KR101340180B1 - Apparatus and method of diagnosing health by separating region of tongue coating from region of tongue substance - Google Patents

Abstract

The present invention relates to a device and a method for diagnosing health by dividing tongue / lingual area, and comprising a tongue area extracting unit for extracting tongue area from a user image, and in the extracted tongue area, tongue and tongue that separate tongue and tongue A separator may include a tongue feature extractor that extracts tongue feature information from the separated tongue and tongue, and a diagnosis result generator that generates a diagnosis result based on the extracted tongue feature information.

Description

Apparatus and method for diagnosing health by dividing tongue and tongue area {APPARATUS AND METHOD OF DIAGNOSING HEALTH BY SEPARATING REGION OF TONGUE COATING FROM REGION OF TONGUE SUBSTANCE}

The present invention is to divide the tongue and tongue, and the present invention discloses a technical idea of relatively separating tongue and tongue, tongue and tongue, tongue and tongue.

In modern society, interest in health is increasing day by day.

Along with this era of interest, as technologies such as data analysis method and tool by real-time data collection are advanced, it becomes possible to monitor health condition and provide personalized health care service.

In recent years, the condition of the tongue has been used as a variety of measures to determine the health.

Healthy tongues usually have a pale pink or slightly white mixed pink color, which is said to be white, black or hairy over the entire or wide area of the dorsal side of the tongue. Pregnancy is often white, off-white, or yellowish white, in which case it is called white.

In addition, when the tongue looks black, it is called black snow or black snow, but it is often caused by bacteria or fungi, and may be discolored by tobacco or drugs to take. Some unknown causes of thickening of epithelial tissues and, in rare cases, thickening of epithelial tissues, are often early signs of tongue cancer.

In recent years, by photographing and analyzing the tongue, it is possible to objectively analyze the tongue.

According to an embodiment of the present invention, a device for diagnosing health by dividing tongue / lingual region may include a tongue region extracting unit that extracts a tongue region from a user image, and separates tongue and tongue from the extracted tongue region. It may include a tongue / tongue tongue separation unit, tongue tongue feature extraction unit for extracting tongue feature information from the separated tongue and tongue tongue, and a diagnosis result generation unit for generating a diagnostic result based on the extracted tongue feature information.

In accordance with an embodiment of the present invention, a method for diagnosing health by dividing tongue / lingual area may include extracting a tongue area from a user image, separating the tongue and tongue from the extracted tongue area, and separating the tongues. And extracting snow feature information from the tongue, and generating a diagnosis result based on the extracted snow feature information.

According to an embodiment of the present invention, it is possible to exclude the unstable region, and to separate the lingual / sulfur region by minimizing the influence of the lighting conditions or the surrounding environment.

According to an embodiment of the present invention, it is possible to obtain a divided region of tongue, and tongue, tongue, tongue, tongue, and tongue.

1 is a block diagram illustrating a device for diagnosing tongue / sulfur area according to an embodiment of the present invention.
Figure 2 is a view for explaining an embodiment of separating the tongue and sultry according to an embodiment of the present invention.
3 is a view specifically illustrating a snow region extraction unit according to an embodiment of the present invention.
4 is a view for explaining an embodiment of converting a user image into LAB color.
5 is a view for explaining an embodiment of extracting a midpoint from a user image according to an embodiment of the present invention.
FIG. 6 is a diagram for explaining an embodiment of extracting initial points from a plurality of divided regions.
FIG. 7 is a view for explaining an embodiment of extracting a boundary of a snow region using extracted initial points.
8 and 9 are diagrams illustrating the separation of tongues and tongues based on color.
FIG. 10 is a block diagram illustrating a method of diagnosing tongue / sulfur area according to an embodiment of the present invention.

Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the accompanying drawings.

In the following description of the present invention, detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear. The terminologies used herein are terms used to properly represent preferred embodiments of the present invention, which may vary depending on the user, the intent of the operator, or the practice of the field to which the present invention belongs. Therefore, the definitions of the terms should be made based on the contents throughout the specification. Like reference symbols in the drawings denote like elements.

FIG. 1 is a block diagram illustrating an apparatus for diagnosing tongue / sulfur area according to an embodiment of the present invention 100.

According to an embodiment of the present invention, the tongue / lingual area of the medical examination 100 is for dividing the tongue and the tongue area, and may relatively distinguish the tongue of the tongue and tongue, tongue and tongue, tongue and tongue.

