A Dual-Layer Cryptographic Framework for Secure Image Data Embedding Using ECC, ChaCha20, and SHA-256

Abstract

Steganography enables hidden communication by embedding secret information within digital images, yet conventional Least Significant Bit (LSB) methods lack essential security features, exposing embedded data to unauthorized access and tampering. This work presents a secure image steganography framework that integrates encryption, hashing, and compression to enhance data protection. The proposed system supports both asymmetric and symmetric encryption by utilizing Elliptic Curve Integrated Encryption Scheme (ECIES) over secp256k1 and the ChaCha20 stream cipher, respectively. Before embedding, the secret message is encrypted and a SHA-256 hash of the ciphertext is generated to enable integrity verification. The encrypted data is encoded using Base64 and converted into a binary sequence, which is embedded into the least significant bits of an RGB image. To further improve efficiency and reduce detectability, the stego image is compressed using zlib. During extraction, the compressed image is decompressed, and the embedded bits are reconstructed to retrieve the encoded ciphertext. The system then verifies data integrity by recomputing the SHA-256 hash before decrypting the message using the appropriate key. A user-friendly interface built with Tkinter enables seamless operation and performance evaluation. Experimental analysis indicates that ChaCha20 achieves faster processing times, whereas ECIES provides stronger key management advantages. The proposed approach ensures confidentiality, integrity, and low visual distortion, making it effective for secure and covert image-based communication systems.