Dot region discriminating method

• US 5,025,481 A
• Filed: 10/04/1989
• Issued: 06/18/1991
• Est. Priority Date: 10/21/1988
• Status: Expired due to Term

• Current Assignee: Ricoh Company Limited
• Original Assignee: Ricoh Company Limited
• Inventors: Ohuchi, Satoshi
• Primary Examiner(s): BRINICH, STEPHEN M
• Application Number:
  US07/417,074
• Time in Patent Office: 622 Days
• Field of Search: 382/9, 382/53, 358/462, 358/464
• US Class Current: 382/176
• CPC Class Codes: H04N 1/40062 Discrimination between different image types, e.g. two-tone, continuous tone

Abstract

A dot region discriminating method discriminates a dot region of an image which is described by a digital input image signal. The method includes detecting one or more extreme points from a first local region within the image. Each of the extreme points corresponds to a peak or valley of a density change of a picture element within the first local region. The method further comprises detecting whether or not one of (1) a center picture element within one of said local regions of the image or (2) one of the second local regions, belongs to a dot region, this second detecting step being based on a distribution of the extreme points detected in the first local region by the first detecting step. Finally, the method involves discriminating the dot region of the image based on the second detection step.

Claims

1. A dot region discriminating method for discriminating a dot region of an image which is described by a digital multilevel input image signal, said dot region discriminating method comprising:

(a) an extreme point detecting step for detecting one or a plurality of extreme points from a first local region within the image, each of said extreme points corresponding to a peak or a valley of a density change of a picture element within said first local region, said first local region corresponding to a matrix which comprises M×

M picture elements;

(b) a dot region detecting step for detecting whether or not

(1) one of a center picture element within one of said second local regions of the image or

(2) one of said second local regions, belongs to a dot region;

wherein the dot region detecting step is based on a distribution of the extreme points detected within said first local region by said extreme point detecting step, and wherein each of said second local regions corresponds to a block which comprises N×

N picture elements; and

(c) a discriminating step for discriminating the dot region of the image based on the detection made by said dot region detecting step.

2. The dot region discriminating method as claimed in claim 1 wherein said dot region detecting step includes substeps of independently counting the extreme points which correspond to the peaks in each of blocks which correspond to said second local regions and the extreme points which correspond to the valleys in each of the blocks and determining a larger one of the counted extreme points in each block as the number of extreme points for the block, and said discriminating step discriminates whether or not a predetermined picture element within an object block belongs to the dot region based on a relationship of the number of extreme points of the object block and the numbers of extreme points of the blocks surrounding the object block.

3. The dot region discriminating method as claimed in claim 1 wherein said extreme point detecting step detects that a center picture element within the matrix which corresponds to said first local region is an extreme point when a density level of the center picture element within the matrix is a maximum or a minimum compared to density levels of picture elements surrounding the center picture element within the matrix.

4. The dot region discriminating method as claimed in claim 3 wherein said extreme point detecting step detects that a center picture element within the matrix which corresponds to said first local region is an extreme point when an absolute value of a density difference between the density levels of the center picture element within the matrix and each of the surrounding picture elements is greater than or equal to a predetermined threshold value.

5. The dot region discriminating method as claimed in claim 4 wherein the surrounding picture elements within the matrix are respectively located in a specific direction from the center picture element within the matrix.

6. The dot region discriminating method as claimed in claim 5 wherein the specific direction is selected from directions which are right, left, up, down or 45°

angle to the center picture element within the matrix.

7. The dot region discriminating method as claimed in any of claim 2 wherein said dot region detecting step detects the dot region by moving each block one picture element at a time.

8. The dot region discriminating method as claimed in claim 7 wherein said dot region detecting step detects that the center picture element of an object block is a dot region when a number of blocks in which the number of extreme points is greater than or equal to a predetermined threshold value is greater than or equal to a predetermined threshold value.

9. The dot region discriminating method as claimed in claim 7 wherein said dot region detecting step detects that the center picture element of an object block is a dot region when a sum total of absolute values of differences between the number of extreme points in an object block and the surrounding blocks is less than or equal to a predetermined threshold value.

10. The dot region discriminating method as claimed in claim 1 wherein said dot region detecting step detects the dot region by moving each block one block at a time.

11. The dot region discriminating method as claimed in claim 10 wherein said dot region detecting step detects that all picture elements within an object block are dot regions when the number of blocks in which the number of extreme points is greater than or equal to the predetermined threshold value is greater than or equal to the predetermined threshold value.

