CN1340273A - Color image segmentation method - Google Patents

Color image segmentation method Download PDF

Info

Publication number
CN1340273A
CN1340273A CN00803603A CN00803603A CN1340273A CN 1340273 A CN1340273 A CN 1340273A CN 00803603 A CN00803603 A CN 00803603A CN 00803603 A CN00803603 A CN 00803603A CN 1340273 A CN1340273 A CN 1340273A
Authority
CN
China
Prior art keywords
value
color image
segmentation method
image segmentation
pixel
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN00803603A
Other languages
Chinese (zh)
Other versions
CN1292593C (en
Inventor
申铉枓
崔良林
B·S·曼朱纳思
邓忆宁
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Samsung Electronics Co Ltd
University of California
Original Assignee
Samsung Electronics Co Ltd
University of California
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Samsung Electronics Co Ltd, University of California filed Critical Samsung Electronics Co Ltd
Publication of CN1340273A publication Critical patent/CN1340273A/en
Application granted granted Critical
Publication of CN1292593C publication Critical patent/CN1292593C/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/40—Analysis of texture
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N9/00—Details of colour television systems
    • H04N9/64—Circuits for processing colour signals
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10024—Color image

Abstract

A color image segmentation method is provided. The color image segmentation method includes the steps of: (a) calculating a predetermined value representing the degree of difference form the color of peripheral pixels by using pixel values of an input image; (b) obtaining a converted image by converting a calculated value into a value of a predetermined scale; and (c) segmenting the converted image. According to the color image segmentation method, a robust and an automatic segmentation is possible, and a segmentation speed is high even when segmenting an image containing much noise.

