Fuzzy inference method and machine, and an image production method and system

• US 20030130968A1
• Filed: 12/10/2002
• Published: 07/10/2003
• Est. Priority Date: 10/13/1997
• Status: Active Grant

• Current Assignee: Hiroshi Kutsumi, Jun Ozawa, Kouji Miura, Satoshi Matsuura, Takeshi Imanaka
• Original Assignee: Hiroshi Kutsumi, Jun Ozawa, Kouji Miura, Satoshi Matsuura, Takeshi Imanaka
• Inventors: Ozawa, Jun, Kutsumi, Hiroshi, Matsuura, Satoshi, Miura, Kouji, Imanaka, Takeshi
• Granted Patent: US 7,072,877 B2
• Time in Patent Office: Days
• US Class Current: 706/8
• CPC Class Codes: G06V 30/413 Classification of content, e.g. text, photographs or tables

Abstract

An image production method has steps of:

detecting one or more descriptive regions exclusive of a background portion of a manuscript;

calculating an image characteristic value of said descriptive region;

selecting one from a plurality of previously defined layout patterns based on said determined image characteristic value, the layout pattern including information at least about a position of the descriptive region; and

producing an image by laying out an image of said manuscript descriptive region or images of an illustration and said manuscript descriptive region according to the decided layout pattern.

Claims

1. A fuzzy inference method for inference based on fuzzy rules stated in the if-then form, comprising the steps of:

clustering inference results prior to dufuzzificating the inference results, and

defuzzificating each unit of inference results clustered through the clustering procedure thereby obtaining at least one real value.

76. A program information storage medium for use in the method as set forth in claim 1, which medium serves to store a program for carrying out all or a part of the steps of the method.

2. A fuzzy inference machine comprising:

fuzzy rule storage means for storing fuzzy rules;

fuzzy inference means for performing fuzzy inference based on the fuzzy rules stored in said fuzzy rule storage means;

clustering means for clustering inference results given by said fuzzy inference means; and

defuzzificating means for defuzzificating a unit of inference results clustered by said clustering means.

77. A program information storage medium for use in the system as set forth in claim 2, which medium serves to store a program for carrying out all or a part the functions of the means of the system.

3. An image production method comprising the steps of:

detecting one or more descriptive regions exclusive of a background portion of a manuscript;

calculating an image characteristic value of said descriptive region;

selecting one from a plurality of previously defined layout patterns based on said determined image characteristic value, the layout pattern including information at least about a position of the descriptive region; and

producing an image by laying out an image of said manuscript descriptive region or images of an illustration and said manuscript descriptive region according to the decided layout pattern.

78. A program information storage medium for use in the method as set forth in claim 3, which medium serves to store a program for carrying out all or a part of the steps of the method

4. An image production method comprising the steps of:

detecting, based on manuscript data, only an image of a descriptive region exclusive of a background portion of a manuscript;

calculating at least one of image characteristic values of a horizontal length, a vertical length, a mean pixel luminance, a color tone, a luminance distribution and a luminance centroid of said descriptive region;

selecting one from a plurality of previously defined layout patterns based on said determined image characteristic value, the layout pattern including at least one of information items about a number of illustrations, a size of an illustration, a position of the illustration and a position of the manuscript descriptive region; and

producing an image by laying out respective images of the illustration and said manuscript descriptive region according to the decided layout pattern.

10. An image production method as set forth in claim 4, wherein the performance of any one of said layout pattern selection, schematic layout pattern selection, layout fine adjustment and segmental layout pattern selection is accomplished by the steps which include:

detecting, based on the manuscript data, only the image of the manuscript, descriptive region exclusive of the background portion of the manuscript;

calculating at least one of the image characteristic values of the horizontal length, vertical length, mean pixel luminance, color tone, luminance distribution and luminance centroid of said descriptive region image; and

applying said determined image characteristic value to a previously prepared table.

