Spaces:

jeyanthangj2004
/

ocr

Runtime error

# Install Tesseract OCR
# Download from: https://github.com/UB-Mannheim/tesseract/wiki
# Add to System PATH

# Install Poppler
# Download from: https://github.com/oschwartz10612/poppler-windows/releases
# Extract and add bin/ to System PATH

Linux (Ubuntu/Debian):

sudo apt-get update
sudo apt-get install tesseract-ocr poppler-utils

macOS:

brew install tesseract poppler

🚀 Installation

Step 1: Clone Repository

git clone https://github.com/javvi51/edocr2.git
cd edocr2

Step 2: Create Virtual Environment

# Windows
python -m venv venv
venv\Scripts\activate

# Linux/Mac
python3 -m venv venv
source venv/bin/activate

Step 3: Install Python Dependencies

pip install -r requirements_webapp.txt

Step 4: Download Pre-trained Models

Download the model files from the GitHub Releases:

recognizer_gdts.keras (67.2 MB)
recognizer_gdts.txt
recognizer_dimensions_2.keras (67.2 MB)
recognizer_dimensions_2.txt

Place them in: edocr2/models/

Quick Download (Linux/Mac):

mkdir -p edocr2/models
cd edocr2/models

wget https://github.com/javvi51/edocr2/releases/download/v1.0.0/recognizer_gdts.keras
wget https://github.com/javvi51/edocr2/releases/download/v1.0.0/recognizer_gdts.txt
wget https://github.com/javvi51/edocr2/releases/download/v1.0.0/recognizer_dimensions_2.keras
wget https://github.com/javvi51/edocr2/releases/download/v1.0.0/recognizer_dimensions_2.txt

cd ../..

Quick Download (Windows PowerShell):

New-Item -ItemType Directory -Force -Path edocr2\models
cd edocr2\models

Invoke-WebRequest -Uri "https://github.com/javvi51/edocr2/releases/download/v1.0.0/recognizer_gdts.keras" -OutFile "recognizer_gdts.keras"
Invoke-WebRequest -Uri "https://github.com/javvi51/edocr2/releases/download/v1.0.0/recognizer_gdts.txt" -OutFile "recognizer_gdts.txt"
Invoke-WebRequest -Uri "https://github.com/javvi51/edocr2/releases/download/v1.0.0/recognizer_dimensions_2.keras" -OutFile "recognizer_dimensions_2.keras"
Invoke-WebRequest -Uri "https://github.com/javvi51/edocr2/releases/download/v1.0.0/recognizer_dimensions_2.txt" -OutFile "recognizer_dimensions_2.txt"

cd ..\..

🎮 Usage

Start the Server

python app.py

You should see:

🔧 Loading OCR models...
✅ Models loaded in X.XX seconds

✅ Server ready!
📱 Open your browser and go to: http://localhost:5000

Using the Web Interface

Open Browser - Navigate to http://localhost:5000
Upload Drawing - Drag & drop or click to browse
Wait for Processing - Takes 10-30 seconds
View Results - See annotated drawing and extracted data
Download - Get all results as ZIP file

Supported File Formats

✅ JPG/JPEG - Engineering drawing images
✅ PNG - Engineering drawing images
✅ PDF - Engineering drawing PDFs (first page only)

Maximum file size: 50 MB

📊 What Gets Extracted

The application extracts:

Tables - Title blocks, revision tables, bill of materials
GD&T Symbols - Geometric dimensioning and tolerancing
Dimensions - Measurements with tolerances
Other Info - Additional text and annotations

📁 Output Files

Results are saved in the results/ folder:

results/
└── 20231218_123456_drawing_name/
    ├── drawing_name_mask.png          # Annotated visualization
    ├── drawing_name.json              # Structured data (JSON)
    └── drawing_name.csv               # Tabular data (CSV)

🐛 Troubleshooting

Models Not Loading

Error: Model files not found!

Solution:

Ensure models are in edocr2/models/
Check file names match exactly
Verify files aren't corrupted (check file sizes)

NumPy Version Error

Error: AttributeError: np.sctypes was removed

Solution:

pip uninstall numpy
pip install numpy==1.26.4

Tesseract Not Found

Error: TesseractNotFoundError

Solution:

Install Tesseract OCR
Add to System PATH
Restart terminal/IDE

PDF Processing Error

Error: PDFInfoNotInstalledError

Solution:

Install Poppler
Add poppler/bin to System PATH
Restart terminal

Port Already in Use

Error: Address already in use

Solution:

# Change port in app.py (last line):
app.run(debug=True, host='0.0.0.0', port=5001)  # Use different port

⚙️ Configuration

Edit app.py to customize:

# Maximum upload size
app.config['MAX_CONTENT_LENGTH'] = 50 * 1024 * 1024  # 50MB

# Server port
app.run(debug=True, host='0.0.0.0', port=5000)

# Processing parameters
frame_thres=0.7      # Frame detection threshold
GDT_thres=0.02       # GD&T detection threshold
cluster_thres=20     # Dimension clustering threshold
max_img_size=1048    # Maximum image size for processing

🔒 Security Notes

⚠️ This is a local development server

Not suitable for production deployment
No authentication/authorization
No HTTPS encryption
Only use on trusted local network

For production deployment, use:

Gunicorn/uWSGI
Nginx reverse proxy
HTTPS certificates
Authentication middleware

📚 API Endpoints

`POST /upload`

Upload and process a drawing file.

Request: multipart/form-data with file field

Response:

{
  "success": true,
  "filename": "drawing.jpg",
  "processing_time": 15.23,
  "stats": {
    "tables_found": 2,
    "gdt_symbols": 5,
    "dimensions": 23,
    "other_info": 8
  },
  "data": { ... },
  "mask_path": "drawing_mask.png",
  "output_dir": "20231218_123456_drawing"
}

`GET /results/<folder>/<filename>`

Retrieve a result file.

`GET /download/<folder>`

Download all results as ZIP.

`GET /health`

Health check endpoint.

📖 Resources

👨‍💻 Credits

Original eDOCr2: Javier Villena Toro
Web Application: Jeyanthan GJ
License: MIT

🤝 Contributing

Issues and pull requests are welcome!

Enjoy using eDOCr2! 🚀