Description

Color image segmentation method
Technical field
The present invention relates to a kind of color image segmentation method, more specifically, relate to a kind of color image segmentation method of cutting apart coloured image.
Background technology
Cutting apart of coloured image is the very part and parcel of Digital Image Processing and application thereof.Conventional color images has the problem that is not easy to cut apart the coloured image that comprises texture.And another carries out the conventional color image segmentation method of cutting apart automatically is unsane for the input picture that comprises noise.And another conventional color image segmentation method, being used for the image that the user prepares to cut apart concerning the input picture that comprises noise is that sane image is cut apart, and does not cut apart automatically but carry out, and has therefore spent many times.
Summary of the invention
For addressing the above problem, one object of the present invention is to provide a kind of can cut apart the coloured image that comprises texture automatically, and presents the color image segmentation method of robustness for the input picture that contains noise.
Another object of the present invention is to provide a kind of processing to comprise the color image processing method of color image segmentation method.
Another object of the present invention is to provide a kind of medium, stores the computer program of carrying out color image segmentation method on it.
Therefore, for achieving the above object, according to an aspect of the present invention, provide a kind of color image segmentation method.This color image segmentation method comprises step: (a) calculated for pixel values of use input picture is represented the predetermined value in various degree of peripheral pixel color; (b) image of changing by the value acquisition that calculated value is converted to predetermined ratio (scale); (c) cut apart the image of being changed.
Best, step (c) is cut apart the image of conversion according to the region growing method.
Color image segmentation method is preferably in step (a) before, also comprises step: the represent pixel value that (p-a) pixel value of image is quantized into predetermined quantity; Wherein pixel value is the pixel value that quantizes.
The represent pixel value preferably is made up of 10-20 value.
Color image segmentation method is preferably in step (a) before, also comprises step: (p-a-1) definition comprises the predetermined window of center pixel; And the predetermined value in various degree of (p-a-2) calculating the expression external pixels color relevant with the pixel in predetermined window.
Preferably step (a) comprises step: (a-1) when d is positive integer, definition be centered close to pixel p and have d * d size window B; And (a-2) when i be number between 1 to C, and Z iWhen being the collection of all pixels among the window B, with location of pixels Z iBe categorized into C class; And (a-3) obtain the J-value relevant with each pixel among the class figure, as follows:
J = S B S W = S T - S W S W
Wherein, m iBe at class Z iIn N iThe mean value of the position of data point,
S T = Σ z ∈ Z | | z - m | | 2 And S W = Σ i = 1 c S i = Σ z ∈ Z i | | z - m i | | 2
D is the integer of preferably considering between 3 to 10.
Predetermined ratio preferably has the gray scale of value 0 to 255.
For achieving the above object, according to a further aspect in the invention, provide a kind of color image segmentation method.This color image segmentation method comprises step: the represent pixel value that (a) pixel value of image is quantized into predetermined quantity; (b) use the represent pixel value that quantizes to calculate the predetermined value in various degree of pixel color in the window of the predetermined size of expression; (c) image that obtains to change by the value that calculated value is converted to predetermined ratio; And (d) use the image of cutting apart conversion based on the dividing method of region growing method.
For realizing another purpose, provide a kind of color image processing method that is used for handling the based target of coloured image according to color image segmentation method.This color image segmentation method comprises step: a) calculated for pixel values of use input picture is represented the predetermined value in various degree of peripheral pixel color; (b) image of changing by the value acquisition that calculated value is converted to predetermined ratio (scale); And (c) cut apart the image of being changed.
For realizing another purpose, a kind of medium is provided, be used to store the program code that execution becomes color images the color image segmentation method in a plurality of zones.This color image segmentation method comprises step: the represent pixel value that (a) pixel value of image is quantized into predetermined quantity; (b) use the represent pixel value that quantizes to calculate the predetermined value in various degree of pixel color in the window of the predetermined size of expression; (c) image that obtains to change by the value that calculated value is converted to predetermined ratio; And (d) use the image of cutting apart conversion based on the dividing method of region growing method.
Description of drawings
By below with reference to the detailed description of accompanying drawing to the preferred embodiments of the present invention, above-mentioned purpose of the present invention and advantage will become apparent, wherein:
Fig. 1 is the flow chart that color image segmentation method according to a preferred embodiment of the invention is described;
Fig. 2 A to 2C explanation is according to the class figure (class-map) and the J-value of the color image segmentation method formation of Fig. 1;
The class figure that Fig. 3 A and 3B explanation are cut apart;
A picture frame of " container " of the test pattern that Fig. 4 A explanation is cut apart as a test pattern and by color image segmentation method according to the present invention;
A picture frame of " section chief " of the test pattern that Fig. 4 B explanation explanation is cut apart as a test pattern and by color image segmentation method according to the present invention;
A picture frame at " beach " of the test pattern that Fig. 4 C explanation is cut apart as a test pattern and by color image segmentation method according to the present invention;
A picture frame in " garden " of the test pattern that Fig. 4 D explanation is cut apart as a test pattern and by color image segmentation method according to the present invention;
A picture frame of " mother and the daughter " of the test pattern that Fig. 4 E explanation is cut apart as a test pattern and by color image segmentation method according to the present invention;
Embodiment
With reference to figure 1, the flow chart of color image segmentation method according to a preferred embodiment of the invention wherein has been described, input color image (step 102), and the pixel value of input picture is quantized into several represent pixel values (step 104).For the image in the natural scene of classifying, the represent pixel value is made up of 10-20 value.In the present embodiment, using the Three Represents pixel value to carry out for convenience of description quantizes.Below, form class figure (step 106) by specifying corresponding to the represent pixel value that quantizes.
Best, the regulation window is in the center of aiming in the entire image processed pixel.That is, when d is a positive integer, be preferably between 3 to 10, the center of regulation window B aligned pixel P, and have the size of d * d.And, suppose i be 1 and C between number, and Zi is the collection of all pixels among the window B.In other words, suppose that Zi is divided into C class.The d of best definition window size is the integer of considering between 3 to 10.
And hypothesis m iBe at class Z iIn N iThe mean value of the position of data point, as follows:
(formula 1)
m i = 1 N i Σ z ∈ Z i Z
And, S TAnd S WBe defined as follows respectively:
(formula 2)
S T = Σ z ∈ Z | | z - m i | | 2 And
(formula 3)
S W = Σ i = 1 c Si = Σ z ∈ Z | | z - m i | | 2
Obtain the J-value (step 108) relevant below with each pixel among the class figure.The J-value defined relevant with each pixel among the class figure is as follows:
(formula 4)
J = S B S W = S T - S W S W
The J-value that is obtained by formula 4 is converted into the gray value between 0 to 255, has the gray scales image (step 110) that these values also can be known as the J-image so that obtain.The J-image has as comprising that reality represents the same form of the three-dimensional land map on the mountain valley of regional center and zone boundary and mountain peak respectively.
Preferably cut apart J-image (step 112) according to the region growing method.The region growing method is well known to those of ordinary skill in the art as the method for cutting apart that is used for digital picture, therefore no longer provides its explanation.
Fig. 2 A to 2C explanation is according to the class figure and the J-value of the color image segmentation method formation of Fig. 1.In the class figure of Fig. 2 A, the J-value of center pixel is 1.720, and in the class figure of Fig. 2 B, the J-value of center pixel is 0, and in the class figure of Fig. 2 C, the J-value of center pixel obtains as 0.855.Here,, be positioned at the pixel of the usefulness on the left side of center pixel+represent, be positioned at the pixel that the usefulness 0 on the right of center pixel represents and the right that is positioned at center pixel and use as in the class figure of Fig. 2 A *The pixel of expression roughly clearly forms under the situation in zone, and the J-value is a big relatively value 1.720.And as in the class figure of Fig. 2 B, the pixel of use+representing and with 0 pixel of representing and usefulness *The pixel of expression is even distribution, and is difficult to form under the situation in zone, and the J-value is 0.And, be positioned at the usefulness on the right of center pixel as in the class figure of Fig. 2 C *The pixel of expression forms the zone, but is difficult under the regional situation of formation with the+pixel represented and with 0 pixel of representing, the J-value is 0.855.That is, bigger J-value is the situation on pixel-by-pixel basis near field border most probably, therefore, uses this point can carry out cutting apart based on the region growing method.
