CN110689586A - Tongue image identification method in traditional Chinese medicine intelligent tongue diagnosis and portable correction color card used for same - Google Patents

Tongue image identification method in traditional Chinese medicine intelligent tongue diagnosis and portable correction color card used for same Download PDF

Info

Publication number
CN110689586A
CN110689586A CN201810741280.XA CN201810741280A CN110689586A CN 110689586 A CN110689586 A CN 110689586A CN 201810741280 A CN201810741280 A CN 201810741280A CN 110689586 A CN110689586 A CN 110689586A
Authority
CN
China
Prior art keywords
color
correction
image recognition
tongue
determining
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
CN201810741280.XA
Other languages
Chinese (zh)
Other versions
CN110689586B (en
Inventor
李梢
阮良
侯思宇
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.)
Tsinghua University
Original Assignee
Tsinghua University
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 Tsinghua University filed Critical Tsinghua University
Priority to CN201810741280.XA priority Critical patent/CN110689586B/en
Publication of CN110689586A publication Critical patent/CN110689586A/en
Application granted granted Critical
Publication of CN110689586B publication Critical patent/CN110689586B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • 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
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
    • 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

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Biomedical Technology (AREA)
  • Public Health (AREA)
  • Epidemiology (AREA)
  • Pathology (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Image Analysis (AREA)
  • Image Processing (AREA)

Abstract

The invention provides a recognizable portable tongue image color correction color chart and a recognition method thereof, which are used in the intelligent tongue diagnosis of traditional Chinese medicine and are used for solving the defects that the traditional tongue image color correction color chart is inconvenient to carry about, is difficult to be recognized by a computer and the like. The color card is small in size and easy to carry in consideration of convenience. But too small a size is not conducive to automatic identification by a computer. The invention adds special geometric patterns in the specific position of the color card, adds the special geometric patterns into the color card area according to the specific size and the geometric positions, and is matched with a specific improved computer identification algorithm, thereby realizing the automatic identification, positioning and color sampling of the color card by the computer and obtaining high identification rate. The invention eliminates the complicated procedure of manually calibrating the color card in color correction and improves the efficiency of color correction processing. The invention also provides a solution for errors caused by low identification rate of the intelligent tongue inspection identifier of the traditional Chinese medicine under low pixels, perspective deformation of a color chart and other factors, and obtains good effect.

