Compressed image data enhancement

• US 20070140570A1
• Filed: 12/21/2005
• Published: 06/21/2007
• Est. Priority Date: 12/21/2005
• Status: Active Grant

• Current Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Curry, Donald
• Granted Patent: US 7,577,302 B2
• Time in Patent Office: Days
• US Class Current: 382/232
• CPC Class Codes: H04N 19/48 using compressed domain processing techniques other than decoding, e.g. modification of transform coefficients, variable length coding [VLC] data or run-length data motion estimation in a transform domain H04N19/547; processing of decoded motion vectors H04N19/513
• CPC Class Codes: H04N 19/503 involving temporal prediction adaptive coding with adaptive selection between spatial and temporal predictive coding H04N19/107; adaptive coding with adaptive selection among a plurality of temporal predictive coding modes H04N19/109
• CPC Class Codes: H04N 19/60 using transform coding

Abstract

An image background remover is disclosed that performs background removal of JPEG (Joint Photographic Experts Group) compressed image data without first performing a complete decompression of the compressed image data.

Claims

1. A method for processing compressed image data, comprising:

decoding compressed image data one block at a time to generate decoded blocks;

stretching first values of the decoded blocks; and

encoding decoded blocks with stretched-first values replacing corresponding first values to generate background removed compressed image data.

2. The method of claim 1, further comprising:

selecting decoded luminance blocks; and

reading quantized DC coefficients of decoded luminance blocks as the first values.

3. The method of claim 2, further comprising:

dequantizing the first values to obtain DC coefficients;

multiplying the DC coefficients by a stretch factor to obtain stretched DC coefficients; and

quantizing the stretched DC coefficients to obtain a quantized-stretched DC coefficients.

4. The method of claim 3, further comprising:

replacing the quantized DC coefficients in the decoded luminance blocks with corresponding quantized-stretched DC coefficients to obtain modified luminance blocks;

encoding the modified luminance blocks to obtain a compressed modified luminance block; and

outputting the compressed modified luminance blocks together with encoded chrominance blocks to form the background removed compressed image data.

5. The method of claim 3, further comprising:

combining functions of the dequantizing, multiplying and quantizing steps into a lookup table; and

obtaining the quantized-stretched DC coefficients based on a quantized DC coefficients using the lookup table.

6. The method of claim 1, further comprising:

obtaining second values from the compressed image data that correspond to luminance DC coefficients;

analyzing the second values to generate an analysis result; and

selecting a white point for the image data based on the analysis result.

7. The method of claim 6, further comprising:

dividing a white point value into a target luminance value to generate a stretch factor.

8. The method of claim 7, further comprising:

obtaining the first and the second luminance DC coefficient values from a background layer and/or a foreground layer of an MRC format data, or from one or more layers of an N-layer format data or a luminance value from a color layer;

generating stretched values based on the second values and the stretch factor; and

replacing each of the first values in the background layer and/or foreground layer, or the layers of the N-layer format data with a corresponding one of the stretched values.

9. The method of claim 6, further comprising:

selecting a number of the second values to be less than a number of the first values;

setting the target luminance value to be a maximum luminance value; and

generating a histogram for analyzing the second values.

10. A xerographic apparatus performing the method of claim 1.

11. A program stored in a machine readable medium, transmitted by a transmitter or received by a receiver, the program executed by a machine to perform a process, comprising:

decoding compressed image data one block at a time to generate decoded blocks;

stretching first values of the decoded blocks; and

encoding decoded blocks with stretched first values replacing corresponding first values to generate background removed compressed image data.

12. The program of claim 11, the process performed by a machine executing the program further comprising:

selecting decoded luminance blocks; and

reading quantized DC coefficients of decoded luminance blocks as the first values.

13. The program of claim 12, the process performed by a machine executing the program further comprising:

dequantizing the first values to obtain DC coefficients;

multiplying the DC coefficient by a stretch factor to obtain a stretched DC coefficient; and

quantizing the stretched DC coefficient to obtain a quantized-stretched DC coefficient.

14. The program of claim 13, the process performed by a machine executing the program further comprising:

replacing the quantized DC coefficients in the decoded luminance blocks with corresponding quantized-stretched DC coefficients to obtain modified luminance blocks;

encoding the modified luminance blocks to obtained compressed modified luminance blocks; and

outputting the compressed modified luminance blocks together with encoded chrominance blocks to form the background removed compressed image data.

15. The program of claim 13, the process performed by a machine executing the program further comprising:

combining functions of the dequantizing, multiplying and quantizing steps into a lookup table; and

obtaining the quantized-stretched DC coefficient based on a quantized DC coefficient using the lookup table.

16. The program of claim 11, the process performed by a machine executing the program further comprising:

obtaining second values from the compressed image data that correspond to luminance DC coefficients;

analyzing the second values to generate an analysis result; and

selecting a white point for the image data based on the analysis result.

17. The program of claim 16, the process performed by a machine executing the program further comprising:

dividing a white point value into a target luminance value to generate a stretch factor.

18. The program of claim 17, the process performed by a machine executing the program further comprising:

obtaining the first and the second luminance DC coefficient values from a background layer and/or a foreground layer of an MRC format data, or from one or more layers of an N-layer format data, or a luminance value from a color layer;

generating stretched values based on the second values and the stretch factor; and

replacing each of the first values in the background layer and/or foreground layer, or the layers of the N-layer format data with a corresponding one of the stretched values.

19. The programs of claim 11, comprising a xerographic machine.

20. An apparatus that processes compressed image data, comprising:

means for locating first values corresponding to luminance in the compressed image data;

means for reading the first values;

means for selecting a compressed luminance block in the compressed image data;

means for decoding the compressed luminance block into quantized luminance coefficients;

means for converting a quantized DC coefficient of the quantized luminance coefficients to a quantized-stretched DC coefficient;

means for replacing the quantized DC coefficient in the quantized luminance coefficients with the quantized-stretched DC coefficient to obtain modified quantized-luminance coefficient blocks; and

means for encoding the modified quantized-luminance coefficient blocks to obtain compressed modified luminance blocks.