Fig. 3 A and 3B have illustrated the class figure of cutting apart.
Must check whether to have carried out well and cut apart, and whether represent the value identical with quantized value for each zone among the class figure of cutting apart.For this purpose, work as J kBe the J-value that obtains corresponding to k-zone, and Mk is k regional pixel number, and N is when being pixel number among the class figure total, average J-value is calculated as follows:
(formula 5)
J = 1 N Σ K M K J K
Whether the value of being calculated is represented as quantized value and regardless of having carried out well for each zone among the class figure of cutting apart cuts apart.
In the situation of the class figure of cutting apart shown in Fig. 3 A, J is 0, and on the other hand, in the situation of the class figure of cutting apart shown in Fig. 3 B, J is 0.05.That is, under the regional situation of fixed qty, special under situation about cutting apart preferably, average J-value is little.This situation occurs and be because when well being cut apart in the zone, the zone comprises equally distributed color classification seldom, and therefore, average J-value is little.
A picture frame of " container " of the test pattern that Fig. 4 A explanation is cut apart as a test pattern and by color image segmentation method according to the present invention.With reference to figure 4A, the J of image is 0.232 before cutting apart, but the J after cutting apart is 0.071.And it is the illustration that test pattern is cut apart well.
A picture frame of " section chief " of the test pattern that Fig. 4 B explanation explanation is cut apart as a test pattern and by color image segmentation method according to the present invention.With reference to figure 4B, the J of image is 0.238 before cutting apart, but the J after cutting apart is 0.105.And it is the illustration that test pattern is cut apart well.
A picture frame at " beach " of the test pattern that Fig. 4 C explanation is cut apart as a test pattern and by color image segmentation method according to the present invention.With reference to figure 4C, the J of image is 0.494 before cutting apart, but the J after cutting apart is 0.093.And it is the illustration that test pattern is cut apart well.
A picture frame in " garden " of the test pattern that Fig. 4 D explanation is cut apart as a test pattern and by color image segmentation method according to the present invention.With reference to figure 4D, the J of image is 0.435 before cutting apart, but the J after cutting apart is 0.088.And it is the illustration that test pattern is cut apart well.
A picture frame of " mother and the daughter " of the test pattern that Fig. 4 E explanation is cut apart as a test pattern and by color image segmentation method according to the present invention.With reference to figure 4E, the J of image is 0.438 before cutting apart, but the J after cutting apart is 0.061.And it is the illustration that test pattern is cut apart well.
That is, described as reference Fig. 4 A to 4E, littler according to the J of color image segmentation method split image of the present invention than the J of the image before cutting apart.
In the above-described embodiments, explained the calculating of specific function as an example.Be not limited to this embodiment in appended scope of the present invention, and, can use the function of the other variation in various degree of the color of representing peripheral pixel obviously for those of ordinary skills.
And above-mentioned color image segmentation method can be realized with computer program.The code of coding and code segment are easily under the technical staff's of computer realm help.And this program can be stored in the computer-readable medium, can be read and carry out by computer, so it can realize color image processing method.Medium comprises magnetizing mediums, light medium and carrier wave.
As mentioned above, according to the present invention, coloured image can automatically be cut apart under the assistance that does not have the user, and is sane for the input picture that contains noise.
Utilizability on the industry
Above-mentioned according to color image segmentation method of the present invention in, a kind of sane cutting apart be possible namely Make is when cutting apart the image that contains many noises or texture. And, do not having such as being carried out by hand by the user The user of cutting apart assists down, and auto Segmentation also is possible, so splitting speed is high. This cromogram Can be applied to such as during employed object-based image is processed among the MPEG-7 as dividing method.

