CN111222371A - Sublingual vein feature extraction device and method - Google Patents
Sublingual vein feature extraction device and method Download PDFInfo
- Publication number
- CN111222371A CN111222371A CN201811419050.8A CN201811419050A CN111222371A CN 111222371 A CN111222371 A CN 111222371A CN 201811419050 A CN201811419050 A CN 201811419050A CN 111222371 A CN111222371 A CN 111222371A
- Authority
- CN
- China
- Prior art keywords
- vein
- tongue
- sublingual
- image
- ventral surface
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 210000003462 vein Anatomy 0.000 title claims abstract description 296
- 238000000605 extraction Methods 0.000 title claims abstract description 49
- 238000000034 method Methods 0.000 title claims abstract description 29
- 230000011218 segmentation Effects 0.000 claims abstract description 47
- 238000012549 training Methods 0.000 claims abstract description 29
- 230000000877 morphologic effect Effects 0.000 claims abstract description 6
- 238000004364 calculation method Methods 0.000 claims description 37
- 238000004422 calculation algorithm Methods 0.000 claims description 20
- 230000006870 function Effects 0.000 claims description 17
- 238000012545 processing Methods 0.000 claims description 16
- 238000004590 computer program Methods 0.000 claims description 12
- 238000003708 edge detection Methods 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 5
- 238000003745 diagnosis Methods 0.000 description 21
- 239000003814 drug Substances 0.000 description 16
- 239000008280 blood Substances 0.000 description 5
- 210000004369 blood Anatomy 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000001514 detection method Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 206010041956 Stasis syndrome Diseases 0.000 description 2
- 208000005634 blind loop syndrome Diseases 0.000 description 2
- 230000004069 differentiation Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000001815 facial effect Effects 0.000 description 2
- 208000011580 syndromic disease Diseases 0.000 description 2
- 206010062717 Increased upper airway secretion Diseases 0.000 description 1
- 208000019693 Lung disease Diseases 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 201000011510 cancer Diseases 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 208000019622 heart disease Diseases 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 208000019423 liver disease Diseases 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000004089 microcirculation Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000006740 morphological transformation Effects 0.000 description 1
- 230000001717 pathogenic effect Effects 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 208000026435 phlegm Diseases 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 230000000452 restraining effect Effects 0.000 description 1
- 238000004381 surface treatment Methods 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 230000001839 systemic circulation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/14—Vascular patterns
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Probability & Statistics with Applications (AREA)
- Human Computer Interaction (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a sublingual vein feature extraction device and a method, and the method comprises the following steps: training by using different tongue surface sample images to generate a tongue ventral surface detector; detecting a lingual surface including lips from the lingual surface image by using a lingual surface detector; intercepting the ventral surface of the tongue according to the position information of the ventral surface of the tongue; carrying out threshold segmentation treatment on the ventral surface of the tongue to obtain a shadow area and a tooth area of the ventral surface of the tongue; creating a tongue ventral surface segmentation template by using the shadow area and the tooth area; segmenting a tongue ventral surface image from the tongue surface image by using a tongue ventral surface segmentation template; classifying the tongue ventral surface image to obtain a classification result image; separating the left and right sublingual veins from the classification result image, and performing morphological treatment on the left and right sublingual veins to obtain left and right sublingual vein templates; calculating a sublingual vein image by using the sublingual left and right vein template, and extracting the sublingual vein characteristics from the sublingual vein image. The sublingual vein feature extraction method improves the accuracy of sublingual vein feature extraction.
Description
Technical Field
The invention relates to the technical field of tongue surface treatment in traditional Chinese medicine, in particular to a sublingual vein feature extraction device and method.
Background
The sublingual collaterals diagnosis method has certain auxiliary diagnosis value for various diseases and symptoms like tongue diagnosis. The syndrome mainly refers to blood stasis syndrome, especially blood stasis syndrome; has especially diagnosis value in malignant tumor, heart and lung diseases, liver diseases and hematopathy. Although it is used for non-specific diagnosis, it can better reflect the overall situation of the patient on pathogenic factors, and has great significance especially on whether qi and blood are excessive and deficient and harmonized, whether channels and collaterals are unobstructed, and whether phlegm and blood stasis are present. According to the extensive studies in traditional Chinese medicine, it mainly reflects the changes in the state of the systemic circulation and microcirculation and in the blood. If the doctor has a certain experience, the doctor can eliminate the influences of age, individual difference, climate interference and the like through detailed inspection, and the result can make up the deficiency of the traditional tongue diagnosis and provide more important information for medical clinic in syndrome differentiation.
However, the diagnosis result obtained by the traditional tongue diagnosis method is often affected by the factors such as the experience accumulation of the doctor and the current environment of the patient, and the subjective dependence is strong, and the basis of objectivity and quantification is lacked. The deep development of the sublingual collaterals diagnosis method still depends on the careful exploration of basic research and the objectification of clinical observation, so that the method and the device are particularly important to create new instruments and methods by using high-tech means, form unified and normative objective indexes and observation methods, find out the rules and explain the mechanism, and are practical in clinic.
