A kind of image discriminating device and discriminating method thereof
The technology of a discrimination method and a discrimination device is applied in the field of image processing, and can solve problems such as the inability to discriminate JPEG decompressed images and the inability to discriminate images I
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0091] Randomly select a color image in uncompressed TIFF format from the UCID image database.
[0092] The first coefficient block set generating module of the image discrimination device cuts the selected image into 32×32 image blocks from the center. Carry out 8×8 blocks, divided into 16 sub-blocks, such as Figure 15 shown. It is found that the 8th and 12th sub-blocks contain a pixel value of 255. After the 8th and 12th sub-blocks are removed, the remaining 14 sub-blocks are sorted in the original order, and then DCT is performed to obtain the DCT coefficient value, such as Figure 16 shown.
[0093] The second coefficient block set generation module is for example Figure 16 The sub-blocks shown are rounded, and all sub-blocks with AC coefficients of zero in the obtained result are removed, and reordered, the following is obtained: Figure 17 The results shown.
[0094] The DCT quantization noise set generation module pairs Figure 17 Take the difference between the...
Embodiment 2
[0096] Randomly download a color image in JPEG format through the Internet, and perform the steps described in Example 1 again.
[0097] First, the image type discrimination and conversion module saves the JPEG image as a TIFF image and converts the TIFF image to a grayscale value. After the image is divided into blocks, the first coefficient block set generation module does not find any Sub-blocks of pixel values, then perform DCT transformation on all sub-blocks,
[0098] The second coefficient block set generating module processes the transformed sub-blocks, and reserves and reorders the sub-blocks 1, 6, 7, 12, 13, and 14 whose AC coefficients are not all zero.
[0099] The DCT quantization noise set generation module takes the difference between the DCT coefficient values between the sorted sub-blocks, calculates the variance through the discrimination module, and obtains a variance value of and compared with the threshold, since less than the threshold T A , so it is...
Embodiment 3
[0101] The image discrimination device and method of the present invention are verified by converting from uncompressed to JPEG and back to uncompressed format.
[0102] Randomly select two TIFF format images TIF1 and TIF2 in the UCID image library, and compress them into JPEG images JPG1 and JPG2 with compression factors Q=99 and Q=100 respectively;
[0103] Decompress JPG1 and JPG2 back to TIFF format images respectively, save and check according to the steps described in Example 1. The final result is:
[0104] Variance value of TIF1 less than the threshold T A , so the image is a JPEG decompressed image. The judgment result is correct.
[0105] Variance value of TIF2 less than the threshold T A , so the image is a JPEG decompressed image. The judgment result is correct.
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


