CN109658382B - Tongue positioning method based on image clustering and gray projection - Google Patents
Tongue positioning method based on image clustering and gray projection Download PDFInfo
- Publication number
- CN109658382B CN109658382B CN201811365181.2A CN201811365181A CN109658382B CN 109658382 B CN109658382 B CN 109658382B CN 201811365181 A CN201811365181 A CN 201811365181A CN 109658382 B CN109658382 B CN 109658382B
- Authority
- CN
- China
- Prior art keywords
- tongue
- image
- positioning
- value
- pixel point
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 37
- 230000011218 segmentation Effects 0.000 claims abstract description 39
- 238000000605 extraction Methods 0.000 claims abstract description 14
- 230000000877 morphologic effect Effects 0.000 claims abstract description 11
- 238000012545 processing Methods 0.000 claims abstract description 6
- 210000000746 body region Anatomy 0.000 claims description 9
- 238000013507 mapping Methods 0.000 claims description 8
- 238000005516 engineering process Methods 0.000 claims description 5
- 230000001174 ascending effect Effects 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 12
- 238000003745 diagnosis Methods 0.000 description 10
- 238000002474 experimental method Methods 0.000 description 8
- 239000003814 drug Substances 0.000 description 7
- 230000004807 localization Effects 0.000 description 5
- 239000011248 coating agent Substances 0.000 description 2
- 238000000576 coating method Methods 0.000 description 2
- 238000003709 image segmentation Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 230000037237 body shape Effects 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000003711 image thresholding Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000036285 pathological change Effects 0.000 description 1
- 231100000915 pathological change Toxicity 0.000 description 1
- 230000001575 pathological effect Effects 0.000 description 1
- 230000007170 pathology Effects 0.000 description 1
- 230000035790 physiological processes and functions Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000037303 wrinkles Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a tongue positioning method based on image clustering and gray projection. Firstly, clustering hue components in an HSI color space to determine three segmentation thresholds, then carrying out image threshold segmentation processing and morphological operation on the hue components to generate a binary image, taking the largest target area in the binary image as a tongue area coarse extraction result, and finally carrying out gray projection on a red component of the tongue image to carry out line and row positioning of the tongue, thereby obtaining a tongue positioning result. The invention has good tongue positioning effect, and the method is simple and effective.
Description
Technical Field
The invention belongs to the technical field of image processing, and is used for automatically positioning a tongue body in a tongue image acquired in automatic tongue diagnosis of traditional Chinese medicine, providing a basis for subsequent tongue body segmentation, tongue body identification and diagnosis work of the automatic tongue diagnosis of traditional Chinese medicine, and particularly relates to a tongue body positioning method based on image clustering and gray projection.
Background
Tongue diagnosis is one of the main contents of "inspection" in traditional Chinese medicine, and is one of the traditional diagnosis methods with the characteristics of traditional Chinese medicine. The tongue is an index reflecting the most sensitive physiological function and pathological change of the human body, and has important application value in the diagnosis and treatment process of traditional Chinese medicine. An image processing technology is applied to establish an objective quantification and identification method of tongue diagnosis information, so that automation of tongue diagnosis of traditional Chinese medicine is realized, and the method has important practical significance for traditional Chinese medicine modernization. In an automated tongue diagnosis system, after tongue images of a patient are acquired by digital acquisition instruments (industrial cameras, etc.), a target area (tongue body) must be automatically segmented first. Therefore, the tongue image is segmented into important bands connecting tongue image acquisition and tongue recognition and diagnosis.
Tongue image segmentation presents certain challenges and difficulties. These difficulties stem from the nature of the tongue image itself: (1) The color of the tongue body is very similar to the color of the face, especially the color of the lips, and is easy to be confused; (2) The tongue body is used as a soft body, has no fixed shape, and has large individual variability of the tongue body shape; (3) From the medical pathology point of view, the tongue body is not smooth, tongue fur and tongue quality are different from person to person, and the pathological characteristics are greatly different; (4) Cracks and tongue fur color blocks of the tongue body can also seriously affect the segmentation of the tongue body.