16. An image production method as set forth in claim 4, wherein the performance of any one of said layout pattern selection, schematic layout pattern selection, layout fine adjustment and segmental layout pattern selection is accomplished by the steps which include:

detecting, based on the manuscript data, only the image of the descriptive region exclusive of the background portion of the manuscript;

calculating at least one of the image characteristic values of the horizontal length, vertical length, mean pixel luminance, color tone, luminance distribution and luminance centroid of said descriptive region image; and

applying said determined image characteristic value to a previously formulated rule.

22. An image production method as set forth in claim 4, wherein the performance of any one of said layout pattern selection, schematic layout pattern selection, layout fine adjustment and segmental layout pattern selection is accomplished by the steps which include:

detecting, based on the manuscript data, only the image of the descriptive region exclusive of the background portion of the manuscript;

calculating at least one of the image characteristic values of the horizontal length, vertical length, mean pixel luminance, color tone, luminance distribution, and luminance centroid of said descriptive region image;

performing fuzzy inference by applying said determined image characteristic value to a previously formulated fuzzy rule;

defuzzificating an inference result for obtaining a real value; and

accomplishing said layout pattern selection, schematic layout pattern selection, layout fine adjustment or segmental layout pattern selection based on the determined real value.

28. An image production method as set forth in claim 4, wherein the performance of any one of said layout pattern selection, schematic layout pattern selection, layout fine adjustment and segmental layout pattern selection is accomplished by the steps which include:

detecting, based on the manuscript data, only the image of the descriptive region exclusive of the background portion of the manuscript;

calculating at least one of the image characteristic values of the horizontal length, vertical length, mean pixel luminance, color tone, luminance distribution and luminance centroid of said descriptive region image;

applying said determined image characteristic value to a previously formulated fuzzy rule for fuzzy inference;

clustering results of said inference;

defuzzificating a unit of the clustered inference results for obtaining at least one real value; and

performing the layout pattern selection, schematic layout pattern selection, layout fine adjustment or segmental layout pattern selection based the determined real value.

34. An image production method as set forth in claim 4, wherein all the illustrations are imparted with a degree of relation with any one of the other illustrations and in the selection of illustrations laid according to said decided layout pattern, one illustration is first decided for selection of the subsequent illustrations based on the degree of relation with the firstly decided illustration.

40. An image production method as set forth in claim 4, wherein all the illustrations are previously imparted with an attribute of season or time, respectively, and an illustration to be laid according to said decided layout pattern is selected based on a season or time when the image is produced.

64. An image production method as set forth in claim 4, wherein all the layout patterns are previously imparted with an attribute of subjectiveness, respectively, and said layout pattern is decided based on at least one of the image characteristic values of the horizontal length, vertical length, mean pixel luminance, color tone, luminance distribution, and luminance centroid of said descriptive region image as well as on subjective information entered by a user producing the image.

70. An image production method as set forth in claim 4, wherein all the illustrations are previously imparted with an attribute of subjectiveness, respectively, and an illustration to be laid according to the decided layout pattern is selected based on subjective information entered by a user producing the image.

79. A program information storage medium for use in the method as set forth in claim 4, which medium serves to store a program for carrying out all or a part of the steps of the method

5. An image production system comprising:

descriptive region detecting means for detecting, based on manuscript data, only an image of a descriptive region exclusive of a background portion of a manuscript;

image characteristic value calculating means for calculating at least one of image characteristic values of a horizontal length, a vertical length, a mean pixel luminance, a color tone, a luminance distribution and a luminance centroid of the descriptive region detected by said descriptive region detecting means;

layout pattern storage means for storing a layout pattern including at least one of information items about a number of illustrations, a size of an illustration, a position of the illustration, and a position of the manuscript descriptive region;

layout pattern selecting means for selecting a layout pattern based on the image characteristic value determined by said image characteristic value calculating means;

illustration storage means for storing the illustrations; and

superimposing means for combining the illustration stored in said illustration storage means with the image of said descriptive region according to the layout pattern selected by said layout pattern selecting means.

13. An image production system as set forth in claim 5, further comprising table storage means for storing a table representing a relation of the image characteristic value determined by said image characteristic value calculating means versus the layout pattern to be selected, the schematic layout pattern to be selected, the layout fine adjustment or the segmental layout pattern to be selected, said table storage means being involved in the performance of any one of said layout pattern selection, schematic layout pattern selection, layout fine adjustment and segmental layout pattern selection.

