Detecting double JPEG compression in images
View Patent ↗Techniques are generally described for detecting double JPEG compression in images. Example detection techniques of double JPEG compression may include receiving JPEG images for analysis and extracting 2-dimensional (2-D) arrays of JPEG coefficients from the images. 2-D difference arrays may be generated from the array of JPEG coefficients, with the entries in the difference array reflecting relative changes in values of pairs of entries in the array of JPEG coefficients. The detection techniques also model the difference arrays using random processes, and evaluate whether the random processes reveal statistical artifacts in the JPEG images. These statistical artifacts result from double JPEG impression performed on the JPEG images.
1. A computer-readable storage medium storing computer-executable instructions to process a JPEG image, the computer-executable instructions comprising:
receiving the JPEG image for analysis;
extracting a 2-dimensional (2-D) array of JPEG coefficients from the JPEG image;
generating a 2-D difference array based on the 2-D array of JPEG coefficients, wherein entries in the 2-D difference array reflect relative changes in values of pairs of entries in the 2-D array of JPEG coefficients;
modeling the 2-D difference arrays using a Markov random process; and
evaluating whether the random process reveals statistical artifacts in the JPEG image, the statistical artifacts resulting from double JPEG impression performed on the JPEG image.
2. The computer-readable storage medium of claim 1 , the extracting the 2-D array of JPEG coefficients instructions comprising extracting a 2-D array of JPEG coefficients for a luminance component of the JPEG image.
3. The computer-readable storage medium of claim 1 , the extracting the 2-D array of JPEG coefficients instructions comprising extracting the 2-D array of JPEG coefficients for at least one chrominance component of the JPEG image.
4. The computer-readable storage medium of claim 1 , the generating the 2-D difference array instructions comprising generating a horizontal difference array, wherein the horizontal difference array reflects relative changes in values between pairs of horizontally-related elements appearing in the 2-D array of JPEG coefficients.
5. The computer-readable storage medium of claim 1 , the generating the 2-D difference array instructions comprising generating a vertical difference array, wherein the vertical difference array reflects relative changes in values between pairs of vertically-related elements appearing in the 2-D array of JPEG coefficients.
6. The computer-readable storage medium of claim 1 , the generating the 2-D difference array instructions comprising generating a main diagonal difference array, wherein the main diagonal difference array reflects relative changes in values between pairs of elements appearing along a main diagonal in the 2-D array of JPEG coefficients.
7. The computer-readable storage medium of claim 1 , the generating the 2-D difference array instructions comprising generating a minor diagonal difference array, wherein the minor diagonal difference array reflects relative changes in values between pairs of elements appearing along a minor diagonal in the 2-D array of JPEG coefficients.
8. The computer-readable storage medium of claim 1 , the instructions further comprising applying thresholding to elements in the 2-D difference array before modeling the 2-D difference arrays using the Markov random process.
9. The computer-readable storage medium of claim 1 , the instructions further comprising computing absolute values of elements in the 2-D array of JPEG coefficients.
10. The computer-readable storage medium of claim 1 , the instructions further comprising populating a transition probability matrix for the 2-D difference array, wherein entries in the transition probability matrix represent probabilities of values of first elements in the 2-D difference array transitioning to values of second elements in the difference array.
11. The computer-readable storage medium of claim 1 , the instructions further comprising detecting statistical artifacts of double JPEG compression in the JPEG image.
12. The computer-readable storage medium of claim 11 , the detecting statistical artifacts instructions comprising detecting rounding errors that occur during the double JPEG compression.
13. The computer-readable storage medium of claim 11 , the detecting statistical artifacts instructions comprising detecting weakened statistical correlation between a plurality of elements in the 2-D difference array.
14. The computer-readable storage medium of claim 11 , the detecting statistical artifacts instructions comprising detecting statistical artifacts resulting from a first JPEG compression of the JPEG image using a first quantization factor from a first quantization step, and a second quantization factor from a second quantization step, wherein the ratio of the first and second quantization factors is an integer.
15. The computer-readable storage medium of claim 1 , the instructions further comprising asserting a signal indicating that the JPEG image is a double JPEG image.
16. The computer-readable storage medium of claim 1 , the instructions further comprising asserting a signal indicating that the JPEG image is a single JPEG image.
17. A system for processing a JPEG image comprising:
a computer readable storage medium storing computer-executable instructions for processing the JPEG image;
a processor that is configured by the computer executable instructions to:
receive the JPEG image for analysis;
extract a 2-dimensional (2-D) array of JPEG coefficients from the JPEG image;
generate a 2-D difference array based on the 2-D array of JPEG coefficients, wherein entries in the 2-D difference array reflect relative changes in values of pairs of entries in the 2-D array of JPEG coefficients;
model the 2-D difference arrays using a random process; and
evaluate whether the random process reveals statistical artifacts in the JPEG image, the statistical artifacts resulting from double JPEG impression performed on the JPEG image.
18. The system of claim 17 , wherein the processor is further configured by the computer-executable instructions to extract the 2-D array of JPEG coefficients for a luminance component of the JPEG image.
19. The system of claim 17 , processor is further configured by the computer-executable instructions to extract the 2-D array of JPEG coefficients for at least one chrominance component of the JPEG image.
20. A computer-implemented method for processing a JPEG image, the method comprising:
receiving the JPEG image for analysis to determine whether the JPEG image exhibits statistical artifacts associated with double JPEG compression;
extracting a 2-dimensional (2-D) array of JPEG coefficients from the JPEG image;
generating a plurality of 2-D difference arrays based on the array of JPEG coefficients, wherein the 2-D difference arrays include a first difference array and a second difference array, wherein entries in the first difference array reflect relative changes in values of pairs of entries in the 2-D array of JPEG coefficients taken along a first dimension of the JPEG image, and wherein entries in the second difference array reflect relative changes in values of pairs of entries in the 2-D array of JPEG coefficients taken along a further dimension of the JPEG image;
applying thresholding to elements within the plurality of 2-D difference arrays;
modeling thresholded elements of the 2-D difference arrays using a random process; and
evaluating whether the random process reveals statistical artifacts in the JPEG image, the statistical artifacts resulting from double JPEG impression performed on the JPEG image.