👁️ AI Smart Vision
AI Smart Vision is a wearable assistive system designed to help visually impaired users understand their surroundings through AI-based visual processing, distance sensing, voice interaction, and audio feedback.
The system uses an NVIDIA Jetson Orin Nano (8GB) as the central processing unit. A 120-degree wide-angle camera and HC-SR04 distance sensor provide environmental information, while a MEMS microphone enables audio input. The processed information is delivered to the user through Bluetooth earbuds.
🎯 Objective
The objective of AI Smart Vision is to develop a wearable AI-assisted system that can process information from the user's surroundings and communicate useful information through audio.
The system combines:
Computer vision Object detection Text recognition Distance sensing Voice processing AI-based decision making Audio feedback ✨ Key Features 👁️ Object Detection
The system uses YOLOv8 for object detection.
The camera captures the surrounding environment and sends the visual information to the Jetson Orin Nano for AI processing.
🔤 Text Recognition
PaddleOCR is used for text recognition from the camera input.
This allows the system to process visible text and convert it into information that can be communicated to the user.
📏 Distance Detection
An HC-SR04 distance sensor provides distance information about nearby objects.
🎙️ Voice Input
A MEMS microphone provides audio input to the system for voice processing.
🧠 AI Decision Engine
The Jetson Orin Nano contains an AI decision engine that combines information from the different input systems and determines the appropriate response.
🔊 Voice Output
Processed information is communicated to the user through Bluetooth earbuds.
🏗️ System Architecture
The core system architecture is:
120° Wide-Angle Camera
│
▼
┌───────────────┐
│ │
│ NVIDIA Jetson │
│ Orin Nano 8GB │
│ │
│ • YOLOv8 │
│ • PaddleOCR │
│ • Voice │
│ Processing │
│ • AI Decision │
│ Engine │
│ │
└───────┬───────┘
│
▼
Bluetooth Earbuds
Audio Out
HC-SR04 ────────────┐
│
MEMS Microphone ─────┤
▼
Jetson Orin Nano
The uploaded block diagram shows the camera and HC-SR04 distance sensor connected to the Jetson Orin Nano, with the MEMS microphone providing AudioIn and Bluetooth earbuds providing AudioOut.
🔋 Power Management System
The wearable system includes a dedicated power management section.
The block diagram specifies:
Rechargeable Li-Ion battery pack Battery Management System DC-DC buck converter Power switch
The power path is:
Rechargeable Li-Ion Battery Pack │ ▼ Battery Management System │ ▼ DC-DC Buck Converter (12V → 5V) │ ▼ Power Switch │ ▼ NVIDIA Jetson Orin Nano
The block diagram specifies the DC-DC buck converter as providing a stable 12V-to-5V output.
🔧 Hardware Components Component Purpose NVIDIA Jetson Orin Nano 8GB Main AI processing unit 120° Wide-Angle Camera Captures the surrounding visual environment HC-SR04 Distance Sensor Provides distance information MEMS Microphone Audio input Bluetooth Earbuds Audio output Rechargeable Li-Ion Battery Pack Portable power source Battery Management System Battery/power management DC-DC Buck Converter Provides regulated power Power Switch Controls system power
These components and their relationships are represented in the uploaded system block diagram.
🧠 AI Processing
The Jetson Orin Nano acts as the central processing platform.
The block diagram identifies four main processing functions:
YOLOv8
Used for object detection from camera input.
PaddleOCR
Used for text recognition from visual input.
Voice Processing
Processes audio input from the MEMS microphone.
AI Decision Engine
Combines the processed information to determine the system's response.
🔄 System Workflow Environment │ ▼ 120° Wide-Angle Camera │ ▼ YOLOv8 Object Detection │ ├──────────────┐ │ │ ▼ ▼ PaddleOCR AI Processing │ │ └──────┬───────┘ │ ▼ AI Decision Engine │ ▼ Voice Processing │ ▼ Bluetooth Earbuds │ ▼ User
Distance information from the HC-SR04 and voice input from the MEMS microphone are also provided to the Jetson Orin Nano.
🖼️ Block Diagram
The complete hardware and processing architecture is shown below.
🛠️ Technology Stack AI / Machine Learning YOLOv8 PaddleOCR AI Decision Engine Hardware Computing NVIDIA Jetson Orin Nano 8GB Sensors & Input 120° Wide-Angle Camera HC-SR04 Distance Sensor MEMS Microphone Output Bluetooth Earbuds Power Rechargeable Li-Ion Battery Pack Battery Management System DC-DC Buck Converter Power Switch 📂 Suggested Repository Structure ai-smart-vision/ │ ├── README.md │ ├── hardware/ │ ├── block_diagram.png │ └── hardware_details.md │ ├── ai/ │ ├── yolo/ │ ├── paddleocr/ │ └── decision_engine/ │ ├── audio/ │ ├── input/ │ └── output/ │ ├── sensors/ │ └── hc_sr04/ │ ├── power/ │ └── power_management.md │ ├── software/ │ └── ... │ ├── screenshots/ │ └── ... │ └── LICENSE
Adjust this structure to match your actual project files before publishing.
🚀 Getting Started
The exact installation procedure depends on the software implementation running on the Jetson Orin Nano.
A typical development workflow is:
- Set up NVIDIA Jetson Orin Nano ↓
- Connect camera and sensors ↓
- Configure AI environment ↓
- Install YOLOv8 ↓
- Install PaddleOCR ↓
- Configure voice processing ↓
- Configure AI decision engine ↓
- Connect Bluetooth audio output ↓
- Test the complete system 📸 Project Images
Add actual project photographs and demonstrations to the repository.
Recommended structure:
screenshots/ ├── prototype.jpg ├── camera.jpg ├── jetson.jpg ├── wearable.jpg ├── object-detection.jpg └── text-recognition.jpg
Then display them in the README:
🔮 Future Development
Possible future development areas include:
Custom YOLOv8 model training Improved object recognition Improved text recognition Better distance awareness Natural voice interaction More intelligent AI decision making Improved wearable hardware design Lower power consumption Smaller hardware enclosure Additional environmental understanding 📊 Project Status
Status: Completed / Prototype Development
AI Smart Vision combines AI, computer vision, embedded computing, sensors, audio processing, and wearable hardware into a single assistive technology platform.
👨💻 Developer
Tamilselvan
Software Developer
⭐ Project
If you find AI Smart Vision interesting, consider giving the repository a ⭐ on GitHub.