Implementasi Invisible Watermark sebagai Proteksi Hak Cipta pada Citra Digital Menggunakan Metode DWT dan SVD

Abstract

In the interconnected digital era, information security has become a critical necessity. One of the primary challenges in information security is the protection of visual content, particularly digital images. The massive exchange of digital images over the internet increases the risk of copyright infringement, whether through unauthorized use, illegal distribution, or unlicensed commercial exploitation. Watermarking techniques have been proposed as a solution to this issue. Watermarking enables the embedding of secret information into digital images without compromising visual quality. It operates by embedding a unique mark into the digital image, allowing for future verification of image ownership. The Discrete Wavelet Transform (DWT) and Singular Value Decomposition (SVD) methods were chosen for their effectiveness in embedding watermark into images. The primary contribution of this research is the development of PyMark Watermarking, a web-based watermarking platform utilizing DWT-SVD methods. This platform integrates with Google Drive via the Google Drive API for storing watermarking results and with Instagram via the Meta Graph API for automating content publication. This application is designed to protect the copyright of digital images, enabling effective watermark embedding without compromising visual quality, and enhancing resistance to manipulation. During testing, the research parameters PSNR, SSIM, BER, and NCC were used as objective measurement tools to assess watermarking quality. The test results demonstrated a high level of imperceptibility with an average PSNR value of 50.62 dB and SSIM of 0.988. Additionally, this method exhibited good resistance to attacks with an average BER of around 0.1 and NCC above 0.9. Furthermore, the PyMark Watermarking website platform proved capable of protecting the watermark after the watermarked image was published online to platforms like Google Drive and Instagram, even when the image had undergone manipulation by third parties.