Answers are grounded in this paper and cite their sources.
Binary semantic masks (BSMs) represent a type of structured discrete data that describe the spatial distribution of target and background regions. As critical outputs of on-board AI processing in remote sensing, transmitting these data securely and efficiently over constrained channels poses a significant challenge, since traditional methods struggle to balance high compression ratios (CRs) with robust data protection. To address this problem, this article proposes a framework for the secure and efficient lossless compression of BSMs, named joint encryption and lossless compression framework for (JELCF)-BSM. The method first exploits spatial redundancy by segmenting the data into homogeneous blocks (HB) and non-HB, followed by encryption via block-level scrambling and diffusion. Subsequently, a modified 2-D binary discrete cosine transform is designed to establish a one-to-one correspondence between binary blocks and transform-domain DC coefficients. Based on this unique representation, an index coding strategy is developed to achieve efficient lossless compression of encrypted semantic masks. Experimental results demonstrate that JELCF-BSM achieves an average CR approximately 82% higher than conventional lossless compression schemes when compressing encrypted BSMs, and 56% higher than compared joint encryption and compression schemes. Security evaluations further confirm that the method exhibits strong robustness against statistical and differential attacks. This method provides a feasible solution for secure and efficient remote sensing semantic data transmission.
No takes yet. Share an insight, caveat, or question.
Leng et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: