Dynamic structure driven image scrambling technique for data protection

Area of Science:

• Computer Science
• Information Security
• Data Structures

Background:

• Digital communication necessitates robust security for sensitive image transmission.
• Images are critical in various sectors, including healthcare, military, and intellectual property.
• Existing image security methods require enhancement for efficiency and effectiveness.

Purpose of the Study:

• To propose a new image scrambling method for enhanced security during digital transmission.
• To leverage dynamic data structures like binary trees and hash tables for improved scrambling and descrambling.
• To ensure the integrity and quality of sensitive images throughout the transmission process.

Main Methods:

• Image pixels are stored in a binary tree using a hash table.
• Pixels are rearranged into a one-dimensional array via tree traversal.
• The array is converted back to a two-dimensional format matching the original image dimensions.

Main Results:

• The proposed method achieves a Peak Signal-to-Noise Ratio (PSNR) below 4 dB between the original and scrambled images.
• A PSNR of infinity is achieved between the original and descrambled images, indicating perfect reconstruction.
• The method demonstrates 100% similarity between the original and descrambled images and shows a negative correlation with scrambled images.

Conclusions:

• The proposed binary tree and hash table-based image scrambling method offers effective security for digital image transmission.
• The technique preserves image quality, evidenced by high PSNR values for descrambled images.
• Future applications include integration into image watermarking and steganography for advanced security solutions.