Design and implementation of highly efficient digital watermarking prototype for securing copyright and authentication of satellite imagery


Satellite imagery is considered as a powerful tool to map and monitor natural and man-made resources changes globally. Moreover, it greatly contributes to boost scientific researches, supports decision making process regarding important issues and has a tremendous potential to improve economic outcomes. However, the dramatic growth in the field of communication and the internet allows satellite imagery to be easily distributed, tampered and duplicated by unauthorized users. As a result, the protection of its intellectual property has become of a great demand. To overcome this issue, the concept of “watermarking” has been introduced as a potential solution for copyright protection and ownership authentication of satellite images. The principle of watermarking here is based on embedding a pattern of invisible identification information (for instance logo) into a satellite image so that it can be detected or extracted later to identify the copyright owner. Unlike other sorts of data, satellite images are complex in nature and contain very sensitive information which can be easily degraded by inserting additional information. This means that applying watermarking technique to satellite images would result negatively on its quality, which totally interferes with the main purpose of building high resolution satellites. Thus, a proper technique should be developed to overcome this problem. Recently, digital watermarking has been a hot research topic in academia and industry, where a wide number of schemes have been proposed and evaluated. This paper deals with the design, implementation and evaluation of a novel watermarking model for preserving the copyright and authentication of DubaiSat-2 images. The proposed algorithm must be robust against intentional and non-intentional attacks such as compression, scaling, rotation, filtering, cropping, contrast enhancement etc. In addition, the algorithm should be fragile in such a way that it should be very sensitive to any minor tampering/modification and it should precisely detect tampered area on the satellite image. Also, the watermark should be imperceptible to the human eye. Five quality indices are used to assess the performance of the proposed prototype, which consist of the peak signal to noise ratio (PSNR), wavelet domain signal to noise ratio (WSNR), structure similarity index measurement (SSIM), normalized correlation (NC) and mean square error (MSE).