Description

Tongue image identification method in traditional Chinese medicine intelligent tongue diagnosis and portable correction color card used for same
Technical Field
The invention provides a recognizable portable tongue image color correction color chart and a recognition method thereof.
Background
The traditional Chinese medicine tongue image diagnosis mode needs a doctor to directly observe the situation of the tongue of a patient and draw a conclusion, and the mode needs the patient and the doctor to face each other, so that the convenience and the popularization are lacked. In recent years, along with the popularization of network communication, especially smart phones, the use of mobile phones for tongue image intelligent acquisition and diagnosis gradually becomes a new trend, and the defects of poor popularization and convenience of the traditional mode can be well overcome. However, the tongue image collection and diagnosis by using the mobile phone is easily interfered by external factors, and has high requirements on light, angle, definition and the like of tongue image shooting, and especially the color restoration problem in color photography, these technical difficulties become problems to be solved urgently in the tongue image collection development of the mobile phone.
The method is one of the commonly used methods at present, and analyzes and fits according to the color change condition of the color card, so as to indirectly infer the color change of the tongue color and finally realize the standardized treatment of the tongue color. However, the color card used in the existing tongue color correction research is usually large in size, the smallest color card has the size of a book, the color card is not convenient to carry about, and an effective solution is also lacked in the intelligent acquisition of the tongue image in traditional Chinese medicine. In addition, the color card used in the existing color correction lacks the function of automatic positioning, and the color card is often identified and positioned manually in practical application. Such a method requires high labor cost, certain operation and processing time, and may have a great influence on the efficiency and accuracy of the tongue image acquisition standardization process.
In addition, for tongue image color identification in intelligent acquisition of tongue images in traditional Chinese medicine, an actual and effective scheme is lacked in the prior art.
Disclosure of Invention
The invention provides a recognizable portable tongue image color correction color chart and a recognition method thereof.
Drawings
FIG. 1 is a flow chart of a method for identifying a color correction target for a tongue image according to an embodiment of the present invention.
FIG. 2 is a tongue color correction target according to one embodiment of the invention.
Fig. 3 is used to illustrate the identification of a color chip location identifier according to one embodiment of the present invention.
Fig. 4 is a diagram for illustrating a shape change of a rectangle after perspective deformation.
Figure 5 shows a graph of the variation of the sample points identifying success and failure.
Detailed Description
In view of the above problems in the prior art, the present inventors propose a portable color correction color chip with automatic positioning function and an identification method thereof. The invention aims to 2, firstly, the positioning work of the color card identification which is usually manually finished in the past is finished by a machine by utilizing a special color card design and an automatic identification method, so that the working efficiency and the normalization of a color correction process are improved, and convenience is provided for large-scale tongue image color correction or color correction under other scenes. The invention provides a color card positioning method based on a position identifier, which is characterized in that a pattern with a special shape is arranged at the vertex of a color card, a symbol is identified and positioned by using an image identification algorithm, and then the positions of all color blocks are deduced through a geometric relation and the corresponding color values are calculated. And secondly, the shape, size, layout and the like of the traditional color card are improved, so that the color card can achieve the effects of convenient carrying and use in practical use.
As shown in fig. 1, the color chart identification and color correction process includes:
the method comprises the following steps: and setting a color card special position identifier. Different special identifiers are respectively designed at four vertexes of the color card. With three vertices taking one sign and the last vertex taking another sign, the two signs having different geometric characteristics. According to a specific embodiment of the present invention, after the position identifier is set, color blocks are filled in the rectangular region formed by the four symbols.
Step two: and acquiring an image by using the color card. And placing the color card and the object in the same scene in the area under the same light condition for shooting to obtain an image simultaneously containing the target object and the color card.
Step three: and searching the image, and searching the positions of 4 symbols by using the geometric characteristics of the symbols. Due to the problem of shooting angle, the rectangular color chart may be distorted into an arbitrary quadrangle due to the perspective effect, so that the coordinates need to be corrected by using the geometric relationship of four vertices to restore the area into a rectangle.
Step four: and calculating and positioning the color block area by utilizing the pre-designed set relation of the color blocks and the color cards to find the center points of the color blocks. And searching the image by using the central point to find the boundary of the color block, thereby determining the position of the whole color block. And finally, calculating the color of the whole color block. According to a specific embodiment, the boundaries of the color blocks are found by using a region growing method.
Step five: and according to the color taking result, performing linear fitting correction by using color block color information in a standard environment, automatically judging the identification result according to the goodness of fit, and judging that the identification is successful when the goodness of fit reaches and/or exceeds a certain preset threshold value.
Finally, after all the steps are completed, the color correction algorithm can be used for carrying out color correction on the image and finally obtaining the tongue image with standardized color.
The embodiments of the present invention will be further described with reference to the accompanying drawings.
Designing a color card:
FIG. 2 is a portable tongue image color correction target supporting automatic positioning according to one embodiment of the present invention. The color chip includes two regions: a hand-held area and a color chip area.
The handheld region is arranged on one side of the color card, so that a sufficient space is ensured when the handheld region is held by a single hand, the use convenience is ensured, and the effect of color correction is influenced in order to fully avoid the phenomenon that the color card is shielded by the hand and/or shadow is generated in the color card region.
According to an embodiment of the present invention, the color chip area is a rectangle and includes 3 portions, which are a location identifier area, an additional information identification area, and a color chip color block area.
According to a particular embodiment, the location identifier zones are distributed at the four vertices of the colour chip area, which have special geometrical characteristics. According to a specific embodiment, the identifier comprises a first identifier and a second identifier depending on the difference in the geometrical characteristics.
According to a particular embodiment of the invention. Wherein the class 1 identifiers comprise three rectangles with a side length ratio of 7:5:3, and the first class identifiers are 3 in number and are respectively located at three of the four vertices (such as the top left, top right, and bottom left vertices) of the color chip area. The second identifier includes three rectangles having a side length ratio of 5:3:1, and is disposed at a fourth vertex of the four vertices of the color chip region (such as at a lower right corner of the color chip region).
The additional information identification area contains simple mark points, for example, mark points that can be analyzed according to a binary rule, for example, four color blocks for recording the version number of the color card are included as parameter settings when the color card is further analyzed subsequently. The color card color block area comprises a plurality of squares which have the same size and are uniformly arranged in a rectangle formed by the position identifiers. Each color block has a black border, which can help the recognition program to better recognize the boundaries of the color blocks. The inner area of the color block is filled with different colors, and the specific colors can be properly selected and/or replaced to meet different practical purposes.
Identification by an identifier:
after the image is acquired by using the color card, the image needs to be subjected to gray scale and binary processing. According to one embodiment of the invention, a dynamic global threshold method is adopted to carry out binary processing on the image, wherein a plurality of groups of candidate threshold values are preset, and the threshold values are substituted into the binary processing one by one. The binarized image is scanned in the manner of fig. 3 in the x-axis and y-axis, respectively, and is identified based on the geometric features of the identifier.
The color chip identifier location algorithm according to one embodiment of the present invention comprises:
(1) and determining a threshold th in the candidate threshold library as an initial binarization threshold, and binarizing the gray level image to obtain a binary image G.
(2) The length of the continuous black or white dots is counted in each line. If black is satisfied: white: black: white: and if the point coordinates in the target area are 1:1:3:1:1, calculating the point coordinates in the target area, and adding the point coordinates into a candidate point library.
(3) And after a candidate point library is obtained, searching in the square r pixel, judging whether other candidate points exist or not, counting the number, if the number of the surrounding candidate points is larger than a preset target n, considering the point as a target point, and adding the target point into a target point library.
(4) The X and Y axis direction is reversed, the steps (2) and (3) are carried out again, and finally the X and Y coordinates of all points are obtained
(5) Selecting the next candidate threshold th from the candidate preset library and repeating the steps (2) - (4)
(6) And summarizing all results, carrying out DBSCAN clustering analysis, selecting an area with the number of the clustering points ranked in the first three as a final area, and acquiring the coordinate of the central point of the final area as the coordinate of the identifier.
And positioning the center point of the color block:
due to perspective, the rectangle of the color chart may be distorted into an arbitrary quadrangle, as shown in fig. 4, where point A, B, D is a type 1 identifier and point C is a type 2 identifier. If the geometric calculation is directly carried out, the obtained result has larger error. In one embodiment according to the invention, the method is to reduce any quadrilateral into a rectangle and then to further process it. And transforming the coordinates of the image in a gray space by using a geometric relation, determining the target coordinates of the pixel points, and keeping the original gray value by the gray value. Since the calculated coordinates are not necessarily integers, and the gray values corresponding to specific integer coordinate points in the target image need to be interpolated, in one embodiment according to the present invention, bilinear interpolation is used for processing. However, such a method has a large amount of calculation, and in a preferred embodiment of the present invention, the determination of coordinates of a center point of a color block is directly performed by using a perspective principle formula, including:
let the coordinate of the center point X of a color block in the standard rectangle be i, j, the length of the original rectangle be a, and the height be b. After perspective transformation, four vertexes of the rectangle are respectively changed into (a)1,b1),(a2,b2),(a3,b3),(a4,b4) Then, the coordinates of point X, after perspective transformation, become:
Figure BDA0001723303520000042