To this end, the tongue / tongue area classification health diagnosis apparatus 100 according to an embodiment of the present invention is a tongue area extraction unit 110, tongue / tongue tongue separation unit 120, tongue characteristics extracting unit 130, and diagnosis The result generator 140 may be included.

The snow area extraction unit 110 according to an embodiment of the present invention may extract snow area from the user image.

Specifically, the tongue area extraction unit 110 according to an embodiment of the present invention uses the at least one of shadow information and color information in the user image, for example, a face image of the user, to primary the tongue area. Can be extracted.

Then, the snow region extraction unit 110 according to an embodiment of the present invention detects an optimal boundary by checking geometric information of a boundary value in the first extracted snow region, and based on the detected optimal boundary, The region may be extracted secondarily, and the snow region may be detected using the second extracted result.

The tongue / sulfur separator 120 according to an embodiment of the present invention may separate tongue and tongue from the extracted tongue region. To this end, the tongue / tongue separation unit 120 according to an embodiment of the present invention can correct the change in color using [Equation 1] when separating the tongue area for tongue diagnosis.

[Equation 1]

At this time, La is the peripheral color, Ld is the color of the surface, Ls can be interpreted as the color of the light source. Also, R is a reflection vector and L can be interpreted as a lighting vector.

That is, the tongue / tongue separator 120 according to an embodiment of the present invention determines at least one of diffuse reflection and specular reflection according to the change of the light source, in the extracted snow region, And tongues can be separated.

In addition, the tongue / tongue separation unit 120 according to an embodiment of the present invention is diffuse reflection and specular reflection (Specific reflection) is affected by changes in the surface of the tongue and light source.

That is, color reflects all of ambient reflection, diffuse reflection, and specular reflection, but it is generally avoided the direct reflection of light in the design, so there is no glare of direct light. The term reflection can be taken into account.

Therefore, when the tongue / tongue separation unit 120 according to an embodiment of the present invention to distinguish the tongue area for tongue diagnosis, it may correct the change in color using [Equation 2].

&Quot; (2) "

For example, the tongue / tongue separation unit 120 according to an embodiment of the present invention may separate the tongue and tongue in consideration of the case where the diffuse reflection is more than a threshold value.

In addition, the tongue / tongue separation unit 120 according to an embodiment of the present invention binarizes the color element (or color element) in the extracted tongue area, and the tongue and tongue on the basis of the binarized color element Can be separated.

As used herein, the color element may include at least one of RGB (Red, Green, Blue), Lab (Luminosity, Red, Yellow), HSV (HSI, Hue, Saturation, Intensity), and YCbCr. .

Snow feature extraction unit 130 according to an embodiment of the present invention may extract snow feature information from the separated tongue and tongue.

For example, the snow feature extraction unit 130 according to an embodiment of the present invention may include at least one of snow quality color information, snow color information, area size, area ratio, snow area horizontal / vertical ratio, and presence / absence information. It can extract as characteristic information.

The diagnosis result generator 140 according to an embodiment of the present invention may generate a diagnosis result based on the extracted feature characteristic information.

Diagnostic result generation unit 140 according to an embodiment of the present invention can be separated by comparing the color of the tongue / tongue. In addition, the separated results may be used to classify the statistical characteristics of the color vector space.

In addition, the diagnosis result generating unit 140 according to an embodiment of the present invention may classify the form of the tongue, and may classify by inputting a vector representing the form of the tongue.

Figure 2 is a view for explaining an embodiment of separating the tongue and sultry according to an embodiment of the present invention.

According to an embodiment of the present invention, the device for diagnosing the tongue / lingual area of the tongue is first extracted snow area by using shadow and color information from the user image as shown by reference numeral 210, and the optimal boundary detection is performed through boundary value geometric information. Regions can be extracted secondarily.

The snow region, which has progressed up to the second extraction, may be separated from the user image as indicated by reference numeral 220.

Next, according to an embodiment of the present invention, the tongue / lingual area diagnosis apparatus for diagnosing the health may divide the tongue area separated as shown by reference numeral 220 and tongue image 240 of the tongue.

At this time, the tongue / tongue area classification medical examination device according to an embodiment of the present invention can separate the tongue and tongue through the color classification and chroma saturation binarization.

3 is a view illustrating in detail the snow area extraction unit 300 according to an embodiment of the present invention.

The snow area extraction unit 300 according to an embodiment of the present invention converts the user image into LAB color to extract only the snow area from the user image, and matches the histogram of the user image for each domain to perform preprocessing. It may include a pre-processing unit 310.