Claims (41)

1. a color image segmentation method is used for color images is become a plurality of zones, comprises step:
(a) calculated for pixel values of use input picture is represented the predetermined value in various degree of peripheral pixel color;
(b) image of changing by the value acquisition that calculated value is converted to predetermined ratio;
(c) cut apart the image of being changed.
2. color image segmentation method as claimed in claim 1,
Wherein step (c) is cut apart the image of this conversion according to the region growing method.
3. at least as the color image segmentation method of one of claim 1 or claim 2, in step (a) before, also comprise step: the represent pixel value that (p-a) pixel value of image is quantized into predetermined quantity; Wherein pixel value is the pixel value that quantizes.
4. color image segmentation method as claimed in claim 3, wherein the represent pixel value is made up of 10-20 value.
5. as the color image segmentation method of at least one claim in claim 1 or claim 2 or the claim 4, in step (a) before, also comprise step:
(p-a-1) definition comprises the predetermined window of center pixel; And
(p-a-2) calculate the predetermined value in various degree of representing the external pixels color relevant with the pixel in the regulation window.
6. color image segmentation method as claimed in claim 3 in step (a) before, also comprises step:
(p-a-1) definition comprises the predetermined window of center pixel; And
(p-a-2) calculate the predetermined value in various degree of representing the external pixels color relevant with the pixel in the regulation window.
7. as the color image segmentation method of at least one claim in claim 1 or the claim 2, wherein step (a) comprises step:
(a-1) when d is positive integer, the definition be centered close to pixel p and have d * d size window B; And
(a-2) when i be number between 1 to C, and Z iWhen being the collection of all pixels among the window B, with location of pixels Z iBe categorized into C class; And
(a-3) obtain the J-value relevant with each pixel among the class figure, as follows:
J = S B S W = S T - S W S W
Wherein, m iBe at class Z iIn N iThe mean value of the position of data point,
S T = Σ z ∈ Z | | z - m | | 2 And S W = Σ i = 1 c S i = Σ z ∈ Z i | | z - m i | | 2
8. color image segmentation method as claimed in claim 3, wherein step (a) comprises step:
(a-1) when d is positive integer, the definition be centered close to pixel p and have d * d size window B; And
(a-2) when i be number between 1 to C, and Z iWhen being the collection of all pixels among the window B, with location of pixels Z iBe categorized into C class; And
(a-3) obtain the J-value relevant with each pixel among the class figure, as follows:
J = S B S W = S T - S W S W
Wherein, m iBe at class Z iIn N iThe mean value of the position of data point,
S T = Σ z ∈ Z | | z - m | | 2 And S W = Σ i = 1 c S i = Σ z ∈ Z i | | z - m i | | 2
9. color image segmentation method as claimed in claim 4, wherein step (a) comprises step:
(a-1) when d is positive integer, the definition be centered close to pixel p and have d * d size window B; And
(a-2) when i be number between 1 to C, and Z iWhen being the collection of all pixels among the window B, with location of pixels Z iBe categorized into C class; And
(a-3) obtain the J-value relevant with each pixel among the class figure, as follows:
J = S B S W = S T - S W S W
Wherein, m iBe at class Z iIn N iThe mean value of the position of data point,
S T = Σ z ∈ Z | | z - m | | 2 And S W = Σ i = 1 c S i = Σ z ∈ Z i | | z - m i | | 2
10. color image segmentation method as claimed in claim 5, wherein step (a) comprises step:
(a-1) when d is positive integer, the definition be centered close to pixel p and have d * d size window B; And
(a-2) when i be number between 1 to C, and Z iWhen being the collection of all pixels among the window B, with location of pixels Z iBe categorized into C class; And
(a-3) obtain the J-value relevant with each pixel among the class figure, as follows:
J = S B S W = S T - S W S W
Wherein, m iBe at class Z iIn N iThe mean value of the position of data point,
S T = Σ z ∈ Z | | z - m | | 2 And S W = Σ i = 1 c S I = Σ z ∈ Z i | | z - m i | | 2
11. color image segmentation method as claimed in claim 6, wherein step (a) comprises step:
(a-1) when d is positive integer, the definition be centered close to pixel p and have d * d size window B; And
(a-2) when i be number between 1 to C, and Z iWhen being the collection of all pixels among the window B, with location of pixels Z iBe categorized into C class; And
(a-3) obtain the J-value relevant with each pixel among the class figure, as follows:
J = S B S W = S T - S W S W
Wherein, m iBe at class Z iIn N iThe mean value of the position of data point,
S T = Σ z ∈ Z | | z - m | | 2 And S W = Σ i = 1 c S i = Σ z ∈ Z i | | z - m i | | 2
12. color image segmentation method as claimed in claim 7, wherein d is the integer of considering between 3 to 10.
13. color image segmentation method as claimed in claim 8, wherein d is the integer of considering between 3 to 10.
14. color image segmentation method as claimed in claim 9, wherein d is the integer of considering between 3 to 10.
15. as the color image segmentation method of claim 10, wherein d is the integer of considering between 3 to 10.
16. as the color image segmentation method of claim 11, wherein d is the integer of considering between 3 to 10.
17. as the color image segmentation method of at least one claim in claim 1 or the claim 2, wherein predetermined ratio preferably has the gray scale of value 0 to 255.
