Project Duration: [JAN/2023 - MAY/2023]
Project Description: Developed a cutting-edge image encryption and decryption system that ensured secure transmission and retrieval of sensitive images. This project involved a multi-step process integrating deep learning, error-correcting codes, and cryptographic techniques.
Key Achievements:
-
Image Normalization with Convolutional Neural Networks (CNNs):
- Utilized Convolutional Neural Networks to preprocess and normalize the Iris image dataset.
- Trained a custom CNN model to extract feature vectors from the images.
-
Error-Correcting Code Integration:
- Integrated RS (Reed-Solomon) error-correcting codes for feature vector encoding and decoding.
- Implemented BCH (Bose-Chaudhuri-Hocquenghem) codes for secure data transmission.
-
Image Encryption and Decryption:
- Employed the extracted feature vectors as encryption keys.
- Performed XOR operations on pixel matrices to obtain encrypted images.
- Successfully decrypted images using the same key on the recipient's end.
-
NIST Randomness Tests:
- Conducted the NIST (National Institute of Standards and Technology) randomness tests to validate the cryptographic strength of the generated keys.
-
Colorful Decryption: (Optional, if you implemented it)
- Enhanced the system to enable colorful decryption of images.
Tools and Technologies:
- Python, TensorFlow, Keras
- Image processing libraries (PIL, OpenCV)
- Error-correcting codes (RS and/or BCH)
- NIST Randomness Test Suite
Key Skills Demonstrated:
- Deep Learning and Convolutional Neural Networks
- Image Processing and Cryptography
- Error-Correcting Codes
- NIST Randomness Testing
- Data Security and Encryption
This project showcased my ability to design and implement a secure image encryption system, combining deep learning and cryptographic techniques. It demonstrated my proficiency in managing large datasets, implementing complex algorithms, and ensuring data security.
Feel free to reach out to discuss this project further or for more details.