12. The dot region discriminating method as claimed in claim 10 wherein said dot region detecting step detects that all picture elements within an object block are dot regions when a sum total of absolute values of differences between the number of extreme points in the object block and the surrounding blocks is less than or equal to a predetermined threshold value.

13. The dot region discriminating method as claimed in claim 1 wherein said extreme point detecting step uses a rectangular region of the image as said first local region.

14. The dot region discriminating method as claimed in claim 1 which further comprises a smoothing step for smoothing the digital multilevel input image signal prior to said extreme point detecting step, said extreme point detecting step detects the one or a plurality of extreme points from said first local region within the image and a smoothened image which is described by a smoothened digital multilevel input image signal, and said dot region detecting step detects the dot region based on a synthesized image which is a logical sum of the image and the smoothened image.

15. The dot region discriminating method as claimed in claim 1/wherein said extreme point detecting step independently detects one or a plurality of extreme points from said first local region within the image based on first and second extreme point detection conditions which are mutually different, and said dot region detecting step detects the dot region based on a logical sum of results of the extreme point detection made with said first and second extreme point detection conditions.

16. The dot region discriminating method as claimed in claim 15 wherein said first extreme point detection condition determines a center picture element within said first local region as an extreme point when a density level of the center picture element within said first local region is a predetermined threshold value greater than or a predetermined threshold value smaller than the density levels of surrounding picture elements within said first local region.

17. The dot region discriminating method as claimed in claim 16 wherein said second extreme point detection condition determines the center picture element within said first local region as an extreme point when the density level of the center picture element within said first local region is greater than or equal to the density levels of all other picture elements which are within a region of L ×

L picture elements including the center picture element within said first local region and the density level of the center picture element within said first local region is a predetermined threshold value greater than or equal to the density levels of the other picture elements which are outside the region comprising L×

L picture elements and inside a region comprising M×

M picture elements, where L<

M.

18. The dot region discriminating method as claimed in claim 16 wherein said second extreme point detection condition determines the center picture element within said first local region as an extreme point when the density level of the center picture element within said first local region is less than or equal to the density levels of all other picture elements indicated which are within a region of L×

L picture elements including the center picture element within said first local region and the density level of the center picture element within said first local region is a predetermined threshold value less than or equal to the density levels of the other picture elements which are outside the region comprising L×

L picture elements and inside a region comprising M×

M picture elements, where L<

M.

19. The dot region discriminating method as claimed in claim 15 which further comprises a smoothing step for smoothing the digital multilevel input image signal prior to said extreme point detecting step, said extreme point detecting step detects the one or a plurality of extreme points from said first local region within the image and a smoothened image which is described by a smoothened digital multilevel input image signal, and said dot region detecting step detects the dot region based on a synthesized image which is a logical sum of the image and the smoothened image.

20. The dot region discriminating method as claimed in claim 1 wherein said extreme point detecting step detects that a center picture element within said first local region is an extreme point when a density level of the center picture element within said first local region is greater than or less than density levels of surrounding picture elements within said first local region.

21. The dot region discriminating method as claimed in claim 20 wherein said extreme point detecting step detects that the center picture element within said first local region is an extreme point when an absolute value of a density difference between the density level of the center picture element within said first local region and an average value of the density levels of two picture elements which are located at symmetrical positions about the center picture element within said first local region is greater than or equal to a predetermined threshold value.

22. The dot region discriminating method as claimed in claim 1 wherein said dot region detecting step includes substeps of subdividing each block into a plurality of small regions and determining the number of extreme points within each block depending on a distribution of the extreme points detected within the small regions of the block.

23. The dot region discriminating method as claimed in claim 22 wherein said substep of determining the number of extreme points determines that the number extreme points is zero when a number of extreme points is obtained for each of the small regions within the block which comprises N×

N picture elements with respect to both the peak and valley if a number of small regions in which the number of extreme points is zero is greater than or equal to a predetermined value.

24. The dot region discriminating method as claimed in claim 23 wherein said substep of determining the number of extreme points obtains a sum of the numbers of extreme points of the small regions for the peaks and for the valleys if the number of small regions in which the number of extreme points is zero is less than the predetermined value and determines that the larger sum is the number of extreme points of the block.

25. The dot region discriminating method as claimed in claim 14 wherein said extreme point detecting step uses a rectangular region of the image as said first local region.

26. The dot region discriminating method as claimed in claim 1 wherein said dot region detecting step detects that the center picture element of an object block or the object block itself is a dot region when a number of picture elements which are adjacent to the center picture element and correspond to extreme points is greater than a predetermined threshold value.

