Additional-data embedding apparatus and additional-data embedding method

• US 7,257,710 B2
• Filed: 08/19/2002
• Issued: 08/14/2007
• Est. Priority Date: 08/21/2001
• Status: Expired due to Fees

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Hirai, Jun, Choi, Yoonki
• Primary Examiner(s): Moazzami; Nasser
• Assistant Examiner(s): REZA, MOHAMMAD W
• Application Number:
  US10/223,532
• Publication Number:
  US 20030053628A1
• Time in Patent Office: 1,821 Days
• Field of Search: 713169-183, 380/238, 380/232
• US Class Current: 713/176
• CPC Class Codes: G06T 1/0028 Adaptive watermarking, e.g. Human Visual System [HVS]-based watermarking
• CPC Class Codes: G06T 2201/0052 Embedding of the watermark in the frequency domain

Abstract

The present invention provides an additional-watermark embedding apparatus for embedding predetermined additional data into original data. The additional-watermark embedding apparatus includes an adder for summing the original data and the additional data, a first amplifier for amplifying the original data according to a predetermined non-linear input-output characteristic, a second amplifier for amplifying the summed data supplied from the adder according to a predetermined non-linear input-output characteristic, a subtractor for taking the difference between the amplified data obtained by the second amplifier and the amplified original data obtained by the first amplifier, and an embedder for embedding modified additional data which is obtained by modifying the waveform of the additional data according to the original data and which is supplied from the subtractor into the original data.

Claims

1. An additional-data embedding apparatus for embedding predetermined additional data into original data, comprising:

adding means for summing the original data and the additional data so as to obtain summed data;

first amplifying means for amplifying the original based upon a predetermined nonlinear input-output characteristic so as to obtain amplified original data;

second amplifying means for amplifying the summed data supplied from the adding means based upon the predetermined nonlinear input-output characteristic so as to obtain amplified summed data;

subtracting means for taking a difference between the amplified summed data obtained from the second amplifying means and the amplified original data obtained from the first amplifying means so as to obtain modified additional data; and

embedding means for embedding the modified additional data supplied from the subtracting means into the original data.

2. The additional-data embedding apparatus according to claim 1, further comprising:

storing means for storing, as a reference amplifier values of the first amplifying means and the second amplifying means for the original data and the additional data, respectively,

wherein the first amplifying means refers to the reference table stored in the storing means to amplify the original data, and the second amplifying means refers to the reference table stored in the storing means to amplify the summed data.

3. The additional-data embedding apparatus according to claim 1, wherein the input-output characteristic of each of the first amplifying means and the second amplifying means is designed so as to satisfy a predetermined integrated result of a waveform-modified version of the additional data.

4. The additional-data embedding apparatus according to claim 1, further comprising:

low-pass filtering means for passing a predetermined frequency component in input data,

wherein, when the first amplifying means and the second amplifying means each exhibit a nonlinear input-output characteristic expressed by one of a second-order and a higher-order function, the embedding means embeds portions of the modified additional data output from the subtracting means, which is filtered by the low-pass filtering means and in which a high-order component is removed, into the original data.

5. An additional-data embedding method of embedding predetermined additional data into original data, comprising:

an adding step of summing the original data and the additional data so as to obtain summed data;

a first amplifying step of amplifying the original data based upon a predetermined nonlinear input-output characteristic so as to obtain amplified original data;

a second amplifying step of amplifying the summed data obtained in the adding step based on a predetermined nonlinear input-output characteristic so as to obtain amplified summed data;

a subtracting step of taking a difference between the amplified summed data obtained in the second amplifying step and the amplified original data obtained in the first amplifying step so as to obtain modified additional data; and

an embedding step of embedding the modified additional data obtained in the subtracting step into the original data.

6. The additional-data embedding method according to claim 5, wherein amplifier values for the first amplifying step and the second amplifying step with respect to the original data and the additional data, respectively, are stored as a reference table in a storing means;

and

the reference table stored in the storing means is referred to in the first amplifying step to amplify the original data, and the reference table stored in the storing means is referred to in the second amplifying step to amplify the summed data.

7. The additional-data embedding method according to claim 5, wherein the input-output characteristic for each of the first amplifying step and the second amplifying step is designed so as to satisfy a predetermined integrated result of a waveform-modified version of the additional data.

8. The additional-data embedding method according to claim 5, further comprising:

a low-pass filtering step of passing a predetermined frequency component in input data,

wherein, in the embedding step, when amplification is performed with a nonlinear input-output characteristic expressed by one of a second-order and a higher-order function in each of the first amplifying step and the second amplifying step, a portion of the modified additional data obtained in the subtracting step, which is filtered in the low-pass filtering step and in which a high-order component is removed, is embedded into the original data.

