Remote sensing image zero watermark method based on coupled neural P system

Abstract

The invention discloses a remote sensing image zero watermark method based on a coupled neural P system. The method comprises the steps of coupled neural P system model construction, zero watermark construction and zero watermark extraction. The method comprises the following steps: firstly, taking a low-frequency component obtained by non-subsampled shear wave transformation as input, constructing a coupled nerve P system model to simulate neuron complex interaction and parallel calculation, introducing a multi-scale morphological gradient optimization algorithm to dynamically modulate external stimulation, and accurately extracting space-time dynamic characteristics of an image; secondly, generating an encrypted feature image through asymmetric Tent mapping, and forming a zero watermark through scrambling and XOR operation; in the extraction stage, the watermark of the to-be-detected image is obtained by using the same algorithm and is compared with the original watermark to complete copyright verification. According to the method, lossless, robust and safe remote sensing image copyright protection is realized through fusion of biological neural network characteristics and a mathematical encryption technology.

Images