27. The dot region discriminating method as claimed in claim 2 wherein said extreme point detecting step detects that a center picture element within the matrix which corresponds to said first local region is an extreme point when a density level of the center picture element within the matrix is a maximum or a minimum compared to density levels of picture elements surrounding the center picture element within the matrix.

28. The dot region discriminating method as claimed in claim 2 wherein said dot region detecting step detects the dot region by moving each block one picture element at a time.

29. The dot region discriminating method as claimed in claim 3 wherein said dot region detecting step detects the dot region by moving each block one picture element at a time.

30. The dot region discriminating method as claimed in claim 4 wherein said dot region detecting step detects the dot region by moving each block one picture element at a time.

31. The dot region discriminating method as claimed in claim 5 wherein said dot region detecting step detects the dot region by moving each block one picture element at a time.

32. The dot region discriminating method as claimed in claim 6 wherein said dot region detecting step detects the dot region by moving each block one picture element at a time.

33. The dot region discriminating method as claimed in claim 2 wherein said dot region detecting step detects the dot region by moving each block one block at a time.

34. The dot region discriminating method as claimed in claim 3 wherein said dot region detecting step detects the dot region by moving each block one block at a time.

35. The dot region discriminating method as claimed in claim 4 wherein said dot region detecting step detects the dot region by moving each block one block at a time.

36. The dot region discriminating method as claimed in claim 5 wherein said dot region detecting step detects the dot region by moving each block one block at a time.

37. The dot region discriminating method as claimed in claim 6 wherein said dot region detecting step detects the dot region by moving each block one block at a time.

38. The dot region discriminating method as claimed in claim 2 wherein said extreme point detecting step uses a rectangular region of the image as said first local region.

39. The dot region discriminating method as claimed in claim 3 wherein said extreme point detecting step uses a rectangular region of the image as said first local region.

40. The dot region discriminating method as claimed in claim 4 wherein said extreme point detecting step uses a rectangular region of the image as said first local region.

41. The dot region discriminating method as claimed in claim 5 wherein said extreme point detecting step uses a rectangular region of the image as said first local region.

42. The dot region discriminating method as claimed in claim 6 wherein said extreme point detecting step uses a rectangular region of the image as said first local region.

43. The dot region discriminating method as claimed in claim 7 wherein said extreme point detecting step uses a rectangular region of the image as said first local region.

44. The dot region discriminating method as claimed in claim 8 wherein said extreme point detecting step uses a rectangular region of the image as said first local region.

45. The dot region discriminating method as claimed in claim 9 wherein said extreme point detecting step uses a rectangular region of the image as said first local region.

46. The dot region discriminating method as claimed in claim 10 wherein said extreme point detecting step uses a rectangular region of the image as said first local region.

47. The dot region discriminating method as claimed in claim 11 wherein said extreme point detecting step uses a rectangular region of the image as said first local region.

48. The dot region discriminating method as claimed in claim 12 wherein said extreme point detecting step uses a rectangular region of the image as said first local region.

49. The dot region discriminating method as claimed in claim 2 which further comprises a smoothing step for smoothing the digital multilevel input signal prior to said extreme point detecting step, said extreme point detecting step detects the one or a plurality of extreme points from said first local region within the image and a smoothened image which is described by a smoothened digital multilevel input image signal, and said dot region detecting step detects the dot region based on a synthesized image which is a logical sum of the image and the smoothened image.

50. The dot region discriminating method as claimed in claim 3 which further comprises a smoothing step for smoothing the digital multilevel input signal prior to said extreme point detecting step, said extreme point detecting step detects the one or a plurality of extreme points from said first local region within the image and a smoothened image which is described by a smoothened digital multilevel input image signal, and said dot region detecting step detects the dot region based on a synthesized image which is a logical sum of the image and the smoothened image.

51. The dot region discriminating method as claimed in claim 4 which further comprises a smoothing step for smoothing the digital multilevel input signal prior to said extreme point detecting step, said extreme point detecting step detects the one or a plurality of extreme points from said first local region within the image and a smoothened image which is described by a smoothened digital multilevel input image signal, and said dot region detecting step detects the dot region based on a synthesized image which is a logical sum of the image and the smoothened image.

52. The dot region discriminating method as claimed in claim 5 which further comprises a smoothing step for smoothing the digital multilevel input signal prior to said extreme point detecting step, said extreme point detecting step detects the one or a plurality of extreme points from said first local region within the image and a smoothened image which is described by a smoothened digital multilevel input image signal, and said dot region detecting step detects the dot region based on a synthesized image which is a logical sum of the image and the smoothened image.