18. color image segmentation method as claimed in claim 3, wherein predetermined ratio is the gray scale with value 0 to 255.
19. color image segmentation method as claimed in claim 4, wherein predetermined ratio is the gray scale with value 0 to 255.
20. color image segmentation method as claimed in claim 5, wherein predetermined ratio is the gray scale with value 0 to 255.
21. color image segmentation method as claimed in claim 6, wherein predetermined ratio is the gray scale with value 0 to 255.
22. color image segmentation method as claimed in claim 7, wherein predetermined ratio is the gray scale with value 0 to 255.
23. color image segmentation method as claimed in claim 8, wherein predetermined ratio is the gray scale with value 0 to 255.
24. color image segmentation method as claimed in claim 9, wherein predetermined ratio is the gray scale with value 0 to 255.
25 color image segmentation methods as claim 10, wherein predetermined ratio is the gray scale with value 0 to 255.
26. as the color image segmentation method of claim 11, wherein predetermined ratio is the gray scale with value 0 to 255.
27. as the color image segmentation method of claim 12, wherein predetermined ratio is the gray scale with value 0 to 255.
28. as the color image segmentation method of claim 13, wherein predetermined ratio is the gray scale with value 0 to 255.
29. as the color image segmentation method of claim 14, wherein predetermined ratio is the gray scale with value 0 to 255.
30. as the color image segmentation method of claim 15, wherein predetermined ratio is the gray scale with value 0 to 255.
31. as the color image segmentation method of claim 16, wherein predetermined ratio is the gray scale with value 0 to 255.
32. an object-based color image processing method is used to handle the coloured image according to color image segmentation method, wherein this color image segmentation method comprises step:
(a) calculated for pixel values of use input picture is represented the predetermined value in various degree of peripheral pixel color;
(b) image of changing by the value acquisition that calculated value is converted to predetermined ratio;
(c) cut apart the image of being changed.
33. as the color image processing method of claim 32, wherein this color image processing method meets the MPEG-7 standard.
34. a color image segmentation method is used for color images is become a plurality of zones, comprises step:
(a) pixel value of image is quantized into the represent pixel value of predetermined quantity;
(b) use the represent pixel value that quantizes to calculate the predetermined value in various degree of pixel color in the window of the predetermined size of expression;
(c) image that obtains to change by the value that calculated value is converted to predetermined ratio; And
(d) use the image of cutting apart conversion based on the dividing method of region growing method.
35. as the color image segmentation method of claim 34, wherein step (a) comprises step:
(a-1) when d is positive integer, the definition be centered close to pixel p and have d * d size window B; And
(a-2) when i be number between 1 to C, and Z iWhen being the collection of all pixels among the window B, with location of pixels Z iBe categorized into C class; And
(a-3) obtain the J-value relevant with each pixel among the class figure, as follows:
J = S B S W = S T - S W S W
Wherein, m iBe at class Z iIn N iThe mean value of the position of data point,
S T = Σ z ∈ Z | | z - m | | 2 And S W = Σ i = 1 c S I = Σ z ∈ Z i | | z - m i | | 2
36. as the color image segmentation method of claim 35, wherein d is the integer of considering between 3 to 10.
37. as the color image segmentation method of claim 34 to claim 36, wherein predetermined ratio is the gray scale with value 0 to 255.
38. a medium that is used to store the program code of the color image segmentation method that execution becomes color images in a plurality of zones, this color image segmentation method comprises step:
(a) pixel value of image is quantized into the represent pixel value of predetermined quantity;
(b) use the represent pixel value that quantizes to calculate the predetermined value in various degree of pixel color in the window of the predetermined size of expression;
(c) image that obtains to change by the value that calculated value is converted to predetermined ratio; And
(d) use the image of cutting apart conversion based on the dividing method of region growing method.
39. as the color image segmentation method of claim 38, wherein step (a) comprises step:
(a-1) when d is positive integer, the definition be centered close to pixel p and have d * d size window B; And
(a-2) when i be number between 1 to C, and Z iWhen being the collection of all pixels among the window B, with location of pixels Z iBe categorized into C class; And
(a-3) obtain the J-value relevant with each pixel among the class figure, as follows:
J = S B S W = S T - S W S W
Wherein, m iBe at class Z iIn N iThe mean value of the position of data point,
S T = Σ z ∈ Z | | z - m | | 2 And S W = Σ i = 1 c S i = Σ z ∈ Z i | | z - m i | | 2
40. as the color image segmentation method of claim 39, wherein d is the integer of considering between 3 to 10.
41. as the color image segmentation method of claim 38 to claim 40, wherein predetermined ratio is the gray scale with value 0 to 255.
CNB008036039A 1999-04-23 2000-03-22 Color image segmentation method Expired - Fee Related CN1292593C (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US13064399P 1999-04-23 1999-04-23
US60/130,643 1999-04-23

