Image Watermarking Menggunakan Algoritma Sobel Canny dan Algoritma Discrete Wavelet Transfrom untuk Melindungi Konten Foto Digital pada Platform Instagram

Abstract

In the digital era, cases of photo content theft and copyright infringement particularly on platforms like Instagram are becoming more frequent. To mitigate this issue, a Steganography approach using the Image Watermarking technique is employed to embed ownership or copyright labels into digital images. This study utilizes the Discrete Wavelet Transform (DWT) algorithm for the embedding process and the Sobel-Canny algorithm for the extraction process. The system is developed using the Python programming language and supports *.jpeg/jpg image formats to perform mathematical operations required for embedding and extraction.Testing was carried out on five image files with varying sizes. The results indicate that larger image sizes and longer processing times result in increased complexity and larger output file sizes. Evaluation was performed using Mean Squared Error (MSE) to measure image distortion and Peak Signal to Noise Ratio (PSNR) to assess visual quality. The best results showed an Mean Squared Error (MSE) of 0.80 and Peak Signal to Noise Ratio (PSNR) of 49.07 dB, confirming that the watermark is imperceptible to the human eye and does not significantly affect image quality.Further testing under Crop and Rotate attacks revealed that the watermark becomes unreadable, suggesting the system is not resistant to major image alterations. However, when subjected to minor modifications, the watermark remained detectable, indicating a degree of robustness against light image processing.