In addition, the snow region extraction unit 300 according to an embodiment of the present invention extracts a midpoint to detect a horizontal symmetry axis and a vertical edge point in the preprocessed user image in order to extract only the snow region from the user image. The unit 320 detects the horizontal symmetry axis and the vertical edge point, and then divides the user image into a plurality of regions. The initial segment extracts initial points from the divided regions. The apparatus may further include a point extractor 340, a boundary extractor 350 that extracts a boundary of the snow region based on the extracted initial points, and a smoothing processor 360 that smoothes the extracted boundary. .

According to one side of the present invention, the smoothing processing unit 360 according to an embodiment of the present invention may smooth the extracted boundary using a Catmull-Rom Spline curve.

4 is a view for explaining an embodiment of converting a user image into LAB color.

Referring to FIG. 4, according to an embodiment of the present disclosure, the apparatus for diagnosing lingual / lingual areas may convert the user image 410 into LAB color using a preprocessor. In other words, according to an embodiment of the present invention, the lingual / lingual area medical examination apparatus converts the user image 410 into a LAB color, and converts the user images 420, 430, and 440 into the LAB color. The histogram can be matched for each domain and preprocessed.

LAB color is a model re-established by the International Standard for Color Measurement, and can be used to correct color differences that vary slightly depending on the type of the monitor or the type of print.

Therefore, the tongue / sulfur area classification medical examination apparatus according to an embodiment of the present invention uses the preprocessor to display the collective color of the tongue / sulfur image according to the type of monitor or printed matter without distorting and expressing them collectively. It can be done.

5 is a view for explaining an embodiment of extracting a midpoint from a user image according to an embodiment of the present invention.

The center extractor according to an embodiment of the present invention may extract a center point from the user image 500 as shown by reference numeral 510.

To this end, the midpoint extractor according to an embodiment of the present invention may extract a midpoint by detecting a horizontal symmetry line and a vertical edge point in the preprocessed user image 500. .

FIG. 6 is a diagram for explaining an embodiment of extracting initial points from a plurality of divided regions.

A snow area extractor according to an embodiment of the present invention divides the user image into a plurality of areas by using the area divider, through the detected horizontal symmetry axis and the vertical edge point as shown by reference numeral 610. An initial point extractor may extract at least one initial point 621 from the plurality of divided regions, as shown at 620.

According to an embodiment of the present invention, the region dividing unit may divide the region into eight as shown by reference numeral 610.

In addition, the initial point extractor according to an embodiment of the present invention may extract the at least one initial point 621 from each of the divided regions.

FIG. 7 is a view for explaining an embodiment of extracting a boundary of a snow region using extracted initial points.

As shown in FIG. 7, the boundary extractor according to an embodiment of the present invention may extract the boundary 710 of the snow area based on the initial points extracted from the user image 700.

The boundary extractor according to an embodiment of the present disclosure may extract the boundary 710 of the snow region using a first upper limit boundary value similar to the gray level and a second upper limit boundary value of the minimum gray level.

The boundary extractor according to an embodiment of the present invention maximizes the difference between the shadow values at the boundary surfaces of the lower boundaries with maximum color difference and the minimum boundaries with local minimum pt. The boundary surface 710 may be selected to extract the boundary 710 of the snow region.

8 and 9 are diagrams illustrating the separation of tongues and tongues based on color.

In the structure of the tongue, there is a significant change in diffusion reflection at the edge portion, and the other regions have relatively low change in diffusion reflection and uniform illumination conditions.

Therefore, in the extracted snow region 810 image, the area of diffusion and the tongue are avoided to avoid the area where the change is large, and the area where the change of the region is severe changes in the shade of LB of LAB or V of HSV. The outer area can be excluded as the boundary between areas is large.

That is, the tongue / sulphur separation unit according to an embodiment of the present invention may extract the tongue region by dividing the color by the relative difference of the color of the inner region, and the color coordinates include various color coordinates such as RGB, LAB, and HSV. Can be used.

To this end, the tongue / tongue separation unit according to an embodiment of the present invention, in the extracted tongue area 810, Hue image 820, Saturation image 830, and various such as LAB, HSV The coordinate image 840 to which the color coordinates are applied may be generated to separate the tongue and tongue.

If there is a difference in shadow in the L region and the V region, it may be assumed that the flexion of the tongue is severe.

