Image processing system and image processing method

• US 6,320,981 B1
• Filed: 08/18/1998
• Issued: 11/20/2001
• Est. Priority Date: 08/28/1997
• Status: Expired due to Fees

• Current Assignee: Fuji Xerox Company Limited (Xerox Holdings Corp.)
• Original Assignee: Fuji Xerox Company Limited (Xerox Holdings Corp.)
• Inventors: Yada, Shinichi
• Primary Examiner(s): Lee, Thomas D.
• Application Number:
  US09/135,574
• Time in Patent Office: 1,190 Days
• Field of Search: 382/168, 382/169, 382/170, 382/171, 382/172, 382/270, 382/273, 358/465, 358/466, 358/443, 358/448
• US Class Current: 382/168
• CPC Class Codes: G06T 9/005 Statistical coding, e.g. Huffman, run length coding
• CPC Class Codes: H04N 19/91 Entropy coding, e.g. variable length coding [VLC] or arithmetic coding

Abstract

Block representative values are set adaptively based on the feature amount concerning the distribution state of a prepared histogram of input image data. Thus, if block truncation coding is executed based on an input image having mixed CG image and scanned image areas, efficient coding can be performed and a high-quality decomposed image can be provided. Further, when the histogram is divided into image data groups, the image data groups can be again divided to provide a high-quality decomposed image, thus efficient coding can be performed and a high-quality decomposed image can be provided more reliably.

Claims

1. An image processing system for assigning any one of block representative values to approximate image data to the image data corresponding to each of pixels making up an input image and then coding the image data in units of block images made up of the pixels, said image processing system comprising:

frequency distribution detection means for detecting a frequency distribution of image data input for each of the block images;

feature amount calculation means for calculating a feature amount concerning a distribution state of an area in the proximity of a distribution range end in the detected frequency distribution;

determination means for determining whether or not the block representative values are to be set from the frequency distribution of the area in the proximity of the distribution range end based on the calculated feature amount; and

representative value setting means for setting the block representative values based on the detected frequency distribution in response to the determination result of said determination means.

2. The image processing system of claim 1, wherein

said feature amount calculation means calculates an index indicating that the frequency distribution in the area in the proximity of the distribution range end leans to the frequency distribution range end side detected by said frequency distribution detection means as the feature amount.

3. The image processing system of claim 1, wherein

said representative value setting means sets N block representative values (where N is a n integer of two or more) and comprises:

a first block representative value setting section for setting n block representative values (n=1,

     2) based on the frequency distribution in the proximity of the distribution range end if said determination means determines that the block representative values are to be set from the frequency distribution in the proximity of the distribution range end; and

a second block representative value setting section, if said first block representative value setting section sets the n block representative values, for setting (N−

n) block representative values based on the frequency distribution of the image data to which the block representative values set by said first block representative value setting section are not assigned in the frequency distribution detected by said frequency distribution detection means.

4. The image processing system of claim 1 further comprising:

means for relating the image data to the block representative values; and

means for coding the image data based on the related block representative values.

5. An image processing system for assigning any one of block representative values to approximate image data to the image data corresponding to each of pixels making up an input image and then coding the image data in units of block images made up of the pixels, said image processing system comprising:

histogram preparation means for preparing a histogram of image data input for each of the block images;

image data dividing means for dividing the image data into image data groups based on the prepared histogram;

feature amount calculation means for calculating a feature amount concerning a distribution state of the histogram corresponding to each of the image data groups;

determination means for determining whether or not the block representative value is to be calculated using the histogram corresponding to each of the image data groups provided by said image data dividing means based on the calculated feature amount;

redividing control means for causing said image data dividing means to again divide the image data, if the block representative value is not calculated using the histogram corresponding to each of the image data groups provided by said image data dividing means in response to the determination result of said determination means; and

block representative value calculation means for calculating the block representative value in response to the determination result of said determination means.

6. The image processing system of claim 5, wherein

the image data group in the proximity of the distribution range end of the prepared histogram is separated and divided.

7. The image processing system of claim 6, wherein

the feature amount of the image data group in the proximity of the distribution range end is calculated.

8. The image processing system of claim 5 further comprising:

means for relating the image data to the block representative values; and

means for coding the image data based on the related block representative values.

9. An image processing system for assigning any one of block representative values to approximate image data to the image data corresponding to each of pixels making up an input image and then coding the image data in units of block images made up of the pixels, said image processing system comprising:

histogram preparation means for preparing a histogram of image data input for each of the block images based on a predetermined condition;

image data dividing means for dividing the image data into image data groups based on the prepared histogram;

feature amount calculation means for calculating a feature amount concerning a distribution state of the histogram corresponding to each of the image data groups;

threshold value calculation means for calculating a threshold value for again dividing the image data groups provided by said image data dividing means based on the calculated feature amount;

redividing control means for causing the provided image data groups to be again divided based on the threshold value; and

block representative value setting means for setting the block representative value based on the histogram corresponding to each of the image data groups after again divided.

10. The image processing system of claim 9, wherein

said redividing control means makes the number of the image data groups after again divided greater than that before again divided.

11. The image processing system of claim 9 further comprising:

means for relating the image data to the block representative values; and

means for coding the image data based on the related block representative values.

12. An image processing system for assigning any one of block representative values to approximate image data to the image data corresponding to each of pixels making up an input image and then coding the image data in units of block images made up of the pixels, said image processing system comprising:

histogram preparation means for preparing a histogram of image data input for each of the block images based on a predetermined condition;

image data dividing means for dividing the image data into M image data groups (where M is an integer of two or more) based on the prepared histogram;

comparison means for comparing the number of image data groups M with the preset number of block representative values N (where N is a natural number);

first representative value setting means for setting M block representative values, if the number of image data groups M is equal to or less than the number of block representative values N based on the comparison result of said comparison means;

second representative value setting means for setting N block representative values, if the number of image data groups M exceeds the number of block representative values N based on the comparison result of said comparison means; and

feature amount calculation means for calculating a feature amount concerning a distribution state of the histogram corresponding to each of the image data groups.

