الفهرس 
abstractOnly 14 pages are availabe for public view
 |  | from | 172 |  |  |
| | | from | 172 | | |
Abstract
In this thesis, several algorithms were proposed to hide information in videos which broadcasting over data networks. To increase the level of insurance, information was encrypted using RSA technology before hiding at the sender side in the video frames, which used as a cover. The embedding has been applied directly on the pixels of video frames in the spatial domain. Two methods were implemented, the first method relied on the use of the concept of parity Least Significant Bit (LSB) for each pixel in every frame in the video and the second method relied on the implementation of the XORing of Least Significant Bit and the bit next to it for each pixel in every frame in the video frames.
The proposed methods of hiding were executed in the frequency domain. The embedding has been applied in the frequency domain after applying Combined (Discrete Wavelet Transform - Discrete Cosine Transform) DWT-DCT Algorithm. First 2D-DWT for three times on each frame has been applied, and then these transformations followed by 2D-DCT for HH3 band and quantization have been applied. The hidden data has been embedded in some coefficients in the middle sub-band. After embedding de-quantization has been applied and the inverse for all the transformations has been applied in reverse order to get a frame again in the spatial domain. Stego video has been reconstructed then was located on a server and waiting for client’s request. The concerned client can extract the secret encrypted message after implementing the same procedure then decrypt it using the private key to get the secret message. The proposed system was tested using Mean Square Error (MSE) and Peak Signal to Noise Ratio (PSNR) for different noise (salt and pepper) densities, which have been added as a sort of effective attack that may change the secret message. The broadcasting video over data network depends on Real-time Transport Protocol (RTP) \ User Datagram Protocol (UDP) during the transmission, so some fields in RTP have been used as covert channels to conceal data. The disadvantage of using RTP is that it does not guarantee any reliability or in order delivery of the packets. Some packets can be lost during video broadcasting over data network and others can be delivered out of order. The challenge was how to retrieve hidden data by applying the three proposed methods that were used to retrieve these lost packets. The first method was Forward Error Correction (FEC). The second method was Interleaving. The third method was Receiver-Based Repair (Interpolation). The video stream was traced using Ethereal program. The proposed methods were tested and compared according to the retrieval percentage of the hidden data at the receiver. The results were presented that the frequency domain was more robust against the salt & pepper noise than the spatial domain, which used the same proposed embedding algorithms. PSNR reached to 73.7 dB in the frequency domain compared to 65.6 dB in the spatial domain for noise density 0.1 and Interleaving method have achieved better rates of retrieval of data hidden in comparison with the FEC method and the Interpolation method where the proportion recovery reached to 99% for 5% dropped packets.