According to one embodiment of the present invention, the lingual / lingual tongue section finds an area where the flexion of the tongue is severe due to the difference in shadow in the L area. For this purpose, the mean of m * n area values is obtained, as shown at 900. Difference between 1 910 and Zone 2 920, Zone 3 930, and Zone 4 940, Difference between Zone 2 920 and Zone 3 930, Zone 4 940, Zone A place representing the difference between the largest region among the differences between the 3 930 and the region 4 940 may be detected. In addition, if the value at the point showing the difference of the largest detected area exceeds a certain threshold, the position of this position (x, y) can be designated as the boundary position.

For example, m and n indicated at 900 may be interpreted as the same value, and c and d may also be interpreted as the same value. In addition, c and d may be represented by zero.

According to one embodiment of the present invention, the tongue / sulphur separation unit may proceed while increasing values of (x, y) one by one for all regions.

According to one embodiment of the present invention, the lingual / lingual tongue is referred to as the boundary of the lateral surface of the tongue where the tongue is bent, and the areas within the tongue are connected to each other by connecting the points. This area can be specified as a mask.

In accordance with an embodiment of the present invention, the tongue / tongue separation unit can search within the specified mask again by the above method.

That is, within the designated mask as shown by the reference numeral 900, the average of the m * n area value is obtained, and the difference between the area 1 (910), the area 2 (920), the area 3 (930), the area 4 (940), If the difference between the region 2 920 and the region 3 930 and the region 4 940 and the difference between the region 3 930 and the region 4 940 exceeds a predetermined threshold, a state exists in the region. I can think of it.

와

의 조건을 만족하는 경우에 설 영역에서 황태가 존재한다고 판단할 수 있다.For example, on RGB coordinates

Wow

If the condition is met, it can be determined that the presence of the Emperor in the snow area.

의 조건을 만족하는 경우에 황태가 존재한다고 판단할 수 있고, HSV 좌표 상에서

의 조건을 만족하는 경우에는 백태가 존재한다고 판단할 수 있다.Next, on the YCbCr coordinates,

If the condition is satisfied, it can be judged that there is an Emperor, and on the HSV coordinate

If the condition is satisfied, it can be determined that the white state exists.

For example, if the difference between R and G exceeds the threshold at the same time, it can be regarded as the dominant region. These color coordinates are also RGB, and LAB (YCbCr) and HSV various color coordinates and thresholds can be used.

This threshold may be used in one color gamut and several color gamuts simultaneously, depending on the type of setting.

All of these values are relative differences regardless of the environment.

FIG. 10 is a block diagram illustrating a method of diagnosing tongue / sulfur area according to an embodiment of the present invention.

According to an embodiment of the present invention, the tongue / lingual area diagnosis health diagnosis method may extract a tongue area from a user image (step 1001).

For example, in accordance with an embodiment of the present invention, in order to extract the tongue area / lingual area, the health diagnosis method converts the user image into LAB color, and performs preprocessing by matching the histogram of the user image for each domain. Can be.

According to an embodiment of the present invention, a method for diagnosing a tongue / sull area according to an embodiment of the present invention detects a horizontal symmetry axis and a vertical edge point in the preprocessed user image, and detects the horizontal symmetry axis and the vertical edge point. The user image may be divided into a plurality of regions. Accordingly, according to an embodiment of the present invention, a method for diagnosing a tongue / lingual tongue area may extract initial points from the divided plurality of areas, and extract a boundary of the tongue area based on the extracted initial points. Can be.

In addition, according to an embodiment of the present invention, the tongue / tongue area classification health diagnosis method may be to extract the boundary of the tongue area by smoothing the extracted boundary.

According to an embodiment of the present invention, the lingual / lingual area of the lingual area may be separated from the lingual area and the tongue in the extracted tongue area (step 1002).

According to one embodiment of the present invention, the tongue / lingual area diagnosis health diagnosis method may extract tongue property information from the separated tongue and tongue (step 1003).

According to an embodiment of the present invention, the tongue / lingual area diagnosis health diagnosis method includes at least one of tongue color information, tongue color information, area size, area ratio, tongue area horizontal / vertical ratio, and presence / absence information. Can be extracted as.

According to an embodiment of the present invention, the tongue / lingual area diagnosis health diagnosis method may generate a diagnosis result based on the extracted tongue feature information (step 1004).

The method for diagnosing health by dividing the lingual / lingual area in accordance with an embodiment of the present invention may be implemented in the form of program instructions that can be executed by various computer means and recorded in a computer readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions recorded on the medium may be those specially designed and constructed for the present invention or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

As described above, the present invention has been described by way of limited embodiments and drawings, but the present invention is not limited to the above embodiments, and those skilled in the art to which the present invention pertains various modifications and variations from such descriptions. This is possible.

