Skin tone based steganography

Skin Tone Based Image SteganographyGirish Ram M

R7121

What is Steganography??Steganography is the ART OF HIDING the secret and confidentional data in another transmission medium to achieve secret communication.

Hiding of data in Image is called Image Steganography.

Why do we need Steganography?? Internet serves as an important role for data transmission and sharing.

It is a worldwide and publicized medium, some confidential data might be stolen, copied, modified, or destroyed

Security problems become an essential issue.Encryption is a well-know procedure for secured data transmission

Why not Encryption??Encryption is the process of encoding messages (or information) into a form that cannot be easily understood by unauthorized people.

Disadv of encryption: They make the secret messages unreadable and unnatural, These unnatural messages usually attract some unintended observer’s attention.

Therefore, Digital Steganography is a new security approach for the transmission of confidential data

Encryption

Steganography

Original image

Secret data

Stegano image

Comparison b/w Steganography & WatermarkingBoth are branches of data hiding.Watermarking is the technique of embedding digital marks inside a container(image) so that there is a logical way of extracting the data embedded.

Here no harming of container.Steganography uses the cover file to deliver its secret message while watermarking considers the cover file as the important data that is to be preserved.

Steganography is used for secret communication while watermarking is used for authentication.

Fig1. Trade off b/w embedding capacity, Robustness and undetectability in Data hiding

Embedding CapacityNaive steganographic technique

Secure steganographic technique

Digital Watermarking

Undetectability Robustness

Steganography pay attention to the degree of invisibility while Watermarking pay most of its attribute to the robustness of the message and its ability to withstand attacks of removal.

Existing Methods for Image SteganographyLSB(Least Significant Bit) Substitution based.Transform domain based.Adaptive method.

Proposed MethodBiometrics based steganography using DWT(Discrete Wavelet Transform)Biometric feature used is Skin tone region of images.In this method secret data is embedded within skin region.

Skin region are not sensitive to HVS(Human Visual System).

Overview of methodFirst: Skin tone detection is performed on input image using HSV (Hue, saturation, value).

Second: Cover image is transformed in frequency domain using Haar-DWT.

Payload number calculated.Finally Secret message is embedded in one of the high frequency sub-band by tracing skin pixels in that band.

Embedding steps are applied to 2 cases:With CroppingWithout CroppingBoth cases uses different embedding

algorithm.

With CroppingFirst cropping on input image is performed and only in cropped region data hiding is performed.

Cropped region works as a key at decoding side.More Secure.

Without CroppingData is embedded in whole Skin region image.Embedding algorithm attempts to preserve histogram of DWT coefficients after embedding.

Skin Tone Detection Using HSVAim: Discriminate b/w skin & non-skin pixels.Transforms a given pixel into an appropriate color space.

Uses a skin classifier to label the pixel whether it is a skin or a non-skin pixel.

Skin detection algo produces a mask – black and white.

White pixel = 1 Skin pixel.Black pixel = 0 Non-Skin pixel.

How to decide weather a pixel is skin or not skin?RGB matrix of the given color image is converted into different color spaces to yield distinguishable regions of skin or near skin tone.

Mainly two kinds of color spaces are available HSV (Hue, Saturation and Value) and YCbCr spaces.

It is experimentally found that the distribution of human skin color constantly resides in a certain range within the color space.

Here HSV color space is chosen.

Cont’d..

RGB image is converted into HSV color space. In HSV, responsible values for skin detection are Hue & Saturation.

Extract Hue & Saturatn dimensions into separate new variables (H & S).

For Skin detection threshold is chosen as [H1, S1] & [H2,S2]. Sobottaka & Pitas defined a face localization based on HSV. They found that the threshold range of human flesh as:

Smin= 0.23, Smax =0.68, Hmin =0° & Hmax=50°.

HSV Color modelPure HueWhite Tints

Hue at 0o is red Hue at 120o is green Hue at 240o is blue

YCbCr Model

Black

Shade

Tones

Converted to HSVOriginal Image

DWT and Haar-DWTThis is one of the Frequency domain in which Steganography can be implemented.

In this work DWT better than DCT.Haar-DWT is simplest form of DWT.DWT splits component into numerous frequency sub bands as:LL – Horizontally and vertically low passLH – Horizontally low pass and vertically high passHL - Horizontally high pass and vertically low passHH - Horizontally and vertically high pass

Human eyes are more sensitive to the low frequency part (LL sub-band).

Hence secret message in hidden other 3 parts.

Haar-DWTAdv: High Computation Speed, Simplicity.Performs row and column transformation of Image matrix.

Procedure1.Haar DWT of an array2.Using Matrix Multiplication: If A is matrix on which we want to perform Haar-DWT, row transform is performed using

Q=A×P , where P=

O/P of row transform becomes I/P at column transform.Column transformation is obtained using R= P’× Q where P’ is transpose of P. Matrix P performs averaging and differencing operation.Result is four sub-bands.

