A novel robust and more secure blind image watermarking for optical remote sensing using DWT-SVD and chaotic maps

With the growth in the internet and digital mediums, watermarking plays a major role to provideownership rights for optical remote sensing technologies, shared data, confidential optical information transition, and secure communication. Therefore, the proposed method provides high-security services in communication technology, which is the demand of today's world. In this paper, DWT and SVD transformations have been used in a new way to embed the watermark in a cover image. Moreover, encryption is used as a way of scrambling data, in order that only authorized users can access information. Therefore, using chaotic sequences and combining the watermark's characteristics, a novel transformation is suggested in which, the resultant watermarked image is changed randomly by Arnold scrambling and Logistic maps to improve the robustness, imperceptibility, and security at the same time. The main distinguishing factor of this work compared to the existing methods is that the maps are applied to the watermarked image, not the watermark, and with no need for the host image to extract the watermark. This procedure increases the security of the approach and solves one of the most important challenges of data watermarking. Experimental results demonstrate the high performance of the proposed method, e.g., resulted PSNRs after applying filtering, rotation, and noise attacks are greater than 70, 50, and 53 dB, respectively. Additionally, the proposed algorithm extracts the watermark with NC near 1 after these attacks. Moreover, the SSIM value for watermarked image and the encrypted one is 0.0109, which implies the low similarity between these two images and supports a high security. These results indicate the superior performance of the proposed method compared to the existing algorithms. All the results are performed on a system with a core i5 processor (2.50 GHz) and 2 GB RAM using MATLAB 2022b software.