Vol. 8 Issue 3
Year:2014
Issue:Apr-Jun
Title:Detecting Hidden Data Using Higher Order Empirical Transition Matrix
Author Name:Swagota Bera and Monisha Sharma
Synopsis:
Steganalysis is the art of detecting the presence of hidden data in any common data. Universal steganalysis is general class of steganalysis techniques which can be implemented with any steganographic embedding algorithm, even an unknown algorithm. In this paper, the detection technique is based on the fact that there occurs variation in the feature vectors in an image before and after hiding. The whole image is divided in blocks of 8x8. There exists interdependency among the pixel values within the image blocks known as intrablock dependency. The statistical feature is calculated on the basis of the transition in the pixel value. The features used here is the transition probability matrix calculated by using the markov statistical process. If the transition probability matrix is found out by considering the transition in the pixel value of second pixel w.r.t to first one and so on, then it is known as one step transition probability matrix. If the transition probability matrix is found out by considering the transition in the pixel value of third pixel w.r.t to first and second one simultaneously and so on then it is known as two step transition probability matrix. Further the values of the quantized DCT pixel is restricted in -4 to 4 values which is known as thresholding. This way a feature set is calculated with optimum dimension for the classification between the cover image and the stego image.
Year:2014
Issue:Apr-Jun
Title:Detecting Hidden Data Using Higher Order Empirical Transition Matrix
Author Name:Swagota Bera and Monisha Sharma
Synopsis:
Steganalysis is the art of detecting the presence of hidden data in any common data. Universal steganalysis is general class of steganalysis techniques which can be implemented with any steganographic embedding algorithm, even an unknown algorithm. In this paper, the detection technique is based on the fact that there occurs variation in the feature vectors in an image before and after hiding. The whole image is divided in blocks of 8x8. There exists interdependency among the pixel values within the image blocks known as intrablock dependency. The statistical feature is calculated on the basis of the transition in the pixel value. The features used here is the transition probability matrix calculated by using the markov statistical process. If the transition probability matrix is found out by considering the transition in the pixel value of second pixel w.r.t to first one and so on, then it is known as one step transition probability matrix. If the transition probability matrix is found out by considering the transition in the pixel value of third pixel w.r.t to first and second one simultaneously and so on then it is known as two step transition probability matrix. Further the values of the quantized DCT pixel is restricted in -4 to 4 values which is known as thresholding. This way a feature set is calculated with optimum dimension for the classification between the cover image and the stego image.
No comments:
Post a Comment