1 0 1 01 0 -1 00 1 0 10 1 0 -1

EncodingSuppose C is original 24-bit color cover image of M×N size.

C= {xij, yij, zij |1<=i<=M, 1<= j . N, xij, yij, zij in {0, 1,.., 255}}Let S is secret data. Here secret data considered is binary image of size a×b.

Two cases of Encoding process:With cropping Without cropping

With CroppingLet size of cropped image is Mc×Nc where Mc≤M and Nc≤N and Mc=Nc. i.e. Cropped region must be exact square as we have to apply DWT later.

Step1:Apply skin tone detection on cover image. This will produce mask image that contains skin and non skin pixels.

Step2: Cropping is done interactively on image (Mc×Nc).

Note: Cropped area must be an exact square and cropped area should contain skin region such as face, hand etc. Cropped rectangle will act as key at receiving side.

Step3: Apply DWT to cropped area.This yields 4 sub-bands denoted as HLL, HHL,HLH,HHH Payload of image is determined based on nos of skin

pixels present in one of high frequency sub-band.Embedding in LL sub-band affects image quality greatly.

Step4: Embedding in high frequency HH sub-band is done only in skin pixels. Skin pixels are traced using skin mask detected earlier secret data is embedded

Embedding is performed in G-plane and B-plane of RBG but strictly not in R-plane as contribution of R plane in skin color is more.

Embedding is done as per raster-scan order.

Step5: Perform IDWT to combine 4 sub-bands.

Step6: A cropped stego image of size Mc×Nc is obtained. Merge the cropped stego image with original image to get the stego image of size M×N.

For merging coefficients of first and last pixels of cropped area in original image is required so that R is calculated.

Finally, Stego image is ready for secret communications.

Flowchart of With Cropping case of Embedding Process

Without CroppingThere is a major difference is actual embedding algorithm.

In this embedding algorithm, Data is hidden such that the histogram of the cover image shouldn’t get modified.

Step1:Apply skin tone detection.

Step2: Separate R, G, B planes, Apply DWT to B plane. Skin pixels from HH sub-band are retrieved and stored in one matrix.

Step3: Apply embedding algorithm to retrieved matrix.

Here, start from first pixel and using pseudorandom sequence its corresponding pixel for pair is found. To create a pseudo random sequence Lehmer’s Congruential generator is used. It generates non overlapping random sequence. Lehmer invented the multiplicative congruential algorithm - which is the basis for many of the random number generators today.

Lehmer's generators involve 3 integer parameters a, c, and m, and an initial value x0, called the seed

xk+1 = axk + c mod m

Pairs are formed using this equation.

Embedding algorithm• Input- matrix of only skin pixels, let it is S and secret

• message bits of size M• Output- Modified matrix of only skin pixels• Begin• 1] Select non-overlapping, random pair from S. Let it be p1, p2.

• If count (message bits)=M then• goto End• Else• goto step 2• 2] if p1=p2 then• goto step 1 (choose other pair)

3] if p1‚ p2 then�goto step 4.4] if message bit =0 then if p1>p2 then swap (p1,p2).Choose next message bit.Else Choose next message bit.goto step 1.5] if message bit=1 then if p1<p2 then swap(p1,p2)Choose next message bit.else Choose next message bit.goto step 1.End

Step4: This modified matrix of only skin pixels is restored in HH sub-band.

Step5: Apply IDWT to merge all the sub-bands. Finally, Stego image is ready.

DecodingCase 1- With Cropping In this case we must need a value of cropped area to retrieve data.

Suppose cropped area value is stored in ‘rect’ variable.

‘rect’ will act as a key at decoder side.By tracing skin pixels in HHH sub-band of DWT secret data is retrieved.

Case 2- Without Cropping In this case extraction of secret data is done without cropping.

Decoding requires finding skin pixel pairs that are used for embedding.

These pixel pairs are found by generating non-overlapping, random sequence using Lehmer’s Congruential generator.

Once correct pixel pairs are found, based on their values either one or zero of secret data is decided.

Flowchart of Decoding Process

Performance MeasurePeak signal to noise ratio (PSNR) is used to evaluatequality of stego image after embedding the secret message.PSNR is defined as:

Xij and Yij represents pixel values of original cover image and stego image respectively.

PSNR usually adopts dB value for quality judgement.

The larger PSNR is, higher the image quality.On the contrary smaller dB value means there is a more distortion.

PSNR values falling below 30dB indicate fairly a low quality. However, high quality strives for 40dB or more.

Capacity and PSNR of 4 final stego images in proposed method.

CASE A- WITHOUT CROPPINGCASE B- WITH CROPPING

ConclusionDigital Steganography is a fascinating scientific area which falls under the umbrella of security systems.

Skin Tone based Steganography is presented that uses skin region of images in DWT domain for embedding secret data.

Two cases for data embedding are considered, with cropping and without cropping.

According to performance measure results, proposed approach provides fine image quality. Since PSNR of all images are above 40db.

thank you..

Questions?