Remote Sensing Image Zero Watermarking Algorithm Based on DFT

Abstract

This paper proposes a remote sensing image zero-watermark algorithm based on DFT, which can be used to protect the rights and interests of digital image copyright owners. First, the high frequency part after DCT transformation is binarized to obtain the characteristic image. Second, block DFT processing the host image. The high frequency part after DFT and the characteristic image are XORed to complete the watermark embedding. In order to test the performance of the watermark, an attack was carried out on the image embedded with the watermark. In order to extract the watermark from the attacked image, the high-frequency part of the data extracted from the attacked image is XORed again with the result of the first XOR. Compare the extracted watermark with the embedded watermark, and calculate its NC value. Most of the NC values are close to 1, indicating that the algorithm is robust. In addition, the PSNR values of the attacked image and the original image are also calculated. The experimental results show that the watermark obtained by the algorithm in this paper has strong invisibility and robustness.