Implementasi dan Analisis Video Watermarking Menggunakan Discrete Cosine Transform dan Logika Fuzzy Berdasarkan Human Visual System

Abstract

Nowadays, usage of digital data format has become more popular, because digital data is easier to distribute. Unfortunately, it makes easier to pirate it. Watermarking is a technique of information hiding in a specific media. This research is to implement and analyze video watermarking using Discrete Cosine Transform and fuzzy logic based on Human Visual System. Discrete cosine transform is a compression technique used to transform digital image and digital signal. Discrete cosine transform separates each data byte into two groups, namely high-frequency (DC) coefficient and low frequency (AC) coefficients. DC coefficient is used to insert a watermark. Fuzzy logic is an appropriate way to map an input space into an output space. Human Visual System is a method of image compression techniques, which is based on human visual characteristics. Human Visual System is used to calculate the sensitivity of the luminance and texture sensitivity. Fuzzy logic is used to determine the amount of insertion on each DC coefficient. The result of the experiment showed that the execution time for insertion is longer than the execution time for extraction. Moreover, the watermarked video has a Mean Structural Similarity of 0.934. This is a relatively good result. The whole watermarked message is always successfully extracted with 100% similarity. Also, the test result confirmed that the watermark video is invulnerable to cutting and adding attack. It is proven by the undamaged watermark message.