نظام تشفير الصور قائم على مفاهيم سلسلة الكتل (Block Chain) باستخدام خوارزميات SHA-256 وMD5

Image Cryptography System Based on Block Chain Concepts Using SHA-256 and MD5 Algorithms

Abstract

Since digital images play a crucial role in various fields such as healthcare and national security, ensuring their safety is essential. Conventional encryption methods are often insufficient for addressing issues related to large image sizes, scalability and resistance to modern cryptographic attacks. This thesis proposes an image encryption architecture that combines ideas from blockchain concepts with Secure Hash Algorithm 256 (SHA-256), Secure Hash Algorithm (SHA-512), and Message Digest (MD5) algorithms to make images more secure, faster, and tamper-resistant.
Hashing is a one-way process that transforms input data of arbitrary size into a fixed-length string, known as a hash value or digest. Hashing is primarily used for data integrity verification, digital signatures, and authentication. Encryption, by contrast, is a two-way process that converts plaintext into ciphertext using an encryption algorithm and a key. Unlike hashing, encryption is reversible: the ciphertext can be decrypted back into the original plaintext using the appropriate key. Its main purpose is to ensure confidentiality, protecting data from unauthorized access while allowing recovery by legitimate parties.
The proposed framework divides each image into blocks, which are then linked like nodes in a blockchain, to increase security and unpredictability. Each block is assigned a unique encryption key derived from SHA-256 and SHA-512 and MD5 creates dependency between blocks through sequential keys. Pixel encryption is achieved by performing XOR operations with MD5 at its core. Cryptographic metadata for keys and block sizes is securely hidden within the image file, ensuring the image can be decrypted more quickly and without any quality loss.
The system was evaluated based on nine performance indicators, including security, performance, speed, and final image quality. Experimental results show that the proposed method achieves a high entropy value of 7.607, very low correlation values of (H: -0.0004, V: -0.0002, D: -0.0006), and excellent security against various statistical and differential attacks as the number of pixels changing rate value of 99.48%. It also successfully recovers the original image after decryption. Speed plays an important role in this research, as the encryption time was 0.468 seconds and the decryption time was 0.312 seconds. This performance is superior to many previous studies and constitutes the rationale for selecting the MD5 algorithm for encryption. The system provides enhanced security and more efficient computation compared to other existing methods, making it highly promising for transmitting and storing sensitive images.