Çankaya GCRIS Standart veritabanının içerik oluşturulması ve kurulumu Research Ecosystems (https://www.researchecosystems.com) tarafından devam etmektedir. Bu süreçte gördüğünüz verilerde eksikler olabilir.

Browsing by Author "Maraha, Heyam Essam Jibrael"

Filter results by typing the first few letters
Now showing 1 - 1 of 1
  • Thumbnail Image
    Master Thesis
    Audio watermarking using DWT of second level decomposition of low frequency and using RMS on approximation coefficients
    (Çankaya Üniversitesi, 2014) Maraha, Heyam Essam Jibrael
    Digital media are distributed through the internet system in a magnificent manner due to the efficiency and inexpensiveness of such a system. Audio clips and other digital signals are transmitted, shared and used easily. This causes a real security problem. Solving this problem leads to finding authentication techniques to provide security for these digital media. Digital watermarking is a suggested solution technique to offer such security for these distributed digital signals. Digital watermarking is the art of embedding information (a watermark) in the original signal, taking into consideration preserving the quality of the host digital signal. There are different types of watermarking media such as image, video and audio; our thesis focuses on audio watermarking techniques. Also, there are a variety of techniques using the watermarking algorithm that depends on the type of algorithm that is used to embed the watermark in the original signal. In our thesis, we have used DWT (Discrete Wavelet Transform) of second level decomposition of the approximation coefficients of each frame. These frames are resulted from segmenting the original audio clips into frames, each one equal to 10 seconds. Also, we have applied the RMS (Root Mean Square) to these approximation coefficients which represent the low frequency of each frame. The aim of using the RMS is to find the higher magnitude approximation coefficients among the segmented frames; then to embed the watermark in a quarter of the selected frames according to the magnitude order. A watermark of 256 x 256 pixels was embedded in two tested audio clips (Pop and Classic) according to the former algorithm. We have applied 17 attacks that try to remove or modify the watermark that is hidden in the original audio clips. Also variety of models such as subjective and objective models were applied to evaluate the efficiency of the proposed algorithm results. The results of the proposed watermarking technique show an effective robustness against most of the applied attacks, especially in MP3 Compression, Quantization, Additive Noise, Resample and Low Pass Filter; however, it shows fragile robustness in Time Stretch and Pitch Shift attacks