Signal-X

🚧 Problem Statement

❗ The Urban Gridlock Crisis

With rapid urbanization and an ever-increasing number of vehicles, cities are choking under intense traffic congestion. What should be smooth commutes turn into daily nightmares, leading to:

• ⏱️ Wasted Time stuck in endless queues
• ⛽ Higher Fuel Consumption and rising transportation costs
• 🌫️ Increased Pollution and carbon emissions
• 🚨 Delayed Emergency Responses for ambulances, fire trucks, and police
• 😠 Driver Frustration and rising road rage incidents

🚦 The Flaws of Conventional Traffic Lights

Traditional traffic systems follow fixed-timer logic, blindly rotating signals without considering real-time traffic flow. This outdated mechanism leads to:

• 🟢 Green lights on empty roads, wasting valuable intersection time
• 🔴 Vehicle pile-ups in busy lanes, creating bottlenecks
• 🚑 Emergency vehicles stuck, unable to move swiftly through intersections
• 🧍‍♂️ Manual control dependency, requiring more manpower
• 🧠 No AI, No Adaptability, making systems unresponsive and inefficient

It’s time to move beyond timers and take the smart route to traffic management.

💡 Our Solution – Signal-X

Signal-X is an AI-driven Smart Traffic Management System that transforms urban mobility using the power of computer vision, IoT, and geospatial intelligence. Designed to be scalable, real-time, and intelligent, Signal-X optimizes traffic flow, reduces congestion, and ensures faster emergency response through dynamic signal control.

🚀 Key Features

• 🧠 AI-Powered Traffic Analysis
  YOLOv8-based object detection continuously monitors vehicle flow across all junction lanes.

• ⏱️ Dynamic Signal Adjustment
  Adaptive signal control based on real-time congestion and traffic density patterns.

• 🌐 IoT Integration
  Raspberry Pi/Arduino-based controllers ensure seamless traffic light control and emergency detection.

• 🗺️ Geospatial Intelligence
  Integration with MapMyIndia's API offers live congestion updates and smart rerouting.

• 🚑 Emergency Vehicle Prioritization
  Automatically detects emergency vehicles (ambulance, fire brigade, police) and gives them green-light priority.

• 🔮 Congestion Prediction
  AI models analyze historical + real-time data to predict and prevent traffic bottlenecks.

🧩 How Signal-X Works – Workflow

Signal-X operates through a seamless, AI-driven pipeline that integrates video processing, intelligent decision-making, and hardware-level execution.

⚙️ Step-by-Step Workflow

1. 🎥 Live Data Collection

   • RTSP streams from CCTV cameras and drones provide real-time traffic video feeds.

2. 🗺️ Traffic API Integration

   • Feeds are passed to MapMyIndia Traffic API for real-time congestion data enrichment.

3. 🧠 Vehicle Detection using YOLOv8

   • Vehicles are identified and classified into 5 categories:
     🚑 Ambulance, 🧬 Organ Transport, 🚒 Fire Brigade, 🚓 Police, 🆘 Disaster Units

4. 🔄 Real-Time Multiprocessing

   • Uses Python’s multiprocessing to analyze all four lanes at a junction simultaneously for accurate density analysis.

5. 🧮 Smart Signal Optimization

   • A custom AI model determines the optimal green-light timing using:
     • 📊 Vehicle Density
     • 🕓 Historical Traffic Patterns (via Reinforcement Learning)
     • 📡 Live API Congestion Data

6. 🔌 IoT-Based Hardware Integration

   • Arduino module integrated into the junction system.
   • If an emergency vehicle is detected, immediate green signal is activated and maintained until it passes.

7. 🖥️ Real-Time Dashboard (React)

   • Officers monitor and visualize traffic flow and system decisions via a React-based dashboard in real-time.

🎯 Why Signal-X?

Unlike traditional fixed-timer systems, Signal-X uses real-time intelligence and automation to enhance traffic efficiency, reduce delays, and save lives.

✅ Smart
✅ Scalable
✅ Emergency-Ready
✅ Real-Time
✅ AI-Optimized

🔧 Technical Architecture

Components Overview

1. Data Collection Layer

   • CCTV cameras & drones capture live video feeds
   • Microphone sensors detect emergency vehicle sirens
   • IoT devices (Raspberry Pi & Arduino) process sensor inputs

2. AI Processing Layer

   • YOLOv8 for vehicle detection and classification
   • DeepSORT for vehicle tracking
   • OpenCV for image processing
   • Flask for API management

3. Backend Processing

   • MongoDB for data storage
   • WebSocket for real-time communication
   • Traffic analysis algorithms

4. Frontend Dashboard

   • React-based responsive interface
   • Real-time traffic analytics visualization
   • MapMyIndia integration for geospatial display

5. Signal Control System

   • Priority-based signal switching for emergency vehicles
   • Adaptive traffic light control based on congestion analysis

💻 Technology Stack

🧰 Tech Stack – What Powers Signal-X 🚦

Signal-X is built using a powerful blend of AI, hardware, geospatial analytics, real-time systems, and scalable web technologies. Here's everything under the hood:

🌐 Frontend – React Dashboard

🧠 AI & Computer Vision – Smart Traffic Brain

🧪 Backend – Core Logic & APIs

🔌 Hardware – IoT-Powered Signal Controller

🎮 Bonus Tools & Utilities

🚀 Signal-X combines real-world sensors, cutting-edge AI, and cloud-based data pipelines to create a future-ready traffic management solution.

🚀 Getting Started

Prerequisites

• Python 3.8+
• Node.js and npm
• MongoDB
• MapmyIndia API key
• Gemini API key
• Arduino IDE

Installation

1. Clone the repository

   git clone https://github.com/ArshTiwari2004/Signal-X.git
   cd Signal-X
   

2. Setup Frontend

   cd traffic-monitoring/frontend/web/signalx-web
   npm install
   

3. Create a .env file in the root of the frontend directory with:

   VITE_MAPPLS_SDK_KEY=your_mapmy_india_sdk_key
   

4. Start the development server:

   npm run dev
   

5. Setup Backend - Create and activate a virtual environment:

   python -m venv venv
   venv\Scripts\activate  # Windows
   source venv/bin/activate  # Linux/Mac
   

6. Install the required packages:

   pip install -r requirements.txt
   

7. Start the Flask server:

   python backend/flask_api.py
   

8. For running the Expo project on mobile ``