Spaces:
Running
Running
File size: 8,200 Bytes
1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 1691ca8 67e2508 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 |
---
title: TextLens - AI-Powered OCR
emoji: π
colorFrom: blue
colorTo: purple
sdk: gradio
sdk_version: 4.0.0
app_file: app.py
pinned: false
license: mit
---
# π TextLens - AI-Powered OCR
[](https://huggingface.co/spaces/GoConqurer/textlens-ocr)
[](https://github.com/KumarAmrit30/textlens-ocr)
[](https://www.python.org/downloads/)
A state-of-the-art Vision-Language Model (VLM) based OCR application that extracts text from images using Microsoft Florence-2 with intelligent fallback systems and enterprise-grade zero downtime deployment.
## π Live Demo
**π Try it now:** [https://huggingface.co/spaces/GoConqurer/textlens-ocr](https://huggingface.co/spaces/GoConqurer/textlens-ocr)

## β¨ Key Features
### π€ Advanced AI-Powered OCR
- **Microsoft Florence-2 VLM**: State-of-the-art vision-language model for text extraction
- **Intelligent Fallback System**: Automatic fallback to EasyOCR if primary model fails
- **Multi-Model Support**: Florence-2-base and Florence-2-large variants
- **Real-time Processing**: Instant text extraction on image upload
### π¨ Modern User Experience
- **Clean UI**: Professional Gradio interface with intuitive design
- **Multiple Input Methods**: Upload files, use webcam, or paste from clipboard
- **Copy-to-Clipboard**: One-click text copying functionality
- **Responsive Design**: Works seamlessly on desktop and mobile devices
- **Dark/Light Theme**: Automatic theme adaptation
### β‘ Performance & Reliability
- **GPU Acceleration**: Supports CUDA, MPS (Apple Silicon), and CPU inference
- **Smart Device Detection**: Automatically uses best available hardware
- **Error Resilience**: Robust error handling with graceful degradation
- **Memory Optimization**: Efficient model loading and cleanup
### π‘οΈ Enterprise Features
- **Zero Downtime Deployment**: Blue-green deployment with health checks
- **Health Monitoring**: Built-in `/health` and `/ready` endpoints
- **Graceful Shutdown**: Signal handling for clean application restarts
- **Production Ready**: Scalable architecture with automated deployment
## π Quick Start
### π Online (Recommended)
**Instant access** - No installation required:
π [**Launch TextLens**](https://huggingface.co/spaces/GoConqurer/textlens-ocr)
### π» Local Development
1. **Clone Repository**
```bash
git clone https://github.com/KumarAmrit30/textlens-ocr.git
cd textlens-ocr
```
2. **Setup Environment**
```bash
python -m venv textlens_env
source textlens_env/bin/activate # Windows: textlens_env\Scripts\activate
pip install -r requirements.txt
```
3. **Launch Application**
```bash
python app.py
```
π Open: `http://localhost:7860`
### π§ͺ Quick Test
```bash
# Verify installation
python -c "from models.ocr_processor import OCRProcessor; print('β
TextLens ready!')"