13. The image processing system of claim 12 further comprising:

means for relating the image data to the block representative values; and

means for coding the image data based on the related block representative values.

14. An image processing system for assigning any one of block representative values to approximate image data to the image data corresponding to each of pixels making up an input image and then coding the image data in units of block images made up of the pixels, said image processing system comprising:

histogram preparation means for preparing a histogram of image data input for each of the block images based on a predetermined condition;

image data dividing means for dividing the image data into image data groups based on the prepared histogram;

redividing control means for causing said image data dividing means to again divide the image data if the number of image data groups differs from the number of block representative values preset;

feature amount calculation means for calculating a feature amount concerning a distribution state of the histogram corresponding to each of the image data groups;

determination means for determining whether or not the block representative value is to be calculated using the histogram corresponding to the image data groups based on the calculated feature amount; and

block representative value calculation means for calculating the block representative value in response to the determination result of said determination means.

15. The image processing system of claim 14, wherein

if the number of image data groups is less than the preset number of block representative values N, said redividing control means selects the image data group to be again divided based on the distribution range of each image data group.

16. The image processing system of claim 15, wherein

the feature amount of the image data group in the proximity of the distribution range end is calculated.

17. The image processing system of claim 14, wherein

if the number of image data groups is equal to or greater than the preset number of block representative values N, said redividing control means selects based on a distance between image data groups corresponding to a difference between the maximum value of image data making up one image data group and the minimum value of image data making up a different image data group, the one image data group and the different image data group to be combined.

18. The image processing system of claim 14, wherein

the image data group in the proximity of the distribution range end of the prepared histogram is separated and divided.

19. The image processing system of claim 14 further comprising:

means for relating the image data to the block representative values; and

means for coding the image data based on the related block representative values.

20. An image processing method for assigning any one of block representative values to approximate image data to the image data corresponding to each of pixels making up an input image and then coding the image data in units of block images made up of the pixels, said image processing method comprising:

a histogram preparation step of preparing a histogram of image data input for each of the block images based on a predetermined condition;

an image data dividing step of dividing the image data into M image data groups (where M is an integer of two or more) based on the prepared histogram;

a comparison step of comparing the number of image data groups M with a preset number of block representative values N (where N is a natural number);

a first representative value setting step of setting M block representative values if the number of image data groups M is equal to or less than the number of block representative values N based on the comparison result in said comparison step;

a second representative value setting step of setting N block representative values if the number of image data groups M exceeds the number of block representative values N based on the comparison result in said comparison step; and

a feature amount calculation step of calculating a feature amount concerning a distribution state of the histogram corresponding to each of the image data groups.

21. The image processing method of claim 20, further comprising the steps of:

relating the image data to the block representative values; and

coding the image data based on the related block representative values.

22. An image processing method for assigning any one of block representative values to approximate image data to the image data corresponding to each of pixels making up an input image and then coding the image data in units of block images made up of the pixels, said image processing method comprising:

a histogram preparation step of preparing a histogram of image data input for each of the block images;

an image data dividing step of dividing the image data into image data groups based on the prepared histogram;

a feature amount calculation step of calculating a feature amount concerning a distribution state of the histogram corresponding to each of the image data groups;

a determination step of determining whether or not the block representative value is to be calculated using the histogram corresponding to each of the image data groups provided in said image data dividing step based on the calculated feature amount;

a redividing control step of causing said image data dividing step to again divide the image data if the block representative value is not calculated using the histogram corresponding to each of the image data groups provided in said image data dividing step in response to the determination result in said determination step; and

a block representative value calculation step of calculating the block representative value in response to the determination result in said determination step.

23. The image processing method of claim 22, further comprising the steps of:

relating the image data to the block representative values; and

coding the image data based on the related block representative values.

24. An image processing method for assigning any one of block representative values to approximate image data to the image data corresponding to each of pixels making up an input image and then coding the image data in units of block images made up of the pixels, said image processing method comprising:

a histogram preparation step of preparing a histogram of image data input for each of the block images based on a predetermined condition;

an image data dividing step of dividing the image data into image data groups based on the prepared histogram;

a redividing control step of causing said image data dividing step to again divide the image data if the number of image data groups differs from the number of block representative values preset;

a feature amount calculation step of calculating a feature amount concerning a distribution state of the histogram corresponding to each of the image data groups;

a determination step of determining whether or not the block representative value is to be calculated using the histogram corresponding to the image data groups based on the calculated feature amount; and

a block representative value calculation step of calculating the block representative value in response to the determination result in said determination step.

25. The image processing method of claim 24, wherein

if the number of image data groups is less than the preset number of block representative values N, said redividing control step selects the image data group to be again divided based on the distribution range of each image data group.

26. The image processing method of claim 24, wherein

if the number of image data groups is equal to or greater than the preset number of block representative values N, said redividing control step selects based on a distance between image data groups corresponding to a difference between the maximum value of image data making up one image data group and the minimum value of image data making up a different image data group, the one image data group and the different image data group to be combined.

27. The image processing method of claim 24, wherein

the image data group in the proximity of the distribution range end of the prepared histogram is separated and divided.

28. The image processing method of claim 24, further comprising the steps of:

relating the image data to the block representative values; and

coding the image data based on the related block representative values.