[AI_Assistance_for_blinds] 🎯

Basic Details

Team Name: [Wildx]

Team Members

• Member 1:[Ann Rose Mathew] - [LBS Institute of Technology for Women]
• Member 2: [Ann Mariya Joju M] - [LBS Institute of Technology for Women]

Hosted Project Link

[https://github.com/annrose2277-glitch/AI_Assistance_for_blinds.git]

Project Description

[This project modernizes traditional navigational aids by integrating computer vision and sensor fusion into an intelligent assistive device. By detecting objects in real-time and converting visual data into auditory feedback via text-to-speech, we provide visually impaired users to enable greater independence and confidence in navigation.]

The Problem statement

[Visually impaired individuals face significant barriers to independent mobility and experience social anxiety due to a reliance on sighted assistance. This dependency is primarily caused by the limitations of traditional white canes, which provide only tactile, ground-level feedback and lack the capability to detect mid-air obstacles, identify specific objects, or provide the environmental context necessary for safe, confident navigation.]

The Solution

[Our prototype transforms the standard white cane into an intelligent navigation assistant. By mounting a lightweight webcam and an AI-processing unit onto the cane

Computer Vision (The Eyes): We utilize the YOLO (You Only Look Once) model. Unlike traditional models, YOLO scans the entire camera frame in one pass, making it incredibly fast for real-time detection.

The "40% Rule" (Smart Filtering): To prevent the user from being overwhelmed by too much information, the system only triggers a high-priority alert when an object’s bounding box covers 40% of the camera's view. This ensures we are reporting the most relevant, immediate obstacles.

Voice Interface (The Guide): Using the Pyttsx3 library, the system converts the AI's visual data into clear speech. Instead of a beep, the user hears: "An object is in front of you: [Object Name]."]

---

Technical Details

Technologies/Components Used

For Software:

• Languages used: [Python]
• Frameworks used: [Ultralytics YOLO model YOLO26n]
• Libraries used: [OpenCV,Pyttsx3,Threading,NumPy,Time]
• Tools used: [Pip,Webcam Hardware,Integrated Development Environment (IDE)]

For Hardware:

• Main components: [List main components]
• Specifications: [Technical specifications]
• Tools required: [List tools needed]

---

Features

List the key features of your project:

• Feature 1: [Real-Time --- Detection Instant identification of obstacles as the user walks.]
• Feature 2: [Proximity Awareness --- The 40% threshold ensures users are warned about objects directly in their path.]
• Feature 3: [Natural Language Feedback --- Uses human-like speech to describe the environment, reducing the learning curve.]
• Feature 4: [Visual Monitoring --- Provides a live feed with bounding boxes on a display for caregivers or for calibration.]

---

Implementation

For Software:

Installation

[pip install ultralytics opencv-python pyttsx3 numpy]

Run

[python main.py]

For Hardware:

Components Required

[List all components needed with specifications]

Circuit Setup

[Explain how to set up the circuit]

---

Project Documentation

For Software:

Screenshots (Add at least 3)

WhatsApp Image 2026-02-14 at 6 46 43 AM detection of a person,bottle,chair ---output showing in terminal #detection of objects

WhatsApp Image 2026-02-14 at 6 49 04 AM detection of cell phone ---output showing in terminal #detection of objects

 Add caption explaining what this shows

Diagrams

System Architecture:

Explain your system architecture - components, data flow, tech stack interaction

Application Workflow:

Add caption explaining your workflow

---

For Hardware:

Schematic & Circuit

 Add caption explaining connections

 Add caption explaining the schematic

Build Photos


 List out all components shown

 Explain the build steps

 Explain the final build

---

Additional Documentation

For Web Projects with Backend:

API Documentation

Base URL: https://api.yourproject.com

Endpoints

GET /api/endpoint

• Description: [What it does]
• Parameters:
  • param1 (string): [Description]
  • param2 (integer): [Description]
• Response:

{
  "status": "success",
  "data": {}
}

POST /api/endpoint

• Description: [What it does]
• Request Body:

{
  "field1": "value1",
  "field2": "value2"
}

• Response:

{
  "status": "success",
  "message": "Operation completed"
}

[Add more endpoints as needed...]

---

For Mobile Apps:

App Flow Diagram

Explain the user flow through your application

Installation Guide

For Android (APK):

1. Download the APK from [Release Link]
2. Enable "Install from Unknown Sources" in your device settings:
   • Go to Settings > Security
   • Enable "Unknown Sources"
3. Open the downloaded APK file
4. Follow the installation prompts
5. Open the app and enjoy!