therefore, when the coordinates i and j of the center point of the color block in the standard rectangle are determined, the coordinates i 'and j' corresponding to the center point in the actually shot image can be deduced.
Color block selection and color value calculation:
after the coordinates of the central point of the color block are obtained, the image is processed by a region growing algorithm in a gray scale space, and the color block has an obvious boundary, so that the color and the gray value in the same region are approximately uniform, and a relatively complete color block region can be obtained by the region growing algorithm. After determining the color block region, considering that the boundary may cause deviation to the value of the average color, the boundary of the color block region needs to be cut, and in a specific embodiment according to the present invention, the specific cutting size is 1/10 of the side length of the color block region; thereby obtaining the final patch areas. Finally, the specific color value of the color patch can be obtained by calculating the average value of the colors of the color patch area. In this way, the color of all color blocks of the whole color card is obtained by processing each color block area. In addition, the value of the additional information identification area is determined, and the additional information such as the color card version is read.
And (3) judging whether the color card identification is successful:
after intensive and specific research, the invention discovers that: 1) the identification of the color card has a failure; thus, 2) means for judging the recognition failure is required.
Specifically, the case where the recognition fails includes:
identification failure due to ambiguity and/or unrecognizability of several location identifiers of the color chart, or
The position of the identified color block is obviously deviated due to serious deformation of the color block image and the like, so that the identified color block area can be some irrelevant areas and/or patterns,
the color block color values obtained under the above conditions are not the actual color values of the color blocks related to the color card, i.e., the identification fails.
Therefore, according to an embodiment of the present invention, after the color chart identification is performed, the identification effect of the color chart is determined, and only the sample values meeting the "identification success" requirement are subjected to the subsequent processing.
On the basis of a large number of experimental verifications, the inventor finds that under a common shooting condition, the color change of a color block of a color card relative to the color change under a standard environment approximately conforms to a linear rule under an RGB channel. The experiment is realized on the basis of the existing 48-color card, and the design RGB values of each color block of the color card are as follows.
Fig. 5 shows the color change of the sampling points in the R channel, where the identification succeeds and fails, and it can be seen that the change of R in the case of successful identification better conforms to the linear law. Accordingly, the present inventors propose a scheme for determining a recognition result by using a linear regression method, specifically including:
suppose a color card has n sampling points, which are embodied as n color blocks with numbers of 1-n and Y1，Y2...YnColor measurements of n color patches in a standard environment, respectively, and y1，y2...ynColor measured values of color patches numbered 1 to n in a general environment (shooting environment). Y isR、yRFor each sampling point to take a value on the R channel, then the least square method is used to obtain the representation YR、yRThe formula of linear approximate relationship between:
YR＝fR(yR)＝kR·yR+bR
wherein f isRFor the mapping function under R channel, kR、bRAs a mapping function fRAnd fitting coefficients of the first order term and the constant term.
Similarly, the mapping function f under the G channel and the B channel can be obtained respectivelyG，fB。
R, G, B channel measurement with a point i to be calibrated
Figure BDA0001723303520000062
It is known to obtain a correction value for point i in a standard environment
Figure BDA0001723303520000063
In one embodiment according to the present invention, the following formula is used to calculate separately
Figure BDA0001723303520000064
The value of (c):
Figure BDA0001723303520000065
wherein all the coefficients k and b in the above formula are determined by using the classical least squares principle
Figure BDA0001723303520000071
And solving the linear equation of the equation (a).
After the equation coefficients k and b are obtained, the fitting equation and the actual measurement value can be subjected to statistical analysis and linear goodness of fit R can be obtained2. Wherein R is2The calculation can be made according to the following formula:
Figure BDA0001723303520000072
wherein y isiRefers to the measured value of the ith sample,
Figure BDA0001723303520000073
the correction value calculated by the linear regression equation of the ith sample is referred to,is the average of all measured values yValues, where 1. ltoreq. i. ltoreq. n.
In one exemplary embodiment according to the present invention, when R2And if the value is more than 0.8, the identification is considered to be successful.
And after the successful identification is judged, carrying out next color correction processing by using the information of the change of the color block of the color card relative to the color in the standard environment, namely correcting the color of the tongue image, thereby obtaining the tongue image after color correction. The color correction processing of the next step may use an interpolation method or a regression method such as linear regression, polynomial regression, or the like.
In a series of experiments that the present inventors have performed, 134 samples of color chart images taken under different conditions were subjected to classification and comparison. Utilizing the above-mentioned R2The evaluation method of more than 0.8 evaluates the color card identification result. The recognition accuracy can reach about 94.2% in a severe environment with low pixels, and the recognition rate can reach 98.5% in a better environment with high pixels, and the recognition rate can reach 96.3% in total, so that the requirements of practical use can be met. The specific results include:

