Answers are grounded in this paper and cite their sources.
Unmanned Aerial Vehicle (UAV) images are widely used in surveillance, border security, disaster monitoring, and military applications, where secure image transfer is crucial. This paper proposes an adaptive feature-aware image encryption framework to protect UAV images against unauthorised access and statistical attacks. The framework integrates image feature analysis, BLAKE3-based key derivation, and a three-dimensional Bülban chaotic system to generate feature-dependent chaotic sequences. An Adaptive Security Controller dynamically determines the block size, diffusion rounds, transient skip, and chaotic warm-up setting according to image characteristics and derived key values. The generated chaotic streams are used for adaptive block-wise permutation and multi-round forward-backwards diffusion to enhance confusion and diffusion. Experimental evaluation achieves an average encrypted-image entropy of 7.9993, NPCR of 99.6169%, and UACI of 33.4590%. The encrypted images also exhibit negligible linear correlation between adjacent pixels and are processed using a 256-bit master key. These results demonstrate favourable statistical characteristics under the evaluated experimental conditions. The proposed framework therefore provides a feature-aware encryption approach for UAV image transmission, while further implementation-level and application-specific validation is required.
No takes yet. Share an insight, caveat, or question.
Yogi et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: