection algorithms to identify specific actions and behaviors, such as physical altercations or fights, in live video streams or recorded footage.
• Fight Classification: Deep neural networks are employed to classify the severity and nature of the altercation based on the detected actions.
• Alerting Mechanism: When a fight is detected and classified, the app generates an immediate alert. It includes images or footage of the scene. This holistic data provides a human perspective to authorities for better decision-making and rapid response. for this system we are using smtp but any alert mechanism can be used.

1. Object Detection: Our system analyzes video feeds in real-time, detecting objects and activities, including potential fights.
2. Classification The detected actions are then processed through deep neural networks that classify the level of violence and threat involved.
3. Alert Generation: If a fight is confirmed, an alert is sent to the nearest police station. The alert includes vital information for a swift response.
4. Response: Authorities receive the alert, enabling them to assess the situation and dispatch personnel to the scene as quickly as possible.

• Here is the main repository which contains the brain of the project.

To Learn more, click here.

• Rakibul Haque (Lead Developer)
• Kamada
• Wakamatsu
• Jin
• Ichimura

We would like to express our gratitude to the organizers of the Ishinomaki Hackathon for providing us with the opportunity to work on this important project.

This project is currently under the MIT license. Feel Free to Use it.

For any inquiries or feedback, please contact us at Contributer's section.