19. An image production system as set forth in claim 5 further comprising:

rule storage means for storing a rule stating a relation of the image characteristic value determined by said image characteristic value calculating means versus the layout pattern to be selected, the schematic layout pattern to be selected, the layout fine adjustment or the segmental layout pattern to be selected; and

inference means for applying said image characteristic value to the rule stored in said rule storage means thereby making decision on said layout pattern, schematic layout pattern, layout fine adjustment, or segmental layout pattern, the rule storage means and the inference means being involved in the performance of any one of said layout pattern selection, schematic layout pattern selection, layout fine adjustment and segmental layout pattern selection.

25. An image production system as set forth in claim 5 further comprising:

rule storage means for storing a fuzzy rule stating a relation of the image characteristic value determined by said image characteristic value calculating means versus the layout pattern to be selected, the schematic layout pattern to be selected, the layout fine adjustment or the segmental layout pattern to be selected;

fuzzy inference means for performing fuzzy inference by applying said image characteristic value to the fuzzy rule stored in said rule storage means thereby making decision on said layout pattern selection, schematic layout pattern selection, layout fine adjustment or segmental layout pattern selection; and

defuzzificating means for obtaining a real value from a fuzzy inference result given by said fuzzy inference means, the rule storage means, fuzzy inference means and defuzzificating means being involved in the performance of any one of said layout pattern selection, schematic layout pattern selection, layout fine adjustment and segmental layout pattern selection.

31. An image production system as set forth in claim 5 further comprising:

rule storage means for storing a fuzzy rule stating a relation of the image characteristic value determined by said image characteristic value calculating means versus the layout pattern to be selected, the schematic layout pattern to be selected, layout fine adjustment, or the segmental layout pattern to be selected;

fuzzy inference means for performing inference by applying said image characteristic value to the fuzzy rule stored in said rule storage means thereby making decision on the layout pattern, schematic layout pattern, layout fine adjustment or segmental layout pattern;

clustering means for partitioning fuzzy inference results given by said fuzzy inference means into one or more cluster units; and

defuzzificating means for obtaining a real value from a unit of the fuzzy inference results clustered by said clustering means, the rule storage means, fuzzy inference means, clustering means and defuzzificating means being involved in the performance of any one of said layout pattern selection, schematic layout pattern selection, layout fine adjustment, and the segmental layout pattern selection.

37. An image production system as set forth in claim 5 further comprising illustration relation level storage means for storing a degree of relation of said illustration with any one of the other illustrations.

43. An image production system as set forth in claim 5 further comprising illustration attribute storage means for storing an attribute of season or time imparted to each of said illustrations based on a content thereof.

46. An image production system as set forth in claim 5, wherein the manuscript data is acquired from an image read by a facsimile machine and supplied via a telephone line.

49. An image production system as set forth in claim 5, wherein the manuscript data is acquired by expanding a text font into an image, the text entered through a keyboard.

52. An image production system as set forth claim 5, wherein said produced image is outputted to a facsimile machine via a telephone line.

55. A facsimile machine comprising the image production system as claimed in claim 5, and a scanner serving as means for optically reading manuscript data.

58. A TV tuner system comprising the image production system as claimed in claim 5, TV broadcasting receiver means for receiving TV broadcasting, and image output means for outputting image information inclusive of TV broadcast image information received by said TV broadcasting receiver means.

61. A television receiver comprising the image production system as claimed in claim 5, TV broadcasting receiver means for receiving TV broadcasting, and image display means for displaying image information inclusive of image information received by said TV broadcasting receiver means.

67. An image production system as set forth in claim 5 further comprising layout subjectiveness attribute storage means for storing an attribute of subjectiveness imparted to each of the layout patterns, and subjective information input means for permitting a user to enter subjective information.

73. An image production system as set forth in claim 5 further comprising illustration subjectiveness attribute storage means for storing an attribute of subjectiveness imparted to the illustration according to a content thereof, and subjective information input means for permitting a user to enter subjective information.