Claims (16)

1. a tongue image recognition method using a portable tongue image color correction color chart, the correction color chart including a plurality of color patches having different colors, respectively, comprising:
A) determining whether color recognition of the correction color chart is successful for an image of the correction color chart photographed under the same photographing condition as that of the tongue image,
B) and a step A) of correcting the color of the tongue image by using the information of the change of the color block of the correction color chart relative to the color in the standard environment when the color of the correction color chart is successfully identified, thereby obtaining the tongue image after color correction.
2. The tongue image recognition method according to claim 1, wherein step a) comprises:
A1) determining the value y of each of the n color patches of the correction color chip on at least one color channel in the capture environmenti，i＝1,2….n，
A2) For the n measured values yiDetermining fitting coefficients k and b of the primary term and the constant term by using the principle of least squares, and further determining corresponding color correction values
Figure FDA0001723303510000011
i＝1,2….n，
A3) Determination of yiDistributed linear goodness of fit R2：
Wherein:
Figure FDA0001723303510000013
is the correction value for the ith sample,
Figure FDA0001723303510000014
is the average of all measured values y, where 1. ltoreq. i.ltoreq.n
A4) When R is2When the value is greater than or equal to a predetermined threshold value, the image recognition is judged to be successful, and when R is greater than or equal to the predetermined threshold value2And judging that the image recognition fails when the image recognition is smaller than a preset threshold value.
3. The tongue image recognition method according to claim 1, wherein step a) comprises:
A1) determining the value y of each of the n color patches of the correction color chip on at least one color channel in the capture environmenti，i＝1,2….n，
A2) For the n measured values yiDetermining fitting coefficients k and b of the primary term and the constant term by using the principle of least squares, and further determining corresponding color correction values
Figure FDA0001723303510000021
i＝1,2….n，
A3') determining yiN-1, 2 …. n. a first subset of the distribution has a goodness of linear fit R2：
Figure FDA0001723303510000022
Wherein:
the summation in the above equation is over the first subset,
Figure FDA0001723303510000023
is the correction value for the ith sample,is one selected from the following items:
the average of all measured values y, where 1. ltoreq. i. ltoreq.n,
the average of the measured values y for the first subset, an
yiA second subset of the measured values y,
A4) when R is2When the value is greater than or equal to a predetermined threshold value, the image recognition is judged to be successful, and when R is greater than or equal to the predetermined threshold value2And judging that the image recognition fails when the image recognition is smaller than a preset threshold value.
4. The tongue image recognition method according to claim 2 or 3, further comprising:
determining the central point of each color block by using the preset set relation of the color blocks and the correction color cards,
the center points of the color patches are used to search the image of the correction color card, thereby determining the boundaries of the color patches and determining the colors of the color patches.
5. The tongue image recognition method according to claim 2 or 3, wherein:
the calibration color chart includes a plurality of identifiers respectively disposed at a plurality of predetermined positions of the calibration color chart,
the plurality of identifiers comprising at least two different symbols, the at least two different symbols having different geometric characteristics,
searching the image, determining the position of each symbol by using the geometric characteristics,
the coordinates are corrected using the positions of the respective symbols, and distortion that may occur due to the perspective effect of photographing is corrected.
6. The tongue image recognition method according to any one of claims 1 to 3, wherein:
the image and tongue image of the corrected color chart are obtained by placing the color chart and the object in the same scene in the area of the same light condition and shooting.
7. The tongue image recognition method according to claim 4, wherein:
determining the boundaries of the color patches includes determining the boundaries of the color patches using a region growing method.
8. A portable correction color chip characterized in that it is adapted for use in the tongue image recognition method according to one of claims 1 to 7 and comprises:
a plurality of color blocks with different colors respectively,
a plurality of identifiers respectively disposed at a plurality of predetermined positions of the calibration color chart,
wherein:
the plurality of identifiers includes at least two different symbols having different geometric features so that the positions of the respective symbols can be determined by searching the image and using the geometric features.
9. The portable correction color chip of claim 8, wherein:
the portable calibration color chip comprises a handheld area and a color chip area,
the hand-held area is arranged at one side of the color card to ensure that sufficient space is available when the hand is held by a single hand and to fully avoid the hand from shielding the color card and/or generating shadow in the color card area,
the color chip area is a rectangle and includes a location identifier area and a color patch area.
10. The portable correction color chip of claim 9, wherein:
the location identifier areas are distributed at four vertices of the color chip area, which have predetermined geometric features,
and the color blocks are arranged in the color block areas.
11. A method for judging whether color identification of a correction color chip is successful, wherein the correction color chip comprises a plurality of color blocks with different colors, and the method is characterized by comprising the following steps:
A1) determining the value y of each of the n color patches of the correction color chip on at least one color channel in the capture environmenti，i＝1,2….n，
A2) For the n measured values yiDetermining corresponding color correction values by using the principle of least squares
Figure FDA0001723303510000031
A3) Determination of yiDistributed linear goodness of fit R2：
Wherein:
Figure FDA0001723303510000033
is the correction value for the ith sample,
Figure FDA0001723303510000034
is the average of all measured values y, where 1. ltoreq. i≤n
A4) When R is2When the value is greater than or equal to a predetermined threshold value, the image recognition is judged to be successful, and when R is greater than or equal to the predetermined threshold value2And judging that the image recognition fails when the image recognition is smaller than a preset threshold value.
12. A method according to claim 11, characterized in that the method is adapted for the identification of tongue images.
13. A method for judging whether color identification of a correction color chip is successful, wherein the correction color chip comprises a plurality of color blocks with different colors, and the method is characterized by comprising the following steps:
A1) determining the value y of each of the n color patches of the correction color chip on at least one color channel in the capture environmenti，i＝1,2….n，
A2) For the n measured values yiDetermining corresponding color correction values by using the principle of least squares
Figure FDA0001723303510000041
A3') determining yiN-1, 2 …. n. a first subset of the distribution has a goodness of linear fit R2：
Figure FDA0001723303510000042
Wherein:
the summation in the above equation is over the first subset,
Figure FDA0001723303510000043
is the correction value for the ith sample,
Figure FDA0001723303510000044
is one selected from the following items:
the average of all measured values y, where 1. ltoreq. i. ltoreq.n,
the average of the measured values y for the first subset, an
yiA second subset of the measured values y,
A4) when R is2When the value is greater than or equal to a predetermined threshold value, the image recognition is judged to be successful, and when R is greater than or equal to the predetermined threshold value2And judging that the image recognition fails when the image recognition is smaller than a preset threshold value.
14. A method according to claim 13, characterized in that the method is adapted for the identification of tongue images.
15. The tongue image recognition method according to claim 3, wherein:
the first subset and the second subset are then selected.
16. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method of one of claims 1 to 15.
CN201810741280.XA 2018-07-08 2018-07-08 Tongue image identification method in traditional Chinese medicine intelligent tongue diagnosis and portable correction color card used for same Active CN110689586B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810741280.XA CN110689586B (en) 2018-07-08 2018-07-08 Tongue image identification method in traditional Chinese medicine intelligent tongue diagnosis and portable correction color card used for same

