Transcript
Data Hiding Using PMM and Mod 4 Method(A New Approach Towards Multimedia Steganography)
Under the esteemed guidance of Mr. Sauvik Bhattacharya
Presented By Group-01 Of CSE 2007-11:
Arko Kundu (T 20071063) Kaushik Chakrarborty (T 20071064)Bharat Mandal (20071013) Ranjit Das (20071016) Abhishek Roy (L 20071063)
Objective
The Primary Objective is to successfully hide a secret message in a video without distorting the quality of the resultant video.Things to look out for ?
Video Format Size of the data to be hiddenLength and size of the videoEmbedding in video can be of the following types:
Embedding in the Image Sequence.(Image Steganography)Embedding in the Audio.(Audio Steganography)Embedding in both Image Sequence and the Audio. (Image Steganography + Audio Steganography)
MOST IMPORTANT -> DCT or DWT ?
DWT Vs DCTA big Question in our way of Video Steganography.
Which one to choose for Image Processing?
A Discrete Cosine Transform (DCT) is a mathematical transformation thattakes a signal and transforms it from spatial domain into frequency domain.It is similar to Discrete Fourier Tranform, using only real numbers forcomputation purpose.
A Discrete Wavelet Transform (DWT) is any wavelet transform for whichthe wavelets are discretely sampled. As with other wavelet transforms, akey advantage it has over Fourier transforms is temporal resolution(Precision of measurement with respect to time).
The main advantage of using a DWT is that it’s computationally very fast.
Recap ( Mod 16 Method)
The Summarized Sending Algorithm is as follows :
1. Select a seed pixel.2. Select the neighbouring pixels.3. Find the intensity of the neighbouring pixels .Find out and store the remainder after
dividing the intensity by 16. Let the remainder be R .4. Subtract R from the intensity and hence the resultant is divisible by 16.5. Now add the decimal equivalent of 4 bit binary message (0000 to 1111) to the
resultant intensity .6. The modified intensity is replaced in the original image matrix.7. At the end we get the stego matrix embedded with the message.
The Summarized Receiving Algorithm is as follows :
1. Select a seed pixel.2. Select the neighbouring pixels.3. Find the intensity of the neighbouring pixels and find out and store the remainder
after dividing it by 16. Let the remainder be R .4. Now R represents the Decimal Value of the embedded message . When converted to
4-bit binary format we get our resultant message at the receiver’s end.
Methodologies(Sending side) Cover Video
Audio
1st half of the
message
Image Frame ‘1’
Split the Video in Image Frames and Audio
Image Frame ‘2’
Image Frame 3Image Frame 3Image Frame n-1
Image Frame ‘n’
2nd half of the
message
Join the Image Frames and Audio to create the Video
Stego Video
Embedding
Embedding
Methodologies(Receiving side) Stego Video
Audio
1st half of the
message
Image Frame ‘1’
Split the Video in Image Frames and Audio
Image Frame ‘2’
Image Frame 3Image Frame 3Image Frame n-1
Image Frame ‘n’
2nd half of the
message
Join the two half’s to create the entire message
Hidden Message
Extracting
Extracting
Methodologies(PMM Sending) The basic technique of mapping in data in stego image is based on the following table :
Methodologies(Seed Pixel Selection)
Seed Pixel Generation for embedding message bits is dependent on the intensity value of the previous pixel selected.
It includes a decision factor (dp) which is dependent on intensity with a fixed way of calculating the next pixel.
x=x1+2+dpy=y1+2+dp
(x,y)=new Pixel position(x1,y1)=previous Pixel position
Intensity >80 &&
<=160
Intensity >160 &&
<=255
dp=1
dp=2
dp=3
Intensity <=80
An Example… (Seed Pixel Selection)
122 45 69 132 256 145 56 79 112 125
156 125 169 123 79 78 12 186 123 69
224 212 145 125 147 86 45 110 236 26
119 248 46 112 48 23 79 45 90 156
119 79 116 189 53 30 130 90 141 171
56 71 26 83 43 75 93 67 116 49
90 112 179 212 201 38 99 119 157 39
83 53 89 115 63 78 90 76 255 255
131 141 176 159 126 146 255 73 86 255
An Example ( Neighboring Pixel Selection )
112 48 23
189 53 30
83 43 75
i+1,j i+1,j-1 i+1,j+1
i,j-1 i,j+1
i-1,j-1 i-1,j+1
i-1,j
Results and DiscussionsT he PSNR , MSE and Coorelation values obtained by applying the Mod 16 Method is shown as follows:
Results and DiscussionsAnalysis of the Video Steganography Technique :
Video Name : baby.aviTotal Message : “University Institute of Technology”
Message in the Video part : “University<Instit>>>”The data is inserted in the frame no. 1, 3, 10 and 11. Their respective PSNR andMSE values are shown below :
Message in the Audio part : “ ute<of<Technology>>>”
PSNR : 140.3128; MSE : 6.0507e-010
Frame No. PSNR MSE
1 61.6923 0.0440
3 61.4383 0.0467
6 61.4190 0.0469
10 34.2765 24.2902
11 34.3523 23.8698
Results and DiscussionsImage Frame Comparison
Frame no. 10 after Embedding
Frame no. 11 after Embedding
Future Scope
In today’s world, we often listen a popular term “Hacking”. So, whatever be the technique will be developed in
future, degree of security related with that has to be kept in mind. It is hoped that Dual Steganography, “Steganographyalong with Cryptography “ may be used to achieve some of the future solution for this above mentioned problem.
In the near future, the most important use of steganographic techniques will probably be lying in the field of “Digital Watermarking”.
Can be used to enhance a country’s security system regarding confidential documents.
Areas of Improvement
Enabling Password Protection.
Flexible Algorithm to adapt to all types of Video formats.
Embedding capacity can be enhanced.
Size of the stego video has to be reduced drastically.
Conclusion
This proposed method presents a Video based steganographicmodel which utilizes cover vide files in the wavelet domain to conceal the presence of other sensitive data in any format.
This approach can also be utilized to work on Grey Scale Videos also.
This method can be extended to map n number of bit also by considering more number of features of the embedding pixels.
This method is also capable of extracting the secret message without the cover video.
top related