80. A program information storage medium for use in the system as set forth in claim 5, which medium serves to store a program for carrying out all or a part the functions of the means of the system.

6. An image production method comprising the steps of:

detecting, based on manuscript data, only an image of a descriptive region exclusive of a background portion of a manuscript;

calculating at least one of image characteristic values of a horizontal length, a vertical length, a mean pixel luminance, a color tone, a luminance distribution and a luminance centroid of said descriptive region image;

selecting, based on said determined image characteristic value, one from a plurality of previously defined schematic layout patterns including at least one of information items about a number of illustrations, an approximate position of an illustration and an approximate position of the descriptive region image;

schematically laying at positions said image of the manuscript descriptive region and said illustration according to the decided schematic layout pattern; and

making fine adjustment of the positions of the illustration and the manuscript descriptive region image based on said image characteristic value and then producing an image.

11. An image production method as set forth in claim 6, wherein the performance of any one of said layout pattern selection, schematic layout pattern selection, layout fine adjustment and segmental layout pattern selection is accomplished by the steps which include:

detecting, based on the manuscript data, only the image of the manuscript descriptive region exclusive of the background portion of the manuscript;

calculating at least one of the image characteristic values of the horizontal length, vertical length, mean pixel luminance, color tone, luminance distribution and luminance centroid of said descriptive region image; and

applying said determined image characteristic value to a previously prepared table.

17. An image production method as set forth in claim 6, wherein the performance of any one of said layout pattern selection, schematic layout pattern selection, layout fine adjustment and segmental layout pattern selection is accomplished by the steps which include:

detecting, based on the manuscript data, only the image of the descriptive region exclusive of the background portion of the manuscript;

calculating at least one of the image characteristic values of the horizontal length, vertical length, mean pixel luminance, color tone, luminance distribution and luminance centroid of said descriptive region image; and

applying said determined image characteristic value to a previously formulated rule.

23. An image production method as set forth in claim 6, wherein the performance of any one of said layout pattern selection, schematic layout pattern selection, layout fine adjustment and segmental layout pattern selection is accomplished by the steps which include:

detecting, based on the manuscript data, only the image of the descriptive region exclusive of the background portion of the manuscript;

calculating at least one of the image characteristic values of the horizontal length, vertical length, mean pixel luminance, color tone, luminance distribution, and luminance centroid of said descriptive region image;

performing fuzzy inference by applying said determined image characteristic value to a previously formulated fuzzy rule;

defuzzificating an inference result for obtaining a real value; and

accomplishing said layout pattern selection, schematic layout pattern selection, layout fine adjustment or segmental layout pattern selection based on the determined real value.

29. An image production method as set forth in claim 6, wherein the performance of any one of said layout pattern selection, schematic layout pattern selection, layout fine adjustment and segmental layout pattern selection is accomplished by the steps which include:

detecting, based on the manuscript data, only the image of the descriptive region exclusive of the background portion of the manuscript;

calculating at least one of the image characteristic values of the horizontal length, vertical length, mean pixel luminance, color tone, luminance distribution and luminance centroid of said descriptive region image;

applying said determined image characteristic value to a previously formulated fuzzy rule for fuzzy inference;

clustering results of said inference;

defuzzificating a unit of the clustered inference results for obtaining at least one real value; and

performing the layout pattern selection, schematic layout pattern selection, layout fine adjustment or segmental layout pattern selection based the determined real value.

35. An image production method as set forth in claim 6, wherein all the illustrations are imparted with a degree of relation with any one of the other illustrations and in the selection of illustrations laid according to said decided layout pattern, one illustration is first decided for selection of the subsequent illustrations based on the degree of relation with the firstly decided illustration.

41. An image production method as set forth in claim 6, wherein all the illustrations are previously imparted with an attribute of season or time, respectively, and an illustration to be laid according to said decided layout pattern is selected based on a season or time when the image is produced.

65. An image production method as set forth in claim 6, wherein all the layout patterns are previously imparted with an attribute of subjectiveness, respectively, and said layout pattern is decided based on at least one of the image characteristic values of the horizontal length, vertical length, mean pixel luminance, color tone, luminance distribution, and luminance centroid of said descriptive region image as well as on subjective information entered by a user producing the image.

