Method and apparatus for removing hidden data based on autoregressive generative model

1. A method of neutralizing an attack of steganography based on an autoregressive generative model, the method being performed by a computer device, the method comprising:

receiving a source image including steganography-based hidden data;

randomly selecting a plurality of target pixels from the source image; and

correcting the source image by inputting the source image and an identifier of the plurality of target pixels to the autoregressive generative model to damage the hidden data,

wherein the autoregressive generative model changes a value of a selected pixel on the basis of a pixel value distribution for pixels adjacent to the pixel in an input image,

wherein the hidden data comprises a series of values from a plurality of pixels in the source image, the plurality of pixels being randomly distributed in the source image, and

wherein the hidden data comprises of a value of at least one pixel of the plurality of target pixels.

2. The method of claim 1 , further comprising detecting an edge of the source image,

wherein the computer device selects the plurality of target pixels from the edge.

3. The method of claim 1 , wherein the autoregressive generative model is PixelCNN or Gated PixelCNN.

4. The method of claim 1 , wherein the autoregressive generative model determines a pixel value of the target pixel such that at least one of a peak signal-to-noise ratio (PSNR) and a structure similarity index (SSIM) is maintained at a reference value or greater on the basis of a likelihood of the pixel value distribution.

5. The method of claim 1 , wherein an input characteristic map and an output characteristic map have a same size, and the autoregressive generative model performs convolution using only some input characteristics.

6. The method of claim 1 , wherein the source image has a standardized format, and the autoregressive generative model is prepared by pre-learning an image distribution of the source image.

7. The method of claim 1 , wherein the computer device changes the target pixel on the basis of the adjacent pixels, which are included in a window having a center at which the target pixel is located.

8. An apparatus for neutralizing an attack of steganography based on an autoregressive generative model, the apparatus comprising:

an input device configured to receive a source image in which steganography-based data is hidden;

a storage device configured to store an autoregressive generative model for changing a specific pixel on the basis of a pixel value distribution for pixels adjacent to the specific pixel in an image; and

a computation device configured to select a plurality of target pixels from the source image, correct the source image by inputting the source image and an identifier of the target pixel to the autoregressive generative model to damage the steganography-based data,

wherein the hidden data comprises a series of values from a plurality of pixels in the source image, the plurality of pixels being randomly distributed in the source image, and

wherein the steganography-based data comprises of a value of at least one pixel of the plurality of target pixels.

9. The apparatus of claim 8 , wherein the autoregressive generative model is PixelCNN or Gated PixelCNN.

10. The apparatus of claim 8 , wherein the autoregressive generative model determines a pixel value of the target pixel such that at least one of a peak signal-to-noise ratio (PSNR) and a structure similarity index (SSIM) is maintained at a reference value or greater on the basis of a likelihood of the pixel value distribution.

11. The apparatus of claim 8 , wherein an input characteristic map and an output characteristic map have the same size, and the autoregressive generative model performs convolution using only some input characteristics.

12. The apparatus of claim 8 , wherein the source image has a standardized format, and the autoregressive generative model is prepared by pre-learning an image distribution of the source image.

13. The apparatus of claim 8 , wherein the computation device changes the target pixel on the basis of the adjacent pixels, which are included in a window having a center at which the target pixel is located.

14. The apparatus of claim 8 ,

wherein the storage device further stores an edge detection program for detecting an edge of an image, and

wherein the computation device detects an edge of the source image using the edge detection program and then selects the target pixel from the edge.

15. A computer-readable recording medium having a computer program recorded thereon to execute the method of claim 1 .