Work

Cryptanalysis and Improvement of a Medical Image-Encryption Algorithm Based on 2D Logistic-Gaussian Hyperchaotic Map

Year: 2025

Type: article

Abstract: The dynamic confrontation between medical image-encryption technology and cryptanalysis enhances the security of sensitive healthcare information. Recently, Lai et al. proposed a color medical image-encryption scheme (LG-IES) based on a 2D Logistic-Gaussian hyperchaotic map (Applied Mathematics and Computation, 2023). This paper identifies that the LG-IES suffers from vulnerabilities stemming from the existence of equivalent keys and the linear solvability of the diffusion equation, enabling successful attacks through crafted chosen-plaintext attacks and known-plaintext attacks. For an M×N image, a system of linear equations with rank r can be constructed, resulting in a reduction of the key space from 232×M×N to 232×(M×N−r). To address these security flaws, the improved ILG-IES integrates the SHA-3 Edge-Pixel Filling Algorithm (SHA-3-EPFA), which includes plaintext-related SHA-3 hashing for parameter generation, a chaos-driven 3 × 3 × 3 Unit Rubik’s Cube rotation to achieve cross-channel fusion, and edge-pixel filling rules for diffusion encryption. ILG-IES outperforms LG-IES in attack resistance (resists CPA/KPA/differential attacks) while maintaining comparable security indicators (e.g., NPCR 99.6%, UACI 33.5%) to reference schemes. In future work, SHA-3-EPFA can be embedded as an independent module into most permutation-diffusion-based image-encryption systems, offering new perspectives for securing sensitive color images.

Source: Electronics

DOI: 10.3390/electronics14214283

Authors: Wanqing Wu, Shiyu Wang

Institution: Hebei University

Language: English

Topic: Chaos-based Image/Signal Encryption

Subfield: Computer Vision and Pattern Recognition

Field: Computer Science

Domain: Physical Sciences

SDG: Good health and well-being

Open Access status: gold

Estimated APC paid: $2,896