Apparatus and method for preventing illegal distribution of digital contents by using a fingerprinting technique

Abstract

An apparatus and method for preventing illegitimate distribution of digital contents on Internet obtains a first wavelet image having a user information-embedding region by performing a wavelet transformation (WT). The user information embedding region of the first wavelet image is wavelet-transformed to obtain a second wavelet image composed of a discrete cosine (DC) region and high-frequency regions second wavelet image. A high-frequency components removed image is obtained by removing high-frequency components in the high-frequency regions of the second wavelet image, and subjected to an inverse WT (IWT) to be outputted as an IWT image. A user information embedding unit embeds user information to the IWT image, wherein data of the user information are sequentially embedded to positions where a difference value between the user information embedding region and the new user information embedding region is small, to thereby reset the user information embedding region as the user information.

1 - 14 . (canceled)  
   
   
       15 . An apparatus for preventing illegitimate distribution of digital contents on Internet by employing a fingerprinting technique, comprising: 
 a first wavelet image obtained by performing a wavelet transformation (WT) to an original image of the digital contents, wherein the first wavelet image has a user information embedding region;    a second wavelet image obtained by performing a WT to the user information embedding region of the first wavelet image, wherein the second wavelet image composed of a discrete cosine (DC) region and high-frequency regions;    a high-frequency components removed image composed of the DC region and regions obtained by removing high-frequency components in the high-frequency regions of the second wavelet image, i.e., by setting high-frequency components other than the DC as “0”, and subjected to an inverse WT (IWT) to be outputted as an IWT image; and    a user information embedding unit for embedding user information, which is provided from an operator, to the IWT image, thereby obtaining a user information embedded image with a new user information embedding region, and embedding the user information to a position determined by a random sequence generated from a location key in a blind information embedding system which does not use the first wavelet image, to thereby reset the user information embedding region as the new user information embedding region.    
   
   
       16 . The apparatus of  claim 15 , wherein the random sequence [locat(K) ε{0,1}, 1≦k≦S(LL n )] is set to have a probability defined in the following equation and wherein the user information is embedded to a position corresponding to “1” of the random sequence, LLn being a user information embedding region:  
         P (1)= ui   —   len/S ( LL   n )    P (0)=1 −ui   —   len/S ( LL   n ).  
   
   
       17 . The apparatus of  claim 16 , wherein ui_len and S(LL n ) respectively represent the size of LL n  and the user information.  
   
   
       18 . A method for preventing illegitimate distribution of digital contents on Internet comprising the steps of: 
 performing a WT to an original image to obtain a first wavelet image;    determining a user information-embedding region in the first wavelet image and performing a WT to the user information-embedding region, thereby obtaining a second wavelet image;    removing high-frequency components from the second wavelet image by setting regions other than the user information-embedding region of a discrete cosine (DC) as “0”, thereby obtaining a high frequency components removed image and performing an inverse WT (IWT) to the high-frequency components removed image, thereby obtaining an IWT image;    embedding user information and a location key provided from an operator to the IWT image to obtain a user information embedded image, thereby obtaining a user information embedded image with a new user information embedding region; and    embedding the user information to a position determined by a random sequence generated from a location key in a blind information embedding system which does not use the first wavelet image, to thereby reset the user information embedding region as the user information.    
   
   
       19 . The method of  claim 18 , wherein the random sequence [locat(K) ε{0,1}, 1≦k≦S(LL n )] is set to have a probability defined in the following equation and wherein the user information is embedded to a position corresponding to “1” of the random sequence, LLn being a user information embedding region:  
         P (1)= ui   —   len/S ( LL   n )    P (0)=1 −ui   —   len/S ( LL   n ).  
   
   
       20 . The method of  claim 18 , wherein ui_len and S(LL n ) respectively represent the size of LL n  and the user information.  
   
   
       21 . The method of  claim 16 , wherein ui_len and S(LL n ) respectively represent the size of LL n  and the user information.