The positioning of the tongue body aims at finding the rectangular area where the tongue body is located. The tongue positioning can help to exclude the middle upper part of lips and most of the areas of human faces except the tongue, avoid interference to tongue segmentation caused by the color approximation of the middle upper part of lips and the tongue, the color approximation of the human faces and the tongue, the wrinkles of the human faces and the like, and help to automatically segment the tongue in the tongue image. Zhang Ling and Qin Jian (1) summarize two prior knowledge based on the observation of tongue images, and on this basis propose a tongue positioning method based on Gray Projection (GP). These two a priori knowledge are: (1) The tongue root area in the tongue image is usually darker than other areas of the face; (2) When generating tongue images, there may be areas of darkness near the tongue contours due to the stretched tongue blocking light. The image rows and columns to which these darker areas relate have lower average gray values than the other image rows and columns. By using the priori knowledge, the tongue positioning method (GP) based on gray projection firstly converts a color image into a gray image, and then the tongue in the tongue image is positioned by the following steps: (1) Calculating the average gray value of each row and the average gray value of each column in the image; (2) The line with the lowest average gray value is found in the first half of the image line and has the line number Rup, the line with the lowest average gray value is found in the second half of the image line and has the line number Rdown, the column with the lowest average gray value is found in the left half of the image column and has the column number Cleft, and the column with the lowest average gray value is found in the right half of the image column and has the column number Cright; (3) Rectangular areas defined by Rup, rdown, cleft and weight were used as a result of tongue positioning.
After exploring a gray projection-based tongue positioning method, the applicant found that: the method has the advantages that tongue positioning is performed on the original tongue image acquired by the image acquisition equipment, and the effect is poor; the tongue body is positioned on the cut tongue image, so that a certain effect is achieved; however, no automatic image clipping method suitable for tongue positioning is proposed at present, and image clipping needs to be performed manually, which is time-consuming and labor-consuming. In order to realize tongue positioning on an original tongue image, the invention provides a tongue positioning algorithm based on image clustering and gray projection, and a good tongue positioning effect is obtained.
Reference is made to:
[1] zhang Ling tongue image segmentation method based on gray projection and threshold value automatic selection, chinese tissue engineering research and clinical rehabilitation 2010,14 (9): 1638-1641.
Disclosure of Invention
The invention aims to realize tongue positioning on an original tongue image, and provides a tongue positioning method based on image clustering and gray projection, which has good tongue positioning effect.
In order to achieve the above purpose, the technical scheme of the invention is as follows: a tongue positioning method based on image clustering and gray projection comprises the following steps:
s1, inputting a tongue image, and performing rough extraction on a tongue body region;
step S2, tongue positioning: and (3) carrying out actual positioning on the tongue body region which is roughly extracted in the step (S1) by using a gray projection technology, then carrying out column positioning by using a binarization result, and finally determining a rectangular region where the tongue body is positioned by combining the positioning results of the rows and the columns to realize tongue body positioning.
In an embodiment of the present invention, the specific implementation steps of the step S1 are as follows:
step S11, mapping a color space: mapping a color tongue image from the RGB color space to the HSI color space:
wherein ,
in formulas (1) to (4), R, G and B represent red, green, and blue components of an image, respectively;
step S12, determining a segmentation threshold value based on clustering: clustering is carried out on the hue components of the HSI color space, and various centers are iteratively updated until the centers are not changed; classifying the tone value corresponding to the pixel point into one of the categories according to the minimum principle of the absolute tone difference between the tone value of each pixel point and various centers, and updating the center of each category into the average tone value of the pixel point; after the image clustering is finished, tone values corresponding to the final class centers are arranged in an ascending order and put into a set T set In (a) and (b);
step S13, image threshold segmentation processing and morphological operation:
first, T is used set The following threshold segmentation is performed on the hue component of the image:
wherein ,Tset (i) Representing T set The i-th element of (a);
then, the threshold value segmentation result B is sequentially carried out img Performing morphological open operation and image filling operation to obtain corrected threshold segmentation result RB img ；
Step S14, extracting the tongue body area roughly: taking the maximum target area in the threshold segmentation result of the step S13 as a rough extraction result O of the tongue area img 。
In an embodiment of the present invention, the number of the categories clustered in the step S12 is set to 3, and the 3 initial category centers are set to the minimum value, the average value, and the maximum value of the tone components, respectively.
In an embodiment of the present invention, the specific implementation steps of the step S2 are as follows:
step S21, line positioning based on gray projection: taking the red component of the original tongue image as an object of gray projection, and performing gray projection operation only on a target pixel point, wherein the target pixel point is a tongue body region coarse extraction result O img White pixels in (a); specifically:
firstly, calculating the average gray value of the target pixel point on each row containing the target pixel point in the image, and then taking the row with the lowest average gray value as the upper boundary of the tongue positioning rectangular area, wherein the row number is R up ；
Next, O is taken img The maximum row of the middle target pixel point is taken as the lower boundary of the tongue rectangular positioning area, and the row number is marked as R down ；
Finally, the row range [ R up R down ]The other target pixel points, namely false tongue pixel points, are removed to obtain a black background point, so that a binarized image is obtained;
step S22, performing column positioning by using the binarized image obtained in step S21 to obtain a left boundary C of the tongue positioning rectangular region left And right boundary C right ；
Step S23, tongue positioning: using row boundaries R up and Rdown Column boundary C left and Cright And determining a rectangular area as a tongue positioning result.
Compared with the prior art, the invention has the following beneficial effects: the invention has good tongue positioning effect, and the method is simple and effective.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
Fig. 2 is a tongue positioning result, wherein: fig. 2 (a) is an original image, fig. 2 (b) is a hue (hue) component threshold segmentation result, fig. 2 (c) is a corrected threshold segmentation result, fig. 2 (d) is a binary image retaining only the maximum target area in the sub-image fig. 2 (c), fig. 2 (e) is a result after the line clipping in the sub-image fig. 2 (d), and fig. 2 (f) is a tongue positioning result.
FIG. 3 shows structural elements used in morphological open operations.
Fig. 4 is a tongue image positioning result with large tongue size difference.
Fig. 5 shows the result of tongue image localization with large difference in tongue shape.
Fig. 6 is a tongue image localization result with a large tongue color difference.
FIG. 7 is a graph showing the result of tongue image localization of thick tongue coating and strong texture.
Detailed Description
The technical scheme of the invention is specifically described below with reference to the accompanying drawings.
The invention provides a tongue positioning method based on image clustering and gray projection, which comprises the following steps:
s1, inputting a tongue image, and performing rough extraction on a tongue body region;
step S2, tongue positioning: and (3) carrying out actual positioning on the tongue body region which is roughly extracted in the step (S1) by using a gray projection technology, then carrying out column positioning by using a binarization result, and finally determining a rectangular region where the tongue body is positioned by combining the positioning results of the rows and the columns to realize tongue body positioning.
The specific implementation steps of the step S1 are as follows:
step S11, mapping a color space: mapping a color tongue image from the RGB color space to the HSI color space:
wherein ,
in formulas (1) to (4), R, G and B represent red, green, and blue components of an image, respectively;
step S12, determining a segmentation threshold value based on clustering: clustering is carried out on the hue components of the HSI color space, and various centers are iteratively updated until the centers are not changed; classifying the tone value corresponding to the pixel point into one of the categories according to the minimum principle of the absolute tone difference between the tone value of each pixel point and various centers, and updating the center of each category into the average tone value of the pixel point; after the image clustering is finished, tone values corresponding to the final class centers are arranged in an ascending order and put into a set T set In (a) and (b); the number of the categories of the clusters is set to be 3, and the 3 initial category centers are respectively set to be the minimum value, the average value and the maximum value of the tone components.
Step S13, image threshold segmentation processing and morphological operation:
first, T is used set The following threshold segmentation is performed on the hue component of the image:
wherein ,Tset (i) Representing T set The i-th element of (a);
then, the threshold value segmentation result B is sequentially carried out img Performing morphological open operation and image filling operation to obtain corrected threshold segmentation result RB img ；
Step S14, extracting the tongue body area roughly: taking the maximum target area in the threshold segmentation result of the step S13 as a rough extraction result O of the tongue area img 。
The specific implementation steps of the step S2 are as follows:
step S21, line positioning based on gray projection: taking the red component of the original tongue image as an object of gray projection, and performing gray projection operation only on a target pixel point, wherein the target pixel point is a tongue body region coarse extraction result O img White pixels in (a); specifically:
firstly, calculating the average gray value of the target pixel point on each row containing the target pixel point in the image, and then taking the row with the lowest average gray value as the upper boundary of the tongue positioning rectangular area, wherein the row number is R up ；
Next, O is taken img The maximum row of the middle target pixel point is taken as the lower boundary of the tongue rectangular positioning area, and the row number is marked as R down ；
Finally, the row range [ R up R down ]The other target pixel points, namely false tongue pixel points, are removed to obtain a black background point, so that a binarized image is obtained;
step S22, performing column positioning by using the binarized image obtained in step S21 to obtain a left boundary C of the tongue positioning rectangular region left And right boundary C right ；
Step S23, tongue positioning: using row boundaries R up and Rdown Column boundary C left and Cright And determining a rectangular area as a tongue positioning result.
The following is a specific implementation procedure of the present invention.
As shown in fig. 1, the tongue positioning method based on image clustering and gray projection of the present invention has the following execution flow:
1. crude extraction of tongue region
(1) Color space mapping: mapping a color image from the RGB color space to the HSI color space:
wherein ,
in formulas (1) - (4), R, G and B represent red, green, and blue components of an image, respectively.
(2) Determining a segmentation threshold based on the clustering: clustering is performed on Hue (Hue) components of the HSI color space, the number of classes is empirically set to 3, three initial class centers are respectively set to the minimum value, the average value and the maximum value of the Hue components, and the three class centers are iteratively updated until the class centers are not changed. And classifying the tone value corresponding to the pixel point into one of three categories according to the minimum principle of absolute tone difference between the tone value of each pixel point and three category centers in each iteration, and updating the center of each category into the average tone value of the pixel point of the category. After the image clustering is finished, the tone values corresponding to the three final class centers are arranged in an ascending order and put into a set T set Is a kind of medium.
(3) Image thresholding and morphological operations: first, T is used set The following threshold segmentation is performed on the hue component of the image:
wherein ,Tset (i) Representing T set I-th element of (a) in the list. Then, the threshold value segmentation result B is sequentially carried out img Performing morphological open operation and image filling operation to obtain corrected threshold segmentation result RB img Wherein, the structural elements used in the morphological opening operation are shown in figure 3. FIG. 2 (a) is an original view, and FIGS. 2 (B) and 2 (c) show initial threshold segmentation results B, respectively img And corrected threshold segmentation result RB img 。
(4) Crude extraction of tongue region: the maximum target (white) area in the threshold segmentation result is taken as a rough extraction result O of the tongue area img . FIG. 2 (d) shows a binarized image corresponding to only the largest target region of FIG. 2 (c), denoted as O img 。
2. Tongue positioning
Firstly, realizing the positioning by using a gray projection technology, then, performing column positioning by using a binarization result, and finally, determining a rectangular area where the tongue body is positioned by combining the positioning results of the rows and the columns to realize the tongue body positioning.
(1) Line location based on gray projection: taking the red component of the original tongue image as an object of gray projection, and performing gray projection operation only on a target pixel point, wherein the target pixel point is a tongue body region coarse extraction result O img White pixels in the display panel. Specifically, first, the average gray value of the target pixel point on each line containing the target pixel point in the image is calculated, then the line with the lowest average gray value is taken as the upper boundary of the tongue positioning rectangular area, and the line number is marked as R up The method comprises the steps of carrying out a first treatment on the surface of the Next, O is taken img The maximum row of the middle target pixel point is taken as the lower boundary of the tongue rectangular positioning area, and the row number is marked as R down . Since we consider that the tongue should be located at the R-th up Line and R down Line-to-line range [ R ] up R down ]The other target pixels should be false tongue pixels and should be cleared, i.e. they become black background pixels. Taking FIG. 2 (d) as an example, FIG. 2 (e) showsA binary image is shown after the elimination of false target pixels using line location.
(2) Column positioning: the left and right boundaries of the target region in fig. 2 (e) are respectively used as the left boundary C of the tongue positioning rectangular region left And right boundary C right 。
(3) Tongue body positioning: using row boundaries (R up and Rdown ) And column boundary (C) left and Cright ) A rectangular area is determined as a result of tongue positioning, as shown in fig. 2 (f).
To verify the effectiveness of the method of the present invention, experiments were performed on a series of tongue images of tongue size, shape, color, thick tongue coating and strong texture, each image having a size of 640 x 480. The method of the invention is compared with the existing Gray Projection (GP) method [1 ]. In order to quantitatively measure the accuracy of tongue positioning, a rectangular region obtained by each method is used as a target region in a binary segmentation result, and error classification errors (ME, misclassification Error) 2 and Kappa indexes (KI, kappa Index) 3 are used as quantitative measures. ME and KI are defined as follows:
wherein ,Bm and Fm Representing the background and the target in the manual standard segmentation result respectively, B a and Fa Respectively representing the background and the target in the segmentation result obtained by the automatic segmentation algorithm, wherein |·| represents the number of elements in the set. The value ranges of ME and KI are 0-1. Lower ME values represent better localization effects, higher KI values represent better localization effects. All experiments were performed on a notebook computer with a CPU of 1.7G Intel Core i5-3317U and a memory of 4G.
1. Qualitative comparison
To qualitatively compare the tongue positioning effect of the two methods, the applicant performed experiments on four sets of tongue images. The first group of tongue images is shown in fig. 4, and consists of ten images with large tongue size difference, wherein the tongue of the first five images is small, and the tongue of the second five images is large. The second group of tongue images is shown in fig. 5, and is composed of ten images with large tongue shape difference, wherein the first four images have square tongue bodies, the middle three images have flat tongue bodies, and the last three images have vertical tongue bodies. The third group of tongue images is shown in fig. 6, and is composed of ten images with large tongue color difference, wherein the first five images have white tongue bodies, and the second five images have red tongue bodies. The fourth group of tongue images is shown in fig. 7, the tongue bodies of the first five images have thicker tongue fur, and the tongue bodies of the last five images have strong textures. Each of figures 4-7 also shows the results of a manual ideal segmentation with lines outlining the actual contours of the tongue (group 2 in figures 4-7); the tongue positioning result obtained by the existing gray projection method is given and indicated by a rectangular frame (group 3 in fig. 4-7); the tongue positioning results obtained by the method of the present invention are shown as rectangular boxes (group 4 in fig. 4-7). As can be seen from fig. 4 to 7, the original gray projection method can be usually positioned to the upper boundary of the tongue body, namely the tongue root, but cannot be positioned to the lower boundary and the left and right boundaries of the tongue body, so that the positioning effect of the tongue body is poor. In contrast, the algorithm of the invention usually successfully locates four boundaries of the tongue body, namely the upper boundary, the lower boundary, the left boundary and the right boundary, and achieves satisfactory tongue body locating effect.
2. Quantitative comparison
In order to quantitatively compare tongue positioning results, a rectangular frame inner area obtained by tongue positioning is directly used as a target area, a rectangular frame outer area is used as a background area, a binary segmentation result is constructed, and the accuracy of the binary segmentation result is measured by using two measures of ME and KI, so that the accuracy of tongue positioning is quantitatively evaluated. The ME and KI values for each image in the four sets of experiments are listed in tables 1-4.
Table 1 quantitative comparison results of the first set of experiments
Table 2 quantitative comparison results of the second set of experiments
Table 3 quantitative comparison results of the third set of experiments
Table 4 quantitative comparison results of the fourth set of experiments
From these tables 1-4, it can be seen that the method of the present invention significantly reduces the ME value per image and increases the KI value per image compared to the existing gray-scale projection (GP) method. The lower ME value and the higher KI value indicate that the method has higher tongue positioning precision than the existing gray projection method.
The above is a preferred embodiment of the present invention, and all changes made according to the technical solution of the present invention belong to the protection scope of the present invention when the generated functional effects do not exceed the scope of the technical solution of the present invention.
Claims (2)
1. A tongue positioning method based on image clustering and gray projection is characterized by comprising the following steps:
s1, inputting a tongue image, and performing rough extraction on a tongue body region;
step S2, tongue positioning: the tongue body area which is roughly extracted in the step S1 is firstly positioned by utilizing a gray projection technology, then is positioned by utilizing a binarization result to carry out column positioning, and finally, the rectangular area where the tongue body is positioned is determined by combining the positioning results of the rows and the columns to realize tongue body positioning;
the specific implementation steps of the step S1 are as follows:
step S11, mapping a color space: mapping a color tongue image from the RGB color space to the HSI color space:
wherein ,
in formulas (1) to (4), R, G and B represent red, green, and blue components of an image, respectively;
step S12, determining a segmentation threshold value based on clustering: clustering is carried out on the hue components of the HSI color space, and various centers are iteratively updated until the centers are not changed; classifying the tone value corresponding to the pixel point into one of the categories according to the minimum principle of the absolute tone difference between the tone value of each pixel point and various centers, and updating the center of each category into the average tone value of the pixel point; after the image clustering is finished, tone values corresponding to the final class centers are arranged in an ascending order and put into a set T set In (a) and (b);
step S13, image threshold segmentation processing and morphological operation:
first, T is used set The elements in (1) threshold the tonal components of the image as followsValue segmentation:
wherein ,Tset (i) Representing T set The i-th element of (a);
then, the threshold value segmentation result B is sequentially carried out img Performing morphological open operation and image filling operation to obtain corrected threshold segmentation result RB img ；
Step S14, extracting the tongue body area roughly: taking the maximum target area in the threshold segmentation result of the step S13 as a rough extraction result O of the tongue area img ；
The specific implementation steps of the step S2 are as follows:
step S21, line positioning based on gray projection: taking the red component of the original tongue image as an object of gray projection, and performing gray projection operation only on a target pixel point, wherein the target pixel point is a tongue body region coarse extraction result O img White pixels in (a); specifically:
firstly, calculating the average gray value of the target pixel point on each row containing the target pixel point in the image, and then taking the row with the lowest average gray value as the upper boundary of the tongue positioning rectangular area, wherein the row number is R up ；
Next, O is taken img The maximum row of the middle target pixel point is taken as the lower boundary of the tongue rectangular positioning area, and the row number is marked as R down ；
Finally, the row range [ R up R down ]The other target pixel points, namely false tongue pixel points, are removed to obtain a black background point, so that a binarized image is obtained;
step S22, performing column positioning by using the binarized image obtained in step S21 to obtain a left boundary C of the tongue positioning rectangular region left And right boundary C right ；
Step S23, tongue positioning: using row boundaries R up and Rdown Column boundary C left and Cright Rectangular areas are determined asAnd positioning the tongue body.
