CN108711156A - A kind of image segmentation Complexity Measurement method - Google Patents
A kind of image segmentation Complexity Measurement method Download PDFInfo
- Publication number
- CN108711156A CN108711156A CN201810442469.9A CN201810442469A CN108711156A CN 108711156 A CN108711156 A CN 108711156A CN 201810442469 A CN201810442469 A CN 201810442469A CN 108711156 A CN108711156 A CN 108711156A
- Authority
- CN
- China
- Prior art keywords
- image
- region
- size
- segmentation
- divisible
- 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.)
- Pending
Links
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
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Facsimile Image Signal Circuits (AREA)
- Image Analysis (AREA)
- Character Input (AREA)
Abstract
The invention discloses a kind of image segmentation Complexity Measurement methods, include the following steps:Step 1, for given coloured image, according to the histogram of its color shade component, initial segmentation is carried out to image;Step 2, (0) setting regions size N calculates the comentropy corresponding to the divisible region of image less than N (0);Step 3, image segmentation necessity index Ps is calculated;Step 4, it calculates image segmentation necessity threshold value Ps (N (0)) and judges whether image needs to divide by threshold value Ps (N (0)) according to image segmentation necessity index Ps.
Description
Technical field
The present invention relates to a kind of image segmentation Complexity Measurement methods.
Background technology
Course after decades of development, for image segmentation algorithm correlative study occur thousands of kinds it is different types of
Algorithm.For normal conditions, image partition method can be divided into 3 classes:(1) partitioning algorithm based on region;(2) global knowledge threshold
Value algorithm;(3) partitioning algorithm based on edge.
However, major part image segmentation algorithm can only all work for a certain specific condition at present, and cannot be the more
Kind image carries out versatility segmentation.Meanwhile it is a problem to be solved that the standard of partitioning algorithm selection, which is remained on,.About
The documents and materials of image segmentation necessity are up to the present still less, even if some documents are involved in the problems, such as this aspect
Do not studied in depth.In document " Zhang K D, Lu H Q, Duan M Y, et al.Automatic Salient
Regions of Interest Extraction Based on Edge and Region Integration[C]//
Proceedings of 2006IEEE International Sympoium on Industrual
Electronics.Montreal,Quebec,Canada:IEEE Industrial Electronics Society,2006,
In l ", author has used a kind of previously given threshold value to express the necessity condition of image segmentation, however, these threshold values are really
The method of determining is that this document does not provide.Therefore, the necessity research of image segmentation still needs to solve there are many problem.
Invention content
In view of the deficiencies of the prior art, the present invention provides a kind of image segmentation Complexity Measurement methods, including walk as follows
Suddenly:
Step 1, for given coloured image, according to the histogram of its color shade component, initial segmentation is carried out to image;
Step 2, (0) setting regions size N calculates the comentropy corresponding to the divisible region of image less than N (0);
Step 3, image segmentation necessity index Ps is calculated;
Step 4, image segmentation necessity threshold value Ps (N (0)) is calculated, according to image segmentation necessity index Ps, by this
Threshold value Ps (N (0)) judges whether image needs to divide.
In step 1, initial segmentation (bibliography is carried out to image using the multi-threshold segmentation method based on histogram:Base
In the multi-threshold segmentation method of grey level histogram, Liu Xinxin, Li Xue, Wang Qiong,《Computer application and software》, 2013).
Step 2 includes:
Step 2-1, the region that setting procedure 1 divides image initial are expressed as { R (i) |I ∈ { 1,2 ... K } }, wherein K
For the alienable areal of image, R (i) indicates that i-th of alienable region of image, i-th of alienable region are corresponding
Size is denoted as N (i);
Step 2-2 is ranked up the alienable region of image is descending, with { R (n, i) |N=1,2 ... K } it indicates
The divisible region of image after sequence, wherein R (n, i) indicates n-th of divisible region after sequence, and the area of R (n, i)
Domain is more than the region of R (n+1, i), and the corresponding area sizes of R (n, i) are denoted as N (n, i);
Step 2-3, setting regions size N (0) calculate the comentropy corresponding to the divisible region of image less than N (0)
SN(0)(K) and the divisible area size of image less than N (0) accounts for the percentage H of image sizeN(0)(K)。
In step 2-3, the comentropy S corresponding to the divisible region of image less than N (0) is calculated by following formulaN(0)
(K) and the divisible area size of image less than N (0) accounts for the percentage H of image sizeN(0)(K):
Wherein, N indicates the size of the given coloured image of step 1.
In step 3, image segmentation necessity index Ps is calculated by following formula:
Step 4 includes:
Step 4-1 passes through if setting regions size N (0) is equal to the size N (n, i) in the maximum divisible region of image
Following formula calculates image segmentation necessity threshold value Ps (N (0)):
If setting regions size N (0) much smaller than image size N (such as setting regions size N (0) only have N ten/
One) image segmentation necessity threshold value Ps (N (0)), is calculated by following formula:
Step 4-2 judges if image segmentation necessity index Ps is less than image segmentation necessity threshold value Ps (N (0))
Image does not have to segmentation.