71. An image production method as set forth in claim 6, wherein all the illustrations are previously imparted with an attribute of subjectiveness, respectively, and an illustration to be laid according to the decided layout pattern is selected based on subjective information entered by a user producing the image.

81. A program information storage medium for use in the method as set forth in claim 6, which medium serves to store a program for carrying out all or a part of the steps of the method

7. An image production system comprising:

descriptive region detecting means for detecting, based on manuscript data, only an image of a descriptive region exclusive of a background portion of a manuscript;

image characteristic value calculating means for calculating at least one of image characteristic values of a horizontal length, a vertical length, a mean pixel luminance, a color tone, a luminance distribution and a luminance centroid of the descriptive region image detected by said descriptive region detecting means;

schematic layout pattern storage means for storing a schematic layout pattern including at least one of information items about a number of illustrations, an approximate position of an illustration and an approximate position of the manuscript descriptive region image;

schematic layout pattern selecting means for selecting a schematic layout pattern based on the image characteristic value determined by said image characteristic value calculating means;

layout fine adjustment means for deciding a precise layout positions in the schematic layout pattern based on the image characteristic value determined by said image characteristic value calculating means, the schematic layout pattern selected by said schematic layout pattern selecting means;

illustration storage means for storing the illustrations; and

superimposing means for combining the illustration stored in said illustration storage means with said image of the descriptive region according to the layout decided by said layout fine adjustment means.

14. An image production system as set forth in claim 7, further comprising table storage means for storing a table representing a relation of the image characteristic value determined by said image characteristic value calculating means versus the layout pattern to be selected, the schematic layout pattern to be selected, the layout fine adjustment or the segmental layout pattern to be selected, said table storage means being involved in the performance of any one of said layout pattern selection, schematic layout pattern selection, layout fine adjustment and segmental layout pattern selection.

20. An image production system as set forth in claim 7 further comprising:

rule storage means for storing a rule stating a relation of the image characteristic value determined by said image characteristic value calculating means versus the layout pattern to be selected, the schematic layout pattern to be selected, the layout fine adjustment or the segmental layout pattern to be selected; and

inference means for applying said image characteristic value to the rule stored in said rule storage means thereby making decision on said layout pattern, schematic layout pattern, layout fine adjustment, or segmental layout pattern, the rule storage means and the inference means being involved in the performance of any one of said layout pattern selection, schematic layout pattern selection, layout fine adjustment and segmental layout pattern selection.

26. An image production system as set forth in claim 7 further comprising:

rule storage means for storing a fuzzy rule stating a relation of the image characteristic value determined by said image characteristic value calculating means versus the layout pattern to be selected, the schematic layout pattern to be selected, the layout fine adjustment or the segmental layout pattern to be selected;

fuzzy inference means for performing fuzzy inference by applying said image characteristic value to the fuzzy rule stored in said rule storage means thereby making decision on said layout pattern selection, schematic layout pattern selection, layout fine adjustment or segmental layout pattern selection; and

defuzzificating means for obtaining a real value from a fuzzy inference result given by said fuzzy inference means, the rule storage means, fuzzy inference means and defuzzificating means being involved in the performance of any one of said layout pattern selection, schematic layout pattern selection, layout fine adjustment and segmental layout pattern selection.

32. An image production system as set forth in claim 7 further comprising:

rule storage means for storing a fuzzy rule stating a relation of the image characteristic value determined by said image characteristic value calculating means versus the layout pattern to be selected, the schematic layout pattern to be selected, layout fine adjustment, or the segmental layout pattern to be selected;

fuzzy inference means for performing inference by applying said image characteristic value to the fuzzy rule stored in said rule storage means thereby making decision on the layout pattern, schematic layout pattern, layout fine adjustment or segmental layout pattern;

clustering means for partitioning fuzzy inference results given by said fuzzy inference means into one or more cluster units; and

defuzzificating means for obtaining a real value from a unit of the fuzzy inference results clustered by said clustering means, the rule storage means, fuzzy inference means, clustering means and defuzzificating means being involved in the performance of any one of said layout pattern selection, schematic layout pattern selection, layout fine adjustment, and the segmental layout pattern selection.