Therefore, the scope of the present invention should not be limited to the described embodiments, but should be determined by the equivalents of the claims, as well as the claims.

100: tongue / tongue area separation medical examination device
110: snow area extraction unit
120: Sulfur / Salt Separator
130: feature extraction section
140: diagnosis result generator

Claims (15)

Snow area extraction unit for extracting snow area from the user image;
A tongue and tongue tongue separator that binarizes the color elements in the extracted tongue region and separates tongue tongue and tongue tongue based on the binarized color elements;
A tongue feature extraction unit for extracting tongue feature information from the separated tongues and tongues; And
A diagnosis result generation unit generating a diagnosis result based on the extracted snow feature information
Apparatus for diagnosing health by dividing the lingual and lingual areas comprising a. The method of claim 1,
The snow area extraction unit,
A center extractor for detecting a horizontal symmetry axis and a vertical edge point in the user image;
An area divider which detects the horizontal symmetry axis and the vertical edge point and divides the user image into a plurality of areas;
An initial point extracting unit configured to extract initial points from the plurality of divided regions;
A boundary extractor configured to extract a boundary of the snow area based on the extracted initial points; And
Smoothing processing unit for smoothing the extracted boundary
Apparatus for diagnosing health by dividing the lingual and lingual areas comprising a. 3. The method of claim 2,
The boundary extractor,
And an apparatus for diagnosing health by dividing a tongue and a tongue area in which the boundary of the tongue area is extracted based on the extracted initial point and a boundary surface that maximizes the difference between shaded values in a preprocessed user image. 3. The method of claim 2,
The smoothing processing unit,
Apparatus for diagnosing health by classifying tongue and tongue areas by smoothing the extracted boundary using a Catmull-Rom Spline curve. The method of claim 1,
The snow area extraction unit,
Apparatus for diagnosing health by dividing the tongue and tongue area by first extracting the tongue area by using at least one of shadow information and color information in the user image. The method of claim 5,
The snow area extraction unit,
Identifying the optimal boundary by checking the geometric information of the boundary value in the first extracted tongue area, and secondly extracting the tongue area based on the detected optimal boundary to separate the tongue and tongue area Device to diagnose. The method of claim 1,
The tongue and tongue tongue separation unit,
Diagnose health by dividing tongue and tongue areas by determining at least one of diffuse reflection and specular reflection according to the change of the light source, and separating tongues and tongues from the extracted tongue area. Device. The method of claim 7, wherein
The tongue and tongue tongue separation unit,
A device for diagnosing health by dividing the tongue and the tongue of the tongue, wherein the tongue and tongue are separated in consideration of the case where the diffuse reflection is greater than or equal to a threshold. delete The method of claim 1,
The tongue feature extraction unit extracts at least one of tongue quality information, tongue color information, area size, area ratio, tongue area horizontal / vertical ratio, and presence / absence information as tongue feature information. A device for diagnosing health by dividing it. Extracting a snow area from a user image;
Binarizing color elements in the extracted tongue area and separating tongue quality and tongue shape based on the binarized color elements;
Extracting snow feature information from the separated tongues and tongues; And
Generating a diagnosis result based on the extracted snow feature information;
Method for diagnosing health by dividing the lingual and lingual area comprising a. 12. The method of claim 11,
Extracting the snow area,
Detecting a horizontal symmetry axis and a vertical edge point in the user image;
Dividing the user image into a plurality of regions by detecting the horizontal symmetry axis and the vertical edge point;
Extracting initial points from the plurality of divided regions;
Extracting a boundary of the snow region based on the extracted initial points; And
Smoothing the extracted boundary;
Method for diagnosing health by dividing the lingual and lingual area comprising a. 12. The method of claim 11,
Extracting the snow area,
Firstly extracting the snow region using at least one of shadow information and color information in the user image; And
Detecting an optimal boundary by checking geometric information of a boundary value in the first extracted snow region and secondly extracting the snow region based on the detected optimal boundary
Method for diagnosing health by dividing the lingual and lingual region, characterized in that it comprises a. 12. The method of claim 11,
The step of separating the tongue and tongue tongue,
Determining at least one of diffuse reflection and specular reflection according to the change of the light source, and separating the tongue and tongue from the extracted tongue region
Method for diagnosing health by dividing the lingual and lingual region, characterized in that it comprises a. A computer-readable recording medium having recorded thereon a program for performing the method of claim 11.

Images