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810741280.XA CN110689586B (en) 2018-07-08 2018-07-08 Tongue image identification method in traditional Chinese medicine intelligent tongue diagnosis and portable correction color card used for same

Publications (2)

Publication Number Publication Date
CN110689586A true CN110689586A (en) 2020-01-14
CN110689586B CN110689586B (en) 2022-12-16

Family

ID=69107473

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810741280.XA Active CN110689586B (en) 2018-07-08 2018-07-08 Tongue image identification method in traditional Chinese medicine intelligent tongue diagnosis and portable correction color card used for same

Country Status (1)

Country Link
CN (1) CN110689586B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112950485A (en) * 2020-11-27 2021-06-11 京东数字科技控股股份有限公司 Color card, image color difference processing method and device, electronic equipment and storage medium
CN113057583A (en) * 2021-03-09 2021-07-02 上海中医药大学 Tongue picture color correction method

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1803087A (en) * 2006-01-19 2006-07-19 上海交通大学 Tongue color automatic recognition method
CN101972138A (en) * 2010-11-08 2011-02-16 哈尔滨工业大学 Integrated portable standardized traditional Chinese medical science tongue image acquiring equipment
CN102095371A (en) * 2010-11-25 2011-06-15 天津大学 Industrial color vision detection device and method
CN102714687A (en) * 2010-01-19 2012-10-03 阿克佐诺贝尔国际涂料股份有限公司 Method and system for determining colour from an image
CN105466430A (en) * 2015-12-31 2016-04-06 零度智控（北京）智能科技有限公司 Unmanned aerial vehicle positioning method and device
CN106546581A (en) * 2016-11-02 2017-03-29 长沙云知检信息科技有限公司 Detection paper card intelligent checking system and detection paper card intelligent analysis method
CN108185993A (en) * 2018-01-31 2018-06-22 潘映含 A kind of tongue is as acquisition method

