Image processing apparatus and method and medium

• US 6,389,162 B2
• Filed: 02/12/1997
• Issued: 05/14/2002
• Est. Priority Date: 02/15/1996
• Status: Expired due to Fees

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Maeda, Mitsuru
• Primary Examiner(s): Tran, Phuoc
• Application Number:
  US08/798,132
• Publication Number:
  US 20010053248A1
• Time in Patent Office: 1,917 Days
• Field of Search: 382/168-172, 382/251, 382/176, 382/224, 382/180, 382/162, 382/165, 382/166, 358/462, 358/464, 358/466, 358/518-522, 358/461, 358/463, 358/465, 358/467
• US Class Current: 382/172
• CPC Class Codes: G06T 9/005 Statistical coding, e.g. Huffman, run length coding
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/162 Quantising the image signal
• CPC Class Codes: H04N 1/4074 using histograms

Abstract

There are provided an image processing apparatus and method which can properly determine the number of quantization levels for quantizing image data by properly removing noise from the background of the image data. A histogram forming unit forms the output frequency distribution of densities of image data. A background density width detecting unit determines a density area exhibiting output frequency values equal to or larger than a predetermined threshold and including a maximum output frequency value in the formed distribution. A quantization level count determining unit decides the number of quantization levels on the basis of the number of maximum values included in the distribution other than the determined density area.

Claims

1. An image processing apparatus comprising:

input means for inputting multi-level image data representing an image;

extracting means for extracting binary image data from the multi-level image data;

dividing means for dividing the image into a plurality of blocks based on the binary image data;

deciding means for deciding a number of quantization levels to be utilized for quantization processing for each of the plurality of blocks based on the binary image data; and

quantizing means for quantizing the multi-level image data in each of said blocks by using the number of quantization levels decided by the deciding means, said quantizing means comprising forming means for forming frequency histograms of said multi-level image data in units of said blocks, wherein the number of quantization levels is determined on the basis of the frequency histograms,

wherein the multi-level image data is constituted by a plurality of color component data, and wherein said binary image data is extracted by the extracting means on the basis of a distribution of said multi-level image data, said distribution being examined on the basis of the frequency histograms of said multi-level image data.

2. The apparatus according to claim 1, wherein said dividing means divides the image into said blocks on the basis of attributes of the image.

3. The apparatus according to claim 2, wherein said attributes include at least a “

character” and

a “

photograph”

.

4. The apparatus according to claim 1, further comprising discriminating means for discriminating characters of a plurality of colors based on the quantized multi-level image data.

5. The apparatus according to claim 4, wherein said discriminating means outputs the characters of a plurality of colors as character codes.

6. An image processing method comprising:

an input step of inputting multi-level image data representing an image;

an extraction step of extracting binary image data from the multi-level image data;

a division step of dividing the image into a plurality of blocks based on the binary image data;

a deciding step of deciding a number of quantization levels to be utilized for quantization processing for each of the plurality of blocks based on the binary image data; and

a quantization step of quantizing the multi-level image data in each of said blocks by using the number of quantization levels by deciding step, said quantizing step comprising a forming step for forming frequency histograms of said multi-level image data in units of said blocks, wherein the number of quantization levels is determined on the basis of the frequency histograms,

wherein the multi-level image data is constituted by a plurality of color component data, and wherein said binary image data is extracted by the extracting step on the basis of a distribution of said multi-level image data, said distribution being examined on the basis of the frequency histograms of said multi-level image data.

7. A computer readable memory that stores program codes of image processing, comprising:

a program code of an input step of inputting multi-level image data representing an image;

a program code of an extraction step of extracting binary image data from the multi-level image data;

a program code of a division step of dividing the image into a plurality of blocks based on the binary image data;

a program code of a deciding step of deciding a number of quantization levels to be utilized for quantization processing for each of the plurality of blocks based on the binary image data; and

a program of quantization step of quantizing the multi-level image data in each of said blocks by using the number of quantization levels decided by deciding step, said quantization step comprising a forming step for forming frequency histograms of said multi-level image data in units of said blocks, wherein the number of quantization levels is determined on the basis of the frequency histograms,

wherein the multi-level image data is constituted by a plurality of color component data, and wherein said binary image data is extracted by the extracting step on the basis of a distribution of said multi-level image data, said distribution being examined on the basis of the frequency histograms of said multi-level image data.

8. An image processing apparatus comprising:

forming means for forming a frequency distribution of densities of image data;

calculating means for calculating a predetermined threshold on the basis of the distribution formed by said forming means;

extracting means for extracting a density exhibiting maximum frequency value in the distribution formed by said forming means;

determining means for determining a density area exhibiting frequency values not less than the threshold and including the density exhibiting the maximum frequency value; and

output means for uniformly outputting densities of the image data which have densities within the density area as densities exhibiting the maximum frequency value,

wherein the image data is color image data.

9. The apparatus according to claim 8, wherein said output means outputs the density exhibiting the maximum frequency value as a background density of the image data.

10. The apparatus according to claim 8, wherein said calculating means comprises

quantizing means for quantizing the distribution on the basis of the density exhibiting the maximum frequency value, and

detecting means for detecting numbers of same frequency values in the distribution quantized by said quantizing means, and

calculates the threshold on the basis of a frequency value exhibiting a maximum number of same frequency values of the numbers of frequency values detected by said detecting means.