9. An additional-data embedding apparatus data, comprising:

transforming means for performing predetermined transformation of the original data;

adding means for summing transformed data obtained from the transforming means and the additional data so as to obtain summed data;

first amplifying means for amplifying the transformed data based on a predetermined nonlinear input-output characteristic so as to obtain amplified transformed data;

second amplifying means for amplifying the summed data supplied from the adding means based on a predetermined nonlinear input-output characteristic so as to obtain amplified summed data;

subtracting means for taking a difference between the amplified summed data obtained from the second amplifying means and the amplified transformed data obtained from the first amplifying means so as to obtain modified additional data;

inverse transforming means for performing inverse transformation on the modified additional data obtained from the subtracting means, the inverse transformation being inverse to the transformation performed by the transforming means; and

embedding means for embedding the modified additional data obtained from the inverse transforming means into the original data.

10. The additional-data embedding apparatus according to claim 9, wherein the transforming means performs predetermined frequency-transformation on the original data in one of a time domain and a space domain;

and

the adding means sums the transformed data obtained by the frequency-transformation performed by the transforming means and the additional data that is a frequency-domain signal in the transforming means, the transformed data comprising frequency coefficients.

11. The additional-data embedding apparatus according to claim 10, wherein the transforming means selects a specific frequency coefficient in the frequency coefficients, and outputs the specific frequency coefficient as the transformed data.

12. The additional-data embedding apparatus according to claim 9, wherein the transforming means level-shifts the original data.

13. The additional-data embedding apparatus according to claim 9, further comprising:

storage means for storing, as a reference table, amplifier values of the first amplifying means and the second amplifying means f or the original data and the additional data, respectively,

wherein the first amplifying means refers to the reference table stored in the storing means to amplify the transformed data, and the second amplifying means refers to the reference table stored in the storing means to amplify the summed data.

14. The additional-data embedding apparatus according to claim 9, wherein the input-output characteristic of each of the first amplifying means and the second amplifying means is designed so as to satisfy a desired integrated result of the waveform-modified version of the additional data.

15. The additional-data embedding apparatus according to claim 9, further comprising:

low-pass filtering means for passing a predetermined frequency component in input data,

wherein, when the first amplifying means and the second amplifying means each exhibit a nonlinear input-output characteristic expressed by one of a second-order and higher-order function, the inverse transforming means performs inverse-transformation on a portion of the modified additional data output from the subtracting means, which is filtered by the low-pass filtering means and in which a high-order component is removed.

16. An additional-data embedding method of embedding predetermined additional data into original data, comprising:

a transforming step of performing a predetermined transformation of the original data;

an adding step of summing transformed data obtained by the predetermined transformation performed in the transforming step and the additional data so as to obtain summed data;

a first amplifying step of amplifying the transformed data based on a predetermined nonlinear input-output characteristic so as to obtain amplified transformed data;

a second amplifying step of amplifying the summed data obtained in the adding step based upon a predetermined nonlinear input-output characteristic so as to obtain amplified summed data;

a subtracting step of taking a difference between the amplified summed data obtained in the second amplifying step and the amplified transformed data obtained in the first amplifying step so as to obtain modified additional data;

an inverse transforming step of performing an inverse transformation of the modified additional data obtained in the subtracting step, the inverse transformation being inverse to the transformation performed in the transforming step; and

an embedding step of embedding the modified additional data obtained in the inverse transforming step into the original data.

17. The additional-data embedding method according to claim 16, wherein, in the transforming step, frequency-transformation is performed on the original data in one of a time domain and a space domain;

and

in the adding step, the transformed data obtained by the frequency-transformation performed in the transforming step and the additional data that is a frequency-domain signal in the transforming step are summed, the transformed data comprising frequency coefficients.

18. The additional-data embedding method according to claim 17, wherein, in the transforming step, a specific frequency coefficient in the frequency coefficients is output as the transformed data.

19. The additional-data embedding method according to claim 16, wherein the original data is level-shifted in the transforming step.

20. The additional-data embedding method according to claim 16, wherein amplifier values for the first amplifying step and the second amplifying step with respect to the original data and the additional data, respectively, are stored as a reference table in a storing means;

and

the reference table stored in the storing means is referred to in the first amplifying step to amplify the transformed data, and the reference table stored in the storing means is referred to in the second amplifying step to amplify the summed data.

21. The additional-data embedding method according to claim 16, wherein the input-output characteristic for each of the first amplifying step and the second amplifying step is designed so as to satisfy a predetermined integrated result of a waveform-modified version of the additional data.

22. The additional-data embedding method according to claim 16, further comprising:

a low-pass filtering step of passing a predetermined frequency component in data input thereto,

wherein, in the inverse transforming step, when amplification is performed with a nonlinear input-output characteristic expressed by one of a second-order and a higher-order function in each of the first amplifying step and the second amplifying step, inverse-transformation is performed on portion of the modified additional data obtained in the subtracting step, which is filtered in the low-pass filtering step and in which a high-order component is removed.