For iOS (IPA) - TestFlight:

1. Download TestFlight from the App Store
2. Open this TestFlight link: [Your TestFlight Link]
3. Click "Install" or "Accept"
4. Wait for the app to install
5. Open the app from your home screen

Building from Source:

# For Android
flutter build apk
# or
./gradlew assembleDebug

# For iOS
flutter build ios
# or
xcodebuild -workspace App.xcworkspace -scheme App -configuration Debug

---

For Hardware Projects:

Bill of Materials (BOM)

Component	Quantity	Specifications	Price	Link/Source
Arduino Uno	1	ATmega328P, 16MHz	₹450	[Link]
LED	5	Red, 5mm, 20mA	₹5 each	[Link]
Resistor	5	220Ω, 1/4W	₹1 each	[Link]
Breadboard	1	830 points	₹100	[Link]
Jumper Wires	20	Male-to-Male	₹50	[Link]
[Add more...]

Total Estimated Cost: ₹[Amount]

Assembly Instructions

Step 1: Prepare Components

1. Gather all components listed in the BOM
2. Check component specifications
3. Prepare your workspace Caption: All components laid out

Step 2: Build the Power Supply

1. Connect the power rails on the breadboard
2. Connect Arduino 5V to breadboard positive rail
3. Connect Arduino GND to breadboard negative rail Caption: Power connections completed

Step 3: Add Components

1. Place LEDs on breadboard
2. Connect resistors in series with LEDs
3. Connect LED cathodes to GND
4. Connect LED anodes to Arduino digital pins (2-6) Caption: LED circuit assembled

Step 4: [Continue for all steps...]

Final Assembly: Caption: Completed project ready for testing

---

For Scripts/CLI Tools:

Command Reference

Basic Usage:

python script.py [options] [arguments]

Available Commands:

• command1 [args] - Description of what command1 does
• command2 [args] - Description of what command2 does
• command3 [args] - Description of what command3 does

Options:

• -h, --help - Show help message and exit
• -v, --verbose - Enable verbose output
• -o, --output FILE - Specify output file path
• -c, --config FILE - Specify configuration file
• --version - Show version information

Examples:

# Example 1: Basic usage
python script.py input.txt

# Example 2: With verbose output
python script.py -v input.txt

# Example 3: Specify output file
python script.py -o output.txt input.txt

# Example 4: Using configuration
python script.py -c config.json --verbose input.txt

Demo Output

Example 1: Basic Processing

Input:

This is a sample input file
with multiple lines of text
for demonstration purposes

Command:

python script.py sample.txt

Output:

Processing: sample.txt
Lines processed: 3
Characters counted: 86
Status: Success
Output saved to: output.txt

Example 2: Advanced Usage

Input:

{
  "name": "test",
  "value": 123
}

Command:

python script.py -v --format json data.json

Output:

[VERBOSE] Loading configuration...
[VERBOSE] Parsing JSON input...
[VERBOSE] Processing data...
{
  "status": "success",
  "processed": true,
  "result": {
    "name": "test",
    "value": 123,
    "timestamp": "2024-02-07T10:30:00"
  }
}
[VERBOSE] Operation completed in 0.23s

---

Project Demo

Video

(https://github.com/user-attachments/assets/82a02595-b917-4043-806c-6508f71014b3)

Explain what the video demonstrates - key features, user flow, technical highlights

Additional Demos

[Add any extra demo materials/links - Live site, APK download, online demo, etc.]

---

AI Tools Used (Optional - For Transparency Bonus)

If you used AI tools during development, document them here for transparency:

Tool Used: [Google Gemini]

Purpose: [What you used it for] -for getting prompt,code

Key Prompts Used: -act as an expert prompt engineer -structured prompt for gemini cli Percentage of AI-generated code: [60%]

Human Contributions: -idea and planning

Note: Proper documentation of AI usage demonstrates transparency and earns bonus points in evaluation!

---

Team Contributions

• [Name 1]: [Specific contributions - e.g., Frontend development, API integration, etc.]
• [Name 2]: [Specific contributions - e.g., Backend development, Database design, etc.]
• [Name 3]: [Specific contributions - e.g., UI/UX design, Testing, Documentation, etc.]

---

License

This project is licensed under the [LICENSE_NAME] License - see the LICENSE file for details.

Common License Options:

• MIT License (Permissive, widely used)
• Apache 2.0 (Permissive with patent grant)
• GPL v3 (Copyleft, requires derivative works to be open source)

---

Made with ❤️ at TinkerHub