53. The dot region discriminating method as claimed in claim 6 which further comprises a smoothing step for smoothing the digital multilevel input signal prior to said extreme point detecting step, said extreme point detecting step detects the one or a plurality of extreme points from said first local region within the image and a smoothened image which is described by a smoothened digital multilevel input image signal, and said dot region detecting step detects the dot region based on a synthesized image which is a logical sum of the image and the smoothened image.

54. The dot region discriminating method as claimed in claim 7 which further comprises a smoothing step for smoothing the digital multilevel input signal prior to said extreme point detecting steps, said extreme point detecting step detects the one or a plurality of extreme points from said first local region within the image and a smoothened image which is described by a smoothened digital multilevel input image signal, and said dot region detecting step detects the dot region based on a synthesized image which is a logical sum of the image and the smoothened image.

55. The dot region discriminating method as claimed in claim 8 which further comprises a smoothing step for smoothing the digital multilevel input signal prior to said extreme point detecting step, said extreme point detecting step detects the one or a plurality of extreme points from said first local region within the image and a smoothened image which is described by a smoothened digital multilevel input image signal, and said dot region detecting step detects the dot region based on a synthesized image which is a logical sum of the image and the smoothened image.

56. The dot region discriminating method as claimed in claim 9 which further comprises a smoothing step for smoothing the digital multilevel input signal prior to said extreme point detecting step, said extreme point detecting step detects the one or a plurality of extreme points from said first local region within the image and a smoothened image which is described by a smoothened digital multilevel input image signal, and said dot region detecting step detects the dot region based on a synthesized image which is a logical sum of the image and the smoothened image.

57. The dot region discriminating method as claimed in claim 10 which further comprises a smoothing step for smoothing the digital multilevel input signal prior to said extreme point detecting step, said extreme point detecting step detects the one or a plurality of extreme points from said first local region within the image and a smoothened image which is described by a smoothened digital multilevel input image signal, and said dot region detecting step detects the dot region based on a synthesized image which is a logical sum of the image and the smoothened image.

58. The dot region discriminating method as claimed in claim 11 which further comprises a smoothing step for smoothing the digital multilevel input signal prior to said extreme point detecting step, said extreme point detecting step detects the one or a plurality of extreme points from said first local region within the image and a smoothened image which is described by a smoothened digital multilevel input image signal, and said dot region detecting step detects the dot region based on a synthesized image which is a logical sum of the image and the smoothened image.

59. The dot region discriminating method as claimed in claim 12 which further comprises a smoothing step for smoothing the digital multilevel input signal prior to said extreme point detecting step, said extreme point detecting step detects the one or a plurality of extreme points from said first local region within the image and a smoothened image which is described by a smoothened digital multilevel input image signal, and said dot region detecting step detects the dot region based on a synthesized image which is a logical sum of the image and the smoothened image.

60. The dot region discriminating method as claimed in claim 13 which further comprises a smoothing step for smoothing the digital multilevel input signal prior to said extreme point detecting step, said extreme point detecting step detects the one or a plurality of extreme points from said first local region within the image and a smoothened image which is described by a smoothened digital multilevel input image signal, and said dot region detecting step detects the dot region based on a synthesized image which is a logical sum of the image and the smoothened image.

61. The dot region discriminating method as claimed in claim 15 wherein said extreme point detecting step uses a rectangular region of the image as said first local region.

62. The dot region discriminating method as claimed in claim 16 wherein said extreme point detecting step uses a rectangular region of the image as said first local region.

63. The dot region discriminating method as claimed in claim 17 wherein said extreme point detecting step uses a rectangular region of the image as said first local region.

64. The dot region discriminating method as claimed in claim 18 wherein said extreme detecting step uses a rectangular region of the image as said first local region.

65. The dot region discriminating method as claimed in claim 19 wherein said extreme point detecting step uses a rectangular region of the image as said first local region.

66. The dot region discriminating method as claimed in claim 20 wherein said extreme point detecting step uses a rectangular region of the image as said first local region.

67. The dot region discriminating method as claimed in claim 21 wherein said extreme point detecting step uses a rectangular region of the image as said first local region.

68. The dot region discriminating method as claimed in claim 22 wherein said extreme point detecting step uses a rectangular region of the image as said first local region.

69. The dot region discriminating method as claimed in claim 23 wherein said extreme point detecting step uses a rectangular region of the image as said first local region.

70. The dot region discriminating method as claimed in claim 24 wherein said extreme point detecting step uses a rectangular region of the image as said first local region.