2. The tongue positioning method based on image clustering and gray projection according to claim 1, wherein the number of the categories of the clusters in the step S12 is set to 3, and the 3 initial category centers are set to the minimum value, the average value and the maximum value of the tone components, respectively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811365181.2A CN109658382B (en) | 2018-11-16 | 2018-11-16 | Tongue positioning method based on image clustering and gray projection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811365181.2A CN109658382B (en) | 2018-11-16 | 2018-11-16 | Tongue positioning method based on image clustering and gray projection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109658382A CN109658382A (en) | 2019-04-19 |
| CN109658382B true CN109658382B (en) | 2023-05-30 |
Family
ID=66110972
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811365181.2A Active CN109658382B (en) | 2018-11-16 | 2018-11-16 | Tongue positioning method based on image clustering and gray projection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109658382B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111079637B (en) * | 2019-12-12 | 2023-09-08 | 武汉轻工大学 | Method, device, equipment and storage medium for segmenting rape flowers in field image |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105118067A (en) * | 2015-09-22 | 2015-12-02 | 成都融创智谷科技有限公司 | Image segmentation method based on Gaussian smoothing filter |
| CN108537779A (en) * | 2018-03-27 | 2018-09-14 | 哈尔滨理工大学 | The method of vertebra segmentation and centroid detection based on cluster |
| CN108682017A (en) * | 2018-04-11 | 2018-10-19 | 浙江工业大学 | Super-pixel method for detecting image edge based on Node2Vec algorithms |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI296110B (en) * | 2004-12-31 | 2008-04-21 | Committee On Chinese Medicine And Pharmacy Dept Of Health | Tongue diagnosis systems and methods |
| CN103729851B (en) * | 2014-01-04 | 2016-06-29 | 闽江学院 | Image partition method based on notable transitional region |
| JP2016150024A (en) * | 2015-02-16 | 2016-08-22 | コニカミノルタ株式会社 | Organ imaging device |
| CN107392904A (en) * | 2017-07-28 | 2017-11-24 | 陆杰 | A kind of partitioning algorithm of the medical image based on mathematical morphology |
| CN108615239B (en) * | 2018-05-10 | 2020-03-27 | 福建中医药大学 | Tongue image segmentation method based on threshold technology and gray level projection |
-
2018
- 2018-11-16 CN CN201811365181.2A patent/CN109658382B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105118067A (en) * | 2015-09-22 | 2015-12-02 | 成都融创智谷科技有限公司 | Image segmentation method based on Gaussian smoothing filter |
| CN108537779A (en) * | 2018-03-27 | 2018-09-14 | 哈尔滨理工大学 | The method of vertebra segmentation and centroid detection based on cluster |
| CN108682017A (en) * | 2018-04-11 | 2018-10-19 | 浙江工业大学 | Super-pixel method for detecting image edge based on Node2Vec algorithms |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109658382A (en) | 2019-04-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108615239B (en) | Tongue image segmentation method based on threshold technology and gray level projection | |
| CN107977671B (en) | Tongue picture classification method based on multitask convolutional neural network | |
| CN110458831B (en) | Scoliosis image processing method based on deep learning | |
| CN110189383B (en) | Traditional Chinese medicine tongue color and fur color quantitative analysis method based on machine learning | |
| CN106023151B (en) | Tongue object detection method under a kind of open environment | |
| CN113667584A (en) | Colony contrast collection | |
| WO2013080868A1 (en) | Image processing device, image processing method, and image processing program | |
| CN113723573A (en) | Tumor tissue pathological classification system and method based on adaptive proportion learning | |
| CN115797352B (en) | Tongue picture image processing system for traditional Chinese medicine health-care physique detection | |
| CN111860538A (en) | Tongue color identification method and device based on image processing | |
| US9852354B2 (en) | Method and apparatus for image scoring and analysis | |
| CN110866932A (en) | Multi-channel tongue edge detection device and method and storage medium | |
| CN104933723B (en) | Tongue image dividing method based on rarefaction representation | |
| CN115965607A (en) | Intelligent traditional Chinese medicine tongue diagnosis auxiliary analysis system | |
| CN111986157B (en) | Digital pathological image quality evaluation system | |
| CN111222371A (en) | Sublingual vein feature extraction device and method | |
| CN110648336B (en) | Method and device for dividing tongue texture and tongue coating | |
| CN109658382B (en) | Tongue positioning method based on image clustering and gray projection | |
| CN107977976A (en) | A kind of image partition method | |
| CN114511567A (en) | Tongue body and tongue coating image identification and separation method | |
| CN107729863B (en) | Human finger vein recognition method | |
| CN108629780B (en) | Tongue image segmentation method based on color decomposition and threshold technology | |
| CN112184696B (en) | Cell nucleus and organelle counting and area calculating method and system thereof | |
| CN117522862A (en) | Image processing method and processing system based on CT image pneumonia recognition | |
| Wilm et al. | Multi-scanner canine cutaneous squamous cell carcinoma histopathology dataset |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |