| Popis: |
U modernom svijetu gdje smo okruženi digitalnim multimedijalnim sadržajima na dnevnoj bazi bitno je naći načine na koje se taj sadržaj može zaštititi. Kako bi ostvarili tu svrhu možemo koristiti razne metode i tehnike, a u ovom radu koncentriramo se na metode u frekvencijskoj domeni koje koriste Diskretnu Kosinusovu Transformaciju (DCT). Izdvojena metoda koju analiziramo prebacuje orignalni audiosignal u DCT domenu te grupira koeficijente u odvojene segmente kojima se svakom posebno računa energija. Vodeni žig koji je generiran iz slike se zatim umeće u najistaknutije vrhove. Izvlačenje se izvodi inverznim postupkom. Eksperimentalni rezultati pokazuju da je metoda izrazito otporna na razne namjerne i nenamjerne napade, a da je vodeni žig neprimjetan. In the modern world where we are surrounded with digital media on a daily basis it is important to find ways to protect that content. It is possible to achieve that purpose using all kinds of different methods and algorthms. In these chapters we analize methods that use Discrete Cosinuse Transform (DCT) and happen in the frequency domain. In our proposed method the original audio is transformed into DCT domain and the DCT coefficients are divided into a fixed number of subsections. Then the energy of each subsection is calculated. Next, watermark is generated from image by image processing algorithm. Watermarks are then embedded into selected peaks of highest energy subsection. They can be extracted by performing the inverse operation of watermark embedding process. Experimental results show us that the watermark is inaudible and this algorithm is robust to various attacks. |