Advantageous effect:A kind of image segmentation Complexity Measurement method provided by the invention, can be well adapted for image mesh
Area size is marked, is to weigh a kind of rationally effective method of image segmentation necessity, and a kind of important image can be become
Divide judge index.
Description of the drawings
The present invention is done with reference to the accompanying drawings and detailed description and is further illustrated, it is of the invention above-mentioned or
Otherwise advantage will become apparent.
Fig. 1 is flow chart of the present invention.
Specific implementation mode
The present invention will be further described with reference to the accompanying drawings and embodiments.
As shown in Figure 1, the present invention provides a kind of image segmentation Complexity Measurement method, include the following steps:
Step 1, for given coloured image, according to the histogram of its color shade component, initial segmentation is carried out to image;
Step 2, (0) setting regions size N calculates the comentropy corresponding to the divisible region of image less than N (0);
Step 3, image segmentation necessity index Ps is calculated;
Step 4, image segmentation necessity threshold value Ps (N (0)) is calculated, according to image segmentation necessity index Ps, by this
Threshold value Ps (N (0)) judges whether image needs to divide.
In step 1, initial segmentation (bibliography is carried out to image using the multi-threshold segmentation method based on histogram:Base
In the multi-threshold segmentation method of grey level histogram, Liu Xinxin, Li Xue, Wang Qiong,《Computer application and software》, 2013).
Step 2 includes:
Step 2-1, the region that setting procedure 1 divides image initial are expressed as { R (i) |I ∈ { 1,2 ... K } }, wherein K
For the alienable areal of image, R (i) indicates that i-th of alienable region of image, i-th of alienable region are corresponding
Size is denoted as N (i);
Step 2-2 is ranked up the alienable region of image is descending, with { R (n, i) |N=1,2 ... K } it indicates
The divisible region of image after sequence, wherein R (n, i) indicates n-th of divisible region after sequence, and the area of R (n, i)
Domain is more than the region of R (n+1, i), and the corresponding area sizes of R (n, i) are denoted as N (n, i);
Step 2-3, setting regions size N (0) calculate the comentropy corresponding to the divisible region of image less than N (0)
SN(0)(K) and the divisible area size of image less than N (0) accounts for the percentage H of image sizeN(0)(K)。
In step 2-3, the comentropy S corresponding to the divisible region of image less than N (0) is calculated by following formulaN(0)
(K) and the divisible area size of image less than N (0) accounts for the percentage H of image sizeN(0)(K):
Wherein, N indicates the size of the given coloured image of step 1.
In step 3, image segmentation necessity index Ps is calculated by following formula:
Step 4 includes:
Step 4-1 passes through if setting regions size N (0) is equal to the size N (n, i) in the maximum divisible region of image
Following formula calculates image segmentation necessity threshold value Ps (N (0)):
If setting regions size N (0) much smaller than image size N (such as setting regions size N (0) only have N ten/
One) image segmentation necessity threshold value Ps (N (0)), is calculated by following formula:
Step 4-2 judges if image segmentation necessity index Ps is less than image segmentation necessity threshold value Ps (N (0))
Image does not have to segmentation.
The present invention provides a kind of image segmentation Complexity Measurement methods, implement the method and approach of the technical solution
Very much, the above is only a preferred embodiment of the present invention, it is noted that those skilled in the art are come
It says, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications also should be regarded as
Protection scope of the present invention.All undefined components in this embodiment can be implemented in the prior art.
Claims (6)
1. a kind of image segmentation Complexity Measurement method, which is characterized in that include the following steps:
Step 1, for given coloured image, according to the histogram of its color shade component, initial segmentation is carried out to image;
Step 2, (0) setting regions size N calculates the comentropy corresponding to the divisible region of image less than N (0);
Step 3, image segmentation necessity index Ps is calculated;
Step 4, it calculates image segmentation necessity threshold value Ps (N (0)) and the threshold value is passed through according to image segmentation necessity index Ps
Ps (N (0)) judges whether image needs to divide.
2. according to the method described in claim 1, it is characterized in that, in step 1, using the multi-threshold segmentation side based on histogram
Method carries out initial segmentation to image.
3. according to the method described in claim 2, it is characterized in that, step 2 includes:
Step 2-1, the region that setting procedure 1 divides image initial are expressed as { R (i) |I ∈ { 1,2 ... K } }, wherein K is figure
As alienable areal, R (i) indicates image i-th of alienable region, the corresponding size in i-th of alienable region
It is denoted as N (i);
Step 2-2 is ranked up the alienable region of image is descending, with { R (n, i) |N=1,2 ... K } indicate sequence
The divisible region of image afterwards, wherein R (n, i) indicates n-th of divisible region after sequence, and the region of R (n, i) is big
In the region of R (n+1, i), the corresponding area sizes of R (n, i) are denoted as N (n, i);
Step 2-3, setting regions size N (0) calculate the comentropy S corresponding to the divisible region of image less than N (0)N(0)(K)
And the divisible area size of image less than N (0) accounts for the percentage H of image sizeN(0)(K)。
4. according to the method described in claim 3, it is characterized in that, in step 2-3, calculated less than N's (0) by following formula
Comentropy S corresponding to the divisible region of imageN(0)(K) and the divisible area size of image less than N (0) accounts for image size
Percentage HN(0)(K):
Wherein, N indicates the size of the given coloured image of step 1.