38. An image production system as set forth in claim 7 further comprising illustration relation level storage means for storing a degree of relation of said illustration with any one of the other illustrations.

44. An image production system as set forth in claim 7 further comprising illustration attribute storage means for storing an attribute of season or time imparted to each of said illustrations based on a content thereof.

47. An image production system as set forth in claim 7, wherein the manuscript data is acquired from an image read by a facsimile machine and supplied via a telephone line.

50. An image production system as set forth in claim 7, wherein the manuscript data is acquired by expanding a text font into an image, the text entered through a keyboard.

53. An image production system as set forth claim 7, wherein said produced image is outputted to a facsimile machine via a telephone line.

56. A facsimile machine comprising the image production system as claimed in claim 7, and a scanner serving as means for optically reading manuscript data.

59. A TV tuner system comprising the image production system as claimed in claim 7, TV broadcasting receiver means for receiving TV broadcasting, and image output means for outputting image information inclusive of TV broadcast image information received by said TV broadcasting receiver means.

62. A television receiver comprising the image production system as claimed in claim 7, TV broadcasting receiver means for receiving TV broadcasting, and image display means for displaying image information inclusive of image information received by said TV broadcasting receiver means.

68. An image production system as set forth in claim 7 further comprising layout subjectiveness attribute storage means for storing an attribute of subjectiveness imparted to each of the layout patterns, and subjective information input means for permitting a user to enter subjective information.

74. An image production system as set forth in claim 7 further comprising illustration subjectiveness attribute storage means for storing an attribute of subjectiveness imparted to the illustration according to a content thereof, and subjective information input means for permitting a user to enter subjective information.

82. A program information storage medium for use in the system as set forth in claim 7, which medium serves to store a program for carrying out all or a part the functions of the means of the system.

8. An image production method comprising the steps of:

detecting, based on manuscript data, only an image of a descriptive region exclusive of a background portion of a manuscript;

calculating at least one of image characteristic values of a horizontal length, a vertical length, a mean pixel luminance, a color tone, a luminance distribution and a luminance centroid of said descriptive region image;

selecting, based on said determined image characteristic value, one from a plurality of previously defined schematic layout patterns including at least one of information items about an approximate position of an illustration and an approximate position of the manuscript descriptive region image;

selecting, based on said image characteristic value, one from a plurality of previously defined segmental layout patterns to be laid in an area of illustration superimposed on the decided schematic layout pattern, the segmental layout pattern including at least one of information items about a number of illustrations, a size of an illustration and a position of the illustration; and

laying out the illustration and the image of the descriptive region according to the decided schematic layout pattern and segmental layout pattern for combining the illustration with the image of the manuscript descriptive region.

12. An image production method as set forth in claim 8, wherein the performance of any one of said layout pattern selection, schematic layout pattern selection, layout fine adjustment and segmental layout pattern selection is accomplished by the steps which include:

detecting, based on the manuscript data, only the image of the manuscript descriptive region exclusive of the background portion of the manuscript;

calculating at least one of the image characteristic values of the horizontal length, vertical length, mean pixel luminance, color tone, luminance distribution and luminance centroid of said descriptive region image; and

applying said determined image characteristic value to a previously prepared table.

18. An image production method as set forth in claim 8, wherein the performance of any one of said layout pattern selection, schematic layout pattern selection, layout fine adjustment and segmental layout pattern selection is accomplished by the steps which include:

detecting, based on the manuscript data, only the image of the descriptive region exclusive of the background portion of the manuscript;

calculating at least one of the image characteristic values of the horizontal length, vertical length, mean pixel luminance, color tone, luminance distribution and luminance centroid of said descriptive region image; and

applying said determined image characteristic value to a previously formulated rule.