```
## π Model Performance
| Model | Size | Speed | Accuracy | Best For |
| -------------------- | ----- | --------- | ------------ | ---------------------- |
| **Florence-2-base** | 270M | β‘ Fast | π High | General OCR, Real-time |
| **Florence-2-large** | 770M | π Medium | π Very High | High accuracy needs |
| **EasyOCR** | ~100M | π Medium | π Good | Fallback, Multilingual |
## π― Supported Use Cases
| Category | Examples | Performance |
| ------------------- | ------------------------------- | ----------- |
| π **Documents** | PDFs, Scanned papers, Forms | βββββ |
| π§Ύ **Receipts** | Shopping receipts, Invoices | ββββ |
| π± **Screenshots** | App interfaces, Error messages | βββββ |
| π **Vehicle** | License plates, VIN numbers | ββββ |
| π **Books** | Printed text, Handwritten notes | ββββ |
| π **Multilingual** | Multiple languages | βββ |
## π§ Configuration
### ποΈ Model Selection
```python
from models.ocr_processor import OCRProcessor
# Fast inference (recommended)
ocr = OCRProcessor(model_name="microsoft/Florence-2-base")
# Maximum accuracy
ocr = OCRProcessor(model_name="microsoft/Florence-2-large")
```
### π¨ UI Customization
Modify `ui/styles.py` to customize appearance:
```python
# Change color scheme
PRIMARY_COLOR = "#1f77b4"
SECONDARY_COLOR = "#ff7f0e"
# Update layout
INTERFACE_WIDTH = "100%"
```
### βοΈ Environment Variables
| Variable | Description | Default |
| ---------------------- | -------------------- | ---------------------- |
| `SPACE_ID` | HuggingFace Space ID | Auto-detected |
| `DEPLOYMENT_STAGE` | deployment stage | `production` |
| `TRANSFORMERS_CACHE` | Model cache path | `~/.cache/huggingface` |
| `CUDA_VISIBLE_DEVICES` | GPU selection | All available |
**Deployment Flow:**
```mermaid
graph LR
A[Code Push] --> B[Validate]
B --> C[Deploy Staging]
C --> D[Health Check]
D --> E[Deploy Production]
E --> F[Verify]
F --> G[Complete β
]
```
## π€ Contributing
We welcome contributions! Here's how to get started:
### π§ Development Setup
1. **Fork & Clone**
```bash
git clone https://github.com/YOUR_USERNAME/textlens-ocr.git
cd textlens-ocr
```
2. **Create Branch**
```bash
git checkout -b feature/your-feature-name
```
3. **Make Changes**
- Add new features or fix bugs
- Update tests and documentation
- Follow code style guidelines
4. **Test Changes**
```bash
python -m pytest tests/
python -c "from models.ocr_processor import OCRProcessor; OCRProcessor()"
```
5. **Submit PR**
```bash
git add .
git commit -m "feat: add your feature description"
git push origin feature/your-feature-name
```
### π Contribution Guidelines
- **Code Style**: Follow PEP 8, use Black formatter
- **Documentation**: Update README and docstrings
- **Tests**: Add tests for new functionality
- **Commits**: Use conventional commit messages
- **Issues**: Link PRs to relevant issues
## π License
This project is licensed under the **MIT License** - see the [LICENSE](LICENSE) file for details.
### π Third-Party Licenses
- **Microsoft Florence-2**: [MIT License](https://github.com/microsoft/Florence)
- **HuggingFace Transformers**: [Apache License 2.0](https://github.com/huggingface/transformers)
- **Gradio**: [Apache License 2.0](https://github.com/gradio-app/gradio)
- **EasyOCR**: [Apache License 2.0](https://github.com/JaidedAI/EasyOCR)
## π Acknowledgments
Special thanks to:
- **Microsoft Research** for the incredible Florence-2 vision-language model
- **HuggingFace** for the transformers library and Spaces platform
- **Gradio Team** for the amazing web interface framework
- **JaidedAI** for EasyOCR fallback capabilities
- **Open Source Community** for continuous support and contributions
## π Project Status
| Component | Status | Version |
| ----------------- | ------------- | ------- |
| **Core OCR** | β
Stable | v1.0.0 |
| **Web UI** | β
Stable | v1.0.0 |
| **Deployment** | β
Production | v1.0.0 |
| **API** | β
Stable | v1.0.0 |
| **Documentation** | β
Complete | v1.0.0 |
### π Stats



---
<div align="center">
**Made with β€οΈ for the AI community**
[β Star this repo](https://github.com/KumarAmrit30/textlens-ocr) β’ [π Try the demo](https://huggingface.co/spaces/GoConqurer/textlens-ocr) β’ [π Read docs](DEPLOYMENT.md)
</div>
|