Optimization of APVO-RDH data hiding algorithm combined with optical image feature extraction for secure image transmission

Abstract

To enhance the security and concealment of image transmission, this paper proposes an optimized Adaptive Pixel Value Ordering reversible data hiding (APVO-RDH) algorithm integrated with optical image feature extraction. Through optical feature analysis including Sobel edge detection, Canny edge detection, and Harris corner detection, the carrier image is adaptively divided into blocks. Secret information is preferentially embedded in texture-smooth regions to reduce distortion and improve hiding capacity. Experimental results demonstrate that the proposed algorithm outperforms traditional Difference Expansion (DE)-RDH and Histogram Shifting (HS)-RDH algorithms in terms of Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index Measure (SSIM), and operational efficiency. The scheme also exhibits good robustness against common transmission interference and can be applied in medical imaging, remote sensing and other secure transmission scenarios.