Publications (2)

Publication Number Publication Date
CN1340273A true CN1340273A (en) 2002-03-13
CN1292593C CN1292593C (en) 2006-12-27

Family

ID=22445651

Family Applications (1)

Application Number Title Priority Date Filing Date
CNB008036039A Expired - Fee Related CN1292593C (en) 1999-04-23 2000-03-22 Color image segmentation method

Country Status (8)

Country Link
EP (1) EP1192809A4 (en)
JP (1) JP3853156B2 (en)
KR (1) KR100436499B1 (en)
CN (1) CN1292593C (en)
AU (1) AU3680700A (en)
MY (1) MY130272A (en)
TW (1) TW469412B (en)
WO (1) WO2000065839A1 (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102629386A (en) * 2012-03-28 2012-08-08 浙江大学 Region segmentation method for colorful textile texture images
CN102087741B (en) * 2009-12-03 2013-01-02 财团法人工业技术研究院 Method and system for processing image by using regional architecture
CN103065317A (en) * 2012-12-28 2013-04-24 中山大学 Partial color transferring method and transferring device based on color classification

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6778697B1 (en) * 1999-02-05 2004-08-17 Samsung Electronics Co., Ltd. Color image processing method and apparatus thereof
CN100530231C (en) * 1999-02-05 2009-08-19 三星电子株式会社 Method for processing colour image
KR100788643B1 (en) 2001-01-09 2007-12-26 삼성전자주식회사 Searching method of image based on combination of color and texture
KR100562937B1 (en) * 2004-08-11 2006-03-22 엘지전자 주식회사 Method and Apparatus for Processing Image of Display Device
KR101223046B1 (en) * 2011-02-08 2013-01-17 경북대학교 산학협력단 Image segmentation device and method based on sequential frame imagery of a static scene
KR101631953B1 (en) * 2014-12-09 2016-06-20 삼성전자주식회사 Method for processing image for blur region detection and image processing apparatus thereof

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100488422B1 (en) * 1996-09-24 2005-09-02 주식회사 팬택앤큐리텔 Grayscale-shaped information encoding / decoding device and method

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102087741B (en) * 2009-12-03 2013-01-02 财团法人工业技术研究院 Method and system for processing image by using regional architecture
CN102629386A (en) * 2012-03-28 2012-08-08 浙江大学 Region segmentation method for colorful textile texture images
CN103065317A (en) * 2012-12-28 2013-04-24 中山大学 Partial color transferring method and transferring device based on color classification

Also Published As

Publication number Publication date
JP2002543691A (en) 2002-12-17
CN1292593C (en) 2006-12-27
TW469412B (en) 2001-12-21
MY130272A (en) 2007-06-29
EP1192809A1 (en) 2002-04-03
KR20010105382A (en) 2001-11-28
AU3680700A (en) 2000-11-10
WO2000065839A1 (en) 2000-11-02
EP1192809A4 (en) 2007-08-22
JP3853156B2 (en) 2006-12-06
KR100436499B1 (en) 2004-06-22

Similar Documents

Publication Publication Date Title
CN1301493C (en) Coding and decoding methods for three-dimensional object data, and device for the methods
CN1207924C (en) Method for testing face by image
CN1292593C (en) Color image segmentation method
EP1355484A3 (en) Image compression method and apparatus, and image coding method and apparatus
CN1633159A (en) A method for removing image noise
CN1470131A (en) Non-linear quantization and similarity matching methods for retrieving image data
CN1577392A (en) Method and device for measuring visual similarity
CN1401106A (en) Image processor and pattern recognition apparatus using the image processor
CN1282109C (en) Image research method and device not affected by light change
CN1773553A (en) Image encoding with dynamic buffer capacity level based compression adjustment
CN1687969A (en) File image compressing method based on file image content analyzing and characteristic extracting
CN1324530C (en) Digital video texture analytic method
CN1495670A (en) Method and device for processing colour image
CN1209735C (en) Color image processing method and apparatus thereof
CN1720550A (en) Method and apparatus for removing false edges from a segmented image
CN101030230A (en) Image searching method and system
CN1165859C (en) Muctilevel image grid data structure and image search method using the same
CN1201333A (en) Interpolation method for binary picture
CN1525387A (en) Device and method for detecting blurring of image
CN1151678C (en) Method and apparatus for encoding contour image of object in video signal
CN1498493A (en) Method for compression and decompression of image data
CN1650611A (en) Image processing device and image processing method
CN1916938A (en) Identifying distance regulator and method thereof and text lines identifier and method thereof
CN1526122A (en) Mesh compression
CN1279491C (en) Image analysis

Legal Events

Date Code Title Description
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C06 Publication
PB01 Publication
C14 Grant of patent or utility model
GR01 Patent grant
C17 Cessation of patent right
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20061227

Termination date: 20100322