5. according to the method described in claim 4, it is characterized in that, in step 3, it is necessary that image segmentation is calculated by following formula
Property index Ps:
6. according to the method described in claim 5, it is characterized in that, step 4 includes:
Step 4-1, if setting regions size N (0) is equal to the size N (n, i) in the maximum divisible region of image, by as follows
Formula calculates image segmentation necessity threshold value Ps (N (0)):
If setting regions size N (0) is much smaller than image size N, image segmentation necessity threshold value Ps is calculated by following formula
(N(0)):
Step 4-2 judges image if image segmentation necessity index Ps is less than image segmentation necessity threshold value Ps (N (0))
Without segmentation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810442469.9A CN108711156A (en) | 2018-05-10 | 2018-05-10 | A kind of image segmentation Complexity Measurement method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810442469.9A CN108711156A (en) | 2018-05-10 | 2018-05-10 | A kind of image segmentation Complexity Measurement method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108711156A true CN108711156A (en) | 2018-10-26 |
Family
ID=63867895
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810442469.9A Pending CN108711156A (en) | 2018-05-10 | 2018-05-10 | A kind of image segmentation Complexity Measurement method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108711156A (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9095273B2 (en) * | 2011-09-26 | 2015-08-04 | Sunnybrook Research Institute | Systems and methods for automated dynamic contrast enhancement imaging |
CN106228555A (en) * | 2016-07-22 | 2016-12-14 | 湖南文理学院 | Thresholding Method for Grey Image Segmentation based on Masi entropy measure |
CN106778633A (en) * | 2016-12-19 | 2017-05-31 | 江苏慧眼数据科技股份有限公司 | A kind of pedestrian recognition method based on region segmentation |
CN108009542A (en) * | 2017-11-01 | 2018-05-08 | 华中农业大学 | Weed images dividing method under rape field environment |
-
2018
- 2018-05-10 CN CN201810442469.9A patent/CN108711156A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9095273B2 (en) * | 2011-09-26 | 2015-08-04 | Sunnybrook Research Institute | Systems and methods for automated dynamic contrast enhancement imaging |
CN106228555A (en) * | 2016-07-22 | 2016-12-14 | 湖南文理学院 | Thresholding Method for Grey Image Segmentation based on Masi entropy measure |
CN106778633A (en) * | 2016-12-19 | 2017-05-31 | 江苏慧眼数据科技股份有限公司 | A kind of pedestrian recognition method based on region segmentation |
CN108009542A (en) * | 2017-11-01 | 2018-05-08 | 华中农业大学 | Weed images dividing method under rape field environment |
Non-Patent Citations (1)
Title |
---|
魏庆 等: "图像分割复杂性测度研究", 《计算机科学》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Hartig | Basic image analysis and manipulation in ImageJ | |
US10641088B2 (en) | Method and device for determining karst development degree of reservoir, computer readable storage medium and device | |
DE102012109959B4 (en) | Automatic enlargement and selection confirmation | |
WO2016001223A9 (en) | Detecting edges of a nucleus using image analysis | |
WO2007148284A3 (en) | A method, a system and a computer program for determining a threshold in an image comprising image values | |
RU2008102965A (en) | MIXING IMAGES | |
Alijani et al. | Envy-free mechanisms with minimum number of cuts | |
Sahari Moghaddam et al. | Estimating stripping of asphalt coating using k-means clustering and machine learning–based classification | |
CN108510499A (en) | A kind of carrying out image threshold segmentation method and device based on fuzzy set and Otsu | |
CN102800051A (en) | Self-adaptation flotation froth image enhancement method based on fractional order differential | |
JP2019067235A5 (en) | ||
CN106875359B (en) | A kind of sample block image repair method based on layering boot policy | |
Zhang et al. | Extracting windows registry information from physical memory | |
Kim et al. | Correlation-based multi-shape granulometry with application in porous silicon nanomaterial characterization | |
CN108711156A (en) | A kind of image segmentation Complexity Measurement method | |
CN107133462A (en) | Data processing method, device and equipment | |
JP2018185265A (en) | Information processor, method for control, and program | |
Zarella et al. | A template matching model for nuclear segmentation in digital images of H&E stained slides | |
CN104299195A (en) | Self-adaptive dual-threshold and dark channel prior image defogging method based on | |
CN102800090A (en) | Blood cell segmentation method | |
JP2009205559A (en) | Road surface marking image processor, road surface marking image processing method, and road surface marking image processing program | |
JP6239966B2 (en) | Wear amount estimation method and wear amount estimation system | |
TW201508536A (en) | Method and device for acquiring virus signature and machine-readable storage medium | |
CN103761525A (en) | Micro-CT digital image processing-based component detection method of cathode carbon block for aluminium electrolysis | |
Shao et al. | An adaptive image contrast enhancement algorithm based on retinex |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181026 |
|
RJ01 | Rejection of invention patent application after publication |