A Robust Image Watermarking Using Modified-Turtle Scheme with High Capacity
Lailatun Adzimah, Gelar Budiman, Sofia Saidah · 2024
In the digital age, multimedia plays an important role in disseminating information worldwide. Multimedia content such as audio, video, and images can be easily accessed and shared, but it is also at high risk of misuse and copyright infringement. This research proposes a modified turtle shell-based watermarking scheme to improve embed capacity and attack resistance. In this scheme, the Least Significant Bit (LSB) technique is used to embed the watermark. We evaluate the embedding quality and capacity using Peak Signal-to-Noise Ratio (PSNR) value which reaches 51.26 dB, payload of 0.122, as well as Normalized Correlation (NC) and entropy values to measure watermark robustness. The attacks tested include Gaussian Noise, Salt & Pepper, Speckle Noise, Compression, Low Pass Filter, and Median Filter. The test results show that the watermark has the highest robustness against Salt & Pepper attack (NC = 0.80, entropy = 6.67) and is more vulnerable against Median Filter attack (NC = 0.52, entropy = 7.34). Overall, the scheme shows good robustness against various attacks with preserved visual quality, although there is vulnerability to some specific types of attacks.