With the rapid development of digital communication, cloud computing, and intelligent applications such as smart healthcare and the Internet of Things (IoT), secure image transmission has become a critical challenge. Traditional encryption algorithms often face limitations including narrow key spaces, insufficient sensitivity to initial parameters, and weak resistance to statistical and differential attacks, leaving encrypted images vulnerable to security breaches. To address this gap, we propose a novel image encryption algorithm that integrates a three-dimensional Chebyshev-Henon hyperchaotic map (3D-CHM), Josephus permutation, wavelet transform, and a Gaussian noise-driven diffusion mechanism. First, the 3D-CHM generates complex, unpredictable chaotic sequences with high sensitivity to initial conditions, providing a substantially larger key space. Second, a pixel substitution strategy combining Josephus permutation and chaotic sequences ensures global pixel shuffling and effectively reduces inter-pixel correlation. Third, a dynamic diffusion process incorporating Gaussian noise further enhances robustness against various attacks. Experimental results demonstrate that the proposed algorithm achieves a vast key space of approximately \(2^{256}\), strong resistance against brute-force, statistical, and differential attacks, and excellent performance metrics (Number of Pixels Change Rate (NPCR) = 99.609463%, Uniform Average Change Intensity (UACI) = 33.462821%). Comparative evaluations with state-of-the-art schemes confirm the superior security, efficiency, and robustness of our method across images of different resolutions. These results highlight the potential of the proposed scheme for practical deployment in privacy-critical scenarios such as medical image protection and IoT-based visual data security.

A blockchain and threshold fully homomorphic encryption (TFHE)-based federated learning framework is proposed to defend against model poisoning and privacy leakage. Standard federated learning (FL) remains vulnerable to inference and poisoning attacks during model aggregation and distribution. To address these threats, we present a secure and efficient FL framework that combines TFHE with Shamir’s secret sharing for privacy-preserving encrypted updates. A multi-layer verification mechanism consisting of verifiable secret sharing (PVSS) and cosine similarity checks is introduced to identify and filter out malicious updates. Furthermore, we replace the central aggregator with a blockchain-based mechanism to eliminate single-point trust. Extensive experiments conducted on MNIST and CIFAR-10 under IID and Non-IID settings demonstrate that our scheme maintains high model accuracy under 20%–40% label-flipping attacks. Compared with baseline methods such as threshold Paillier and PPVFL, our approach reduces encryption and decryption overhead by up to 28.4%, while achieving a lower accuracy drop of only 2.32% on MNIST and 7.44% on CIFAR-10 under 40% poisoning. These results highlight the robustness, efficiency, and practicality of the proposed scheme for real-world federated learning deployments.