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1803087A (en) * 2006-01-19 2006-07-19 上海交通大学 Tongue color automatic recognition method
CN102714687A (en) * 2010-01-19 2012-10-03 阿克佐诺贝尔国际涂料股份有限公司 Method and system for determining colour from an image
CN101972138A (en) * 2010-11-08 2011-02-16 哈尔滨工业大学 Integrated portable standardized traditional Chinese medical science tongue image acquiring equipment
CN102095371A (en) * 2010-11-25 2011-06-15 天津大学 Industrial color vision detection device and method
CN105466430A (en) * 2015-12-31 2016-04-06 零度智控（北京）智能科技有限公司 Unmanned aerial vehicle positioning method and device
US20170193271A1 (en) * 2015-12-31 2017-07-06 ZEROTECH (Shenzhen) Intelligence Robot Co., Ltd. Positioning method and positioning device for unmanned aerial vehicle
CN106546581A (en) * 2016-11-02 2017-03-29 长沙云知检信息科技有限公司 Detection paper card intelligent checking system and detection paper card intelligent analysis method
CN108185993A (en) * 2018-01-31 2018-06-22 潘映含 A kind of tongue is as acquisition method

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112950485A (en) * 2020-11-27 2021-06-11 京东数字科技控股股份有限公司 Color card, image color difference processing method and device, electronic equipment and storage medium
CN112950485B (en) * 2020-11-27 2023-11-03 京东科技控股股份有限公司 Color card, image color difference processing method and device, electronic equipment and storage medium
CN113057583A (en) * 2021-03-09 2021-07-02 上海中医药大学 Tongue picture color correction method

Also Published As

Publication number Publication date
CN110689586B (en) 2022-12-16

Similar Documents

Publication Publication Date Title
CN107609557B (en) Pointer instrument reading identification method
CN110400278B (en) Full-automatic correction method, device and equipment for image color and geometric distortion
CN109859226B (en) Detection method of checkerboard corner sub-pixels for graph segmentation
JP2004054530A (en) Device, method, and program for reading two-dimensional code and recording medium for this program
KR102073468B1 (en) System and method for scoring color candidate poses against a color image in a vision system
CN110111387B (en) Dial plate characteristic-based pointer meter positioning and reading method
CN106017313B (en) Edge detection deviation correction value calculation method, edge detection deviation correction method and device
CN107356213B (en) Optical filter concentricity measuring method and terminal equipment
CN112288724B (en) Defect detection method and device, electronic equipment and storage medium
CN109948393A (en) A kind of localization method and device of bar code
CN111369605A (en) Infrared and visible light image registration method and system based on edge features
CN114399502A (en) Appearance defect detection method and system suitable for LED chip and storage medium
US20220277434A1 (en) Measurement System, Method for Generating Learning Model to Be Used When Performing Image Measurement of Semiconductor Including Predetermined Structure, and Recording Medium for Storing Program for Causing Computer to Execute Processing for Generating Learning Model to Be Used When Performing Image Measurement of Semiconductor Including Predetermined Structure
CN110689586B (en) Tongue image identification method in traditional Chinese medicine intelligent tongue diagnosis and portable correction color card used for same
CN104182757A (en) Method of acquiring actual coverage area of measured target and device
CN110569845A (en) test paper image correction method and related device
CN110852207A (en) Blue roof building extraction method based on object-oriented image classification technology
CN113609984A (en) Pointer instrument reading identification method and device and electronic equipment
CN116958125A (en) Electronic contest host power supply element defect visual detection method based on image processing
CN109977959A (en) A kind of train ticket character zone dividing method and device
CN114187363A (en) Method and device for obtaining radial distortion parameter value and mobile terminal
CN110543798B (en) Two-dimensional code identification method and device
EP2853089B1 (en) Pattern processing apparatus, pattern processing method, and pattern processing program
CN107644442B (en) Spatial position calibration method of double-camera module
CN110751690B (en) Visual positioning method for milling machine tool bit

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