24. An image production method as set forth in claim 8, wherein the performance of any one of said layout pattern selection, schematic layout pattern selection, layout fine adjustment and segmental layout pattern selection is accomplished by the steps which include:

detecting, based on the manuscript data, only the image of the descriptive region exclusive of the background portion of the manuscript;

calculating at least one of the image characteristic values of the horizontal length, vertical length, mean pixel luminance, color tone, luminance distribution, and luminance centroid of said descriptive region image;

performing fuzzy inference by applying said determined image characteristic value to a previously formulated fuzzy rule;

defuzzificating an inference result for obtaining a real value; and

accomplishing said layout pattern selection, schematic layout pattern selection, layout fine adjustment or segmental layout pattern selection based on the determined real value.

30. An image production method as set forth in claim 8, wherein the performance of any one of said layout pattern selection, schematic layout pattern selection, layout fine adjustment and segmental layout pattern selection is accomplished by the steps which include:

detecting, based on the manuscript data, only the image of the descriptive region exclusive of the background portion of the manuscript;

calculating at least one of the image characteristic values of the horizontal length, vertical length, mean pixel luminance, color tone, luminance distribution and luminance centroid of said descriptive region image;

applying said determined image characteristic value to a previously formulated fuzzy rule for fuzzy inference;

clustering results of said inference;

defuzzificating a unit of the clustered inference results for obtaining at least one real value; and

performing the layout pattern selection, schematic layout pattern selection, layout fine adjustment or segmental layout pattern selection based the determined real value.

36. An image production method as set forth in claim 8, wherein all the illustrations are imparted with a degree of relation with any one of the other illustrations and in the selection of illustrations laid according to said decided layout pattern, one illustration is first decided for selection of the subsequent illustrations based on the degree of relation with the firstly decided illustration.

42. An image production method as set forth in claim 8, wherein all the illustrations are previously imparted with an attribute of season or time, respectively, and an illustration to be laid according to said decided layout pattern is selected based on a season or time when the image is produced.

66. An image production method as set forth in claim 8, wherein all the layout patterns are previously imparted with an attribute of subjectiveness, respectively, and said layout pattern is decided based on at least one of the image characteristic values of the horizontal length, vertical length, mean pixel luminance, color tone, luminance distribution, and luminance centroid of said descriptive region image as well as on subjective information entered by a user producing the image.

72. An image production method as set forth in claim 8, wherein all the illustrations are previously imparted with an attribute of subjectiveness, respectively, and an illustration to be laid according to the decided layout pattern is selected based on subjective information entered by a user producing the image.

83. A program information storage medium for use in the method as set forth in claim 8, which medium serves to store a program for carrying out all or a part of the steps of the method.

9. An image production system comprising:

descriptive region detecting means for detecting, based on manuscript data, only an image of a descriptive region exclusive of a background portion of a manuscript;

image characteristic value calculating means for calculating at least one of image characteristic values of a horizontal length, a vertical length, a mean pixel luminance, a color tone, a luminance distribution and a luminance centroid of the descriptive region image detected by said descriptive region detecting means;

schematic layout pattern storage means for storing a schematic layout pattern including at least one of information items about an approximate position of an illustration and an approximate position of the manuscript descriptive region image;

schematic layout pattern selecting means for selecting said schematic layout pattern based on the image characteristic value determined by said image characteristic value calculating means;

segmental layout pattern storage means for storing a segmental layout pattern including at least one of information items about a number of illustrations, a size of an illustration and a position of the illustration;

segmental layout pattern selecting means for selecting said segmental layout pattern based on the image characteristic value determined by said image characteristic value calculating means;

illustration storage means for storing the illustrations; and

superimposing means for combining the illustration stored in said illustration storage means with said image of the descriptive region according to the layout selected by said segmental layout pattern selecting means.

15. An image production system as set forth in claim 9, further comprising table storage means for storing a table representing a relation of the image characteristic value determined by said image characteristic value calculating means versus the layout pattern to be selected, the schematic layout pattern to be selected, the layout fine adjustment or the segmental layout pattern to be selected, said table storage means being involved in the performance of any one of said layout pattern selection, schematic layout pattern selection, layout fine adjustment and segmental layout pattern selection.

21. An image production system as set forth in claim 9 further comprising:

rule storage means for storing a rule stating a relation of the image characteristic value determined by said image characteristic value calculating means versus the layout pattern to be selected, the schematic layout pattern to be selected, the layout fine adjustment or the segmental layout pattern to be selected; and

inference means for applying said image characteristic value to the rule stored in said rule storage means thereby making decision on said layout pattern, schematic layout pattern, layout fine adjustment, or segmental layout pattern, the rule storage means and the inference means being involved in the performance of any one of said layout pattern selection, schematic layout pattern selection, layout fine adjustment and segmental layout pattern selection.

27. An image production system as set forth in claim 9 further comprising:

rule storage means for storing a fuzzy rule stating a relation of the image characteristic value determined by said image characteristic value calculating means versus the layout pattern to be selected, the schematic layout pattern to be selected, the layout fine adjustment or the segmental layout pattern to be selected;

fuzzy inference means for performing fuzzy inference by applying said image characteristic value to the fuzzy rule stored in said rule storage means thereby making decision on said layout pattern selection, schematic layout pattern selection, layout fine adjustment or segmental layout pattern selection; and

defuzzificating means for obtaining a real value from a fuzzy inference result given by said fuzzy inference means, the rule storage means, fuzzy inference means and defuzzificating means being involved in the performance of any one of said layout pattern selection, schematic layout pattern selection, layout fine adjustment and segmental layout pattern selection.

33. An image production system as set forth in claim 9 further comprising:

rule storage means for storing a fuzzy rule stating a relation of the image characteristic value determined by said image characteristic value calculating means versus the layout pattern to be selected, the schematic layout pattern to be selected, layout fine adjustment, or the segmental layout pattern to be selected;

fuzzy inference means for performing inference by applying said image characteristic value to the fuzzy rule stored in said rule storage means thereby making decision on the layout pattern, schematic layout pattern, layout fine adjustment or segmental layout pattern;

clustering means for partitioning fuzzy inference results given by said fuzzy inference means into one or more cluster units; and

defuzzificating means for obtaining a real value from a unit of the fuzzy inference results clustered by said clustering means, the rule storage means, fuzzy inference means, clustering means and defuzzificating means being involved in the performance of any one of said layout pattern selection, schematic layout pattern selection, layout fine adjustment, and the segmental layout pattern selection.

39. An image production system as set forth in claim 9 further comprising illustration relation level storage means for storing a degree of relation of said illustration with any one of the other illustrations.

45. An image production system as set forth in claim 9 further comprising illustration attribute storage means for storing an attribute of season or time imparted to each of said illustrations based on a content thereof.

48. An image production system as set forth in claim 9, wherein the manuscript data is acquired from an image read by a facsimile machine and supplied via a telephone line.

51. An image production system as set forth in claim 9, wherein the manuscript data is acquired by expanding a text font into an image, the text entered through a keyboard.

54. An image production system as set forth claim 9, wherein said produced image is outputted to a facsimile machine via a telephone line.

57. A facsimile machine comprising the image production system as claimed in claim 9, and a scanner serving as means for optically reading manuscript data.

60. A TV tuner system comprising the image production system as claimed in claim 9, TV broadcasting receiver means for receiving TV broadcasting, and image output means for outputting image information inclusive of TV broadcast image information received by said TV broadcasting receiver means.

63. A television receiver comprising the image production system as claimed in claim 9, TV broadcasting receiver means for receiving TV broadcasting, and image display means for displaying image information inclusive of image information received by said TV broadcasting receiver means.

69. An image production system as set forth in claim 9 further comprising layout subjectiveness attribute storage means for storing an attribute of subjectiveness imparted to each of the layout patterns, and subjective information input means for permitting a user to enter subjective information.

75. An image production system as set forth in claim 9 further comprising illustration subjectiveness attribute storage means for storing an attribute of subjectiveness imparted to the illustration according to a content thereof, and subjective information input means for permitting a user to enter subjective information.

84. A program information storage medium for use in the system as set forth in claim 9, which medium serves to store a program for carrying out all or a part the functions of the means of the system.