11. The apparatus according the claim 8, further comprising acquiring means for acquiring image data, of the image data, which has an even density, and

wherein said forming means forms a distribution of frequency values of densities of the image data acquired by said acquiring means.

12. The apparatus according to claim 8, further comprising:

a plurality of encoding means for encoding said image data;

dividing means for dividing image data into a plurality of areas in unit of attributes;

counting means for counting maximum values present in an output distribution of densities of image data corresponding to the area; and

selecting means for selecting one of said encoding means to encode the image data corresponding to the area on the basis of a result obtained by said counting means.

13. The apparatus according to claim 12, wherein said encoding means includes at least multi-level image encoding, binary image encoding, and shape encoding.

14. The apparatus according to claim 8, wherein said forming means forms luminance image data from the color image data, and forms a frequency distribution of the luminance image data.

15. An image processing method comprising:

a formation step of forming a frequency distribution of densities of image data;

a calculation step of calculating a predetermined threshold on the basis of the distribution formed in said formation step;

an extraction step of extracting a density exhibiting a maximum frequency value in the distribution formed in said formation step;

a determination step of determining a density area exhibiting frequency values not less than the threshold and including the density exhibiting the maximum frequency value; and

an output step of uniformly outputting densities of the image data which have densities within the density area as densities exhibiting the maximum frequency value,

wherein the image data is color image data.

16. A computer readable memory that stores program codes of an image processing, comprising:

a program code of a formation step of forming a frequency distribution of densities of image data;

a program code of a calculation step of calculating a predetermined threshold on the basis of the distribution formed in said formation step;

a program code of an extraction step of extracting a density exhibiting a maximum frequency value in the distribution formed in said formation step;

a program code of a determination step of determining a density area exhibiting frequency values not less than the threshold and including the density exhibiting the maximum frequency value; and

a program code of an output step of uniformly outputting densities of the image data which have densities within the density area as densities exhibiting the maximum frequency value,

wherein the image data is color image data.

17. An image processing apparatus comprising:

forming means for forming a frequency distribution of densities of image data;

determining menas for determining a density area exhibiting frequency values not less than a predetermined threshold and including a maximum frequency value in the distribution formed by said forming means; and

first deciding means for deciding numbers of quantization levels on the basis of a number of maximum values included in the distribution other than the density area determined by said determining means.

18. The apparatus according to claim 17, further comprising:

extracting means for extracting densities exhibiting maximum and minimum frequency values; and

second deciding means for deciding a quantization level for quantization of the image data from the distribution on the basis of the density exhibiting the maximum frequency value and extracted by said extracting means.

19. The apparatus according to claim 18, wherein if there are a plurality of maximum values in a predetermined density range and a predetermined frequency range, said extracting means extracts a largest maximum value among the plurality of maximum values.

20. The apparatus according to claim 18, further comprising:

quantizing means for quantizing the image data on the basis of the quantization level; and

first output means for outputting a density of the image data on the quantization level which is quantized by said quantizing means as a maximum density on the quantization level.

21. The apparatus according to claim 17, further comprising second output means for uniformly outputting densities of the image data having densities within the density area as densities exhibiting the maximum frequency values.

22. The apparatus according to claim 21, wherein said second output means outputs the density exhibiting the maximum frequency value as a background density of the image data.

23. The apparatus according to claim 22, further comprising detecting means for detecting an area having a smaller density difference than the image data, and

wherein said second output means outputs a background density of the image data on the basis of a distribution of densities of the image data in the area detected by said detecting means.

24. The apparatus according to claim 17, further comprising dividing means for dividing the image data into a plurality of areas in units of attributes, and

wherein said forming means forms a frequency distribution of the densities of the image data corresponding to the area.

25. The apparatus according to claim 17, wherein said image data is image data represented by a plurality of signals, and said forming means forms a frequency distribution of constituent elements of the image.

26. The apparatus according to claim 25, wherein each signal of the image data is represented by a luminance and a chromaticity.

27. The apparatus according to claim 26, wherein said determining means determines density areas each exhibiting frequency values not less than a predetermined threshold and including a density exhibiting a maximum frequency value in a frequency distribution of density values of each of the signals, formed by said forming means, and

further comprising judging means for judging the number of quantization levels for quantizing the image data and colors included in the image data on the basis of the number of maximum values included in the distribution other than each of density areas determined by said determining means.

28. The apparatus according to claim 25, wherein each signal of the image data is represented by an intensity of a reference color stimulus.

29. An image processing method comprising:

a formation step of forming a frequency distribution of densities of image data;

a determination step of determining a density area exhibiting frequency values not less than a predetermined threshold and including a maximum frequency value in the distribution formed at said formation step; and

a first decision step of deciding numbers of quantization levels on a basis of the number of maximum values included in the distribution other than the density area determined at said determination step.

30. A computer readable memory that stores program codes of an image processing, comprising:

a program code of a formation step of forming a frequency distribution of densities of image data;

a program code of a determination step of determining a density area exhibiting frequency values not less than a predetermined threshold and including a maximum frequency value in the distribution formed at said formation step; and

a program code of a first decision step of deciding numbers of quantization levels on the basis of a number of maximum values included in the distribution other than the density area determined at said determination step.