CAPTCHA Recognition Using Deep Learning is a powerful project designed to solve complex CAPTCHA challenges with advanced AI techniques. It combines a custom-trained YOLO model for character detection and a neural network for recognition, achieving high accuracy in diverse and distorted CAPTCHA formats. The project also includes an API for seamless integration into real-world applications.
- High Accuracy: Solves CAPTCHAs with over 90% accuracy.
- Custom YOLO Model: Detects and localizes characters in challenging images.
- Extensive Dataset: Includes a collection of difficult CAPTCHA images and a personalized dataset for training.
- API for Deployment: Offers a user-friendly API to integrate CAPTCHA-solving functionality into your projects.
- Comprehensive Codebase: Fully documented scripts for training, predictions, and API usage.
- Python 3.8+
- TensorFlow 2.x
- YOLO libraries
- Install dependencies using:
Access the API at
python makeapi.py
http://localhost:8000. (with useapi.py)
- Security Testing: Identify weaknesses in CAPTCHA implementations.
- Automation: Streamline tasks requiring CAPTCHA solving.
This project is licensed under the MIT License.
Contributions are welcome! Feel free to open issues or submit pull requests.
For questions or feedback, reach out to Taha.