The development of research work facing the computerized traditional Chinese medicine sublingual collaterals diagnosis method can further promote the convergence development of modern information science and the traditional Chinese medicine, solve the important basic problem of restraining the exertion of the characteristic advantages of the traditional Chinese medicine for the standardization of the traditional Chinese medicine differentiation and the modernization of the traditional Chinese medicine clinical, teaching and scientific research means, realize the modernization of the traditional Chinese medicine, have important theoretical value and practical significance, and are an important link of the modernization of the traditional Chinese medicine tongue diagnosis. At present, computer methods such as image processing and pattern recognition provide reference bases for traditional Chinese medicine diagnosis technologies, however, in the prior art, the accuracy in tongue and ventral surface segmentation is not high, and sublingual veins segmented in the aspect of sublingual vein segmentation are not complete enough, so that the extracted sublingual vein features are not accurate enough, and the accuracy of a doctor on the judgment result of the sublingual vein pulse diagnosis in traditional Chinese medicine is influenced.
Disclosure of Invention
The invention mainly aims to provide a sublingual vein feature extraction device and method applied to a traditional Chinese medicine facial diagnosis, and aims to solve the technical problem that the accuracy of sublingual vein feature extraction is low in the prior art.
To achieve the above object, the present invention provides a sublingual vein feature extraction device, comprising a processor adapted to implement various computer program instructions and a memory adapted to store a plurality of computer program instructions, the computer program instructions being loaded by the processor and performing the steps of: inputting different tongue surface sample images through an input unit to construct a plurality of positive and negative samples; processing a plurality of positive and negative samples by using an opencv _ createsamples program in an opencv open source library to generate a training data set; training the training data set by using an opencv _ traincacade program in an opencv open source library to generate a tongue ventral surface detector; detecting the ventral and lingual surfaces containing lips from the lingual surface image to be detected by using a ventral and lingual surface detector; determining the position information of the lingual surface and the ventral surface based on the lingual surface image to be detected, and intercepting the lingual surface and the ventral surface containing lips according to the position information of the lingual surface and the ventral surface; performing threshold segmentation processing on the intercepted tongue ventral surface to obtain a shadow area and a tooth area of the tongue ventral surface; creating a tongue ventral surface segmentation template by using a shadow area and a tooth area of the tongue ventral surface; inputting a tongue surface image to be detected and a tongue ventral surface segmentation template into a grabCut function in an opencv open source library to segment the tongue ventral surface image; classifying the tongue ventral surface images by using a kmeans clustering algorithm in an opencv open source library to obtain classified result images; separating the left and right sublingual veins from the classification result image, and performing morphological treatment on the left and right sublingual veins to obtain left and right sublingual vein templates; calculating a sublingual vein image by using the sublingual left and right vein template, and extracting sublingual vein features from the sublingual vein image.
Preferably, the step of creating a lingual ventral surface segmentation template by using the dark shadow area and the tooth area of the lingual ventral surface comprises: extracting a contour line of a tongue ventral surface shadow area by using a canny edge detection algorithm, extracting all n coordinate points of the contour line, and pairing all the n coordinate points to form n (n-1)/2 edges; for each edge, checking whether the remaining (n-2) points are on the same side of the edge; if all the points are on one side of the edge, adding the edge into the convex hull set until all the edges are traversed, and taking the convex hull set as the contour line of the tongue ventral surface shadow area; creating a single-channel template image with the size of the tongue surface sample image, and mapping the contour line of the tongue ventral surface shadow area to the corresponding position of the single-channel template image; setting all pixel values of the non-white area inside the contour line as 1, setting all pixel values of the white area inside the contour line as 3, setting all pixel values of the area outside the contour line as 0, and setting the pixel value of the corresponding position of the single-channel template image as 0 according to the lingual tooth area to obtain the lingual segmentation template.
Preferably, the step of classifying the tongue ventral surface image by using a kmeans clustering algorithm in an opencv open source library to obtain a classification result image comprises the following steps: converting the RGB tongue ventral surface image into an Lab color space, and classifying the tongue ventral surface image by using a kmeans clustering algorithm in an opencv open source library according to the pixel values of a color channel a and a color channel b of the tongue ventral surface image in the Lab color space to obtain a classification result image.
Preferably, the step of calculating the sublingual vein image by using the sublingual left and right vein template comprises: and respectively performing AND operation on the sublingual left and right vein templates and each color channel of the tongue ventral surface image, and combining operation results to obtain the sublingual vein image.
Preferably, the sublingual vein features include R, G, B color values and H, S, V color values of the sublingual vein, a length and a width of the left vein, a length and a width of the right vein, a tongue length ratio of the left vein, and a tongue length ratio of the right vein, wherein:
the calculation method for extracting the R color value of the sublingual vein comprises the following steps: dividing the sum of all pixel values of an R channel of the sublingual vein image by the number of sublingual vein pixels of the R channel to obtain an R color value; G. b, H, S, V color values are calculated according to the method of extracting R color values;
the calculation method for extracting the length and width of the left vein comprises the following steps: calculating the minimum circumscribed rectangle of the left vein in the sublingual vein image, wherein the length and the width of the rectangle are respectively the length and the width of the left vein under the tongue, and the calculation step of the minimum circumscribed rectangle of the left vein comprises the following steps:
(1) extracting a left vein contour line and calculating a left vein convex hull according to the contour line;
(2) randomly selecting an edge AB on the left vein convex hull as an initial edge, wherein A and B are a left end point and a right end point, rotating the AB by an angle theta by taking the end point A as a center, and enabling the edge AB to be parallel to an x axis of a coordinate transverse axis, so that all points of the left vein convex hull rotate by the angle theta around the point A;
(3) taking the AB edge as the upper boundary or the lower boundary of the circumscribed rectangle, finding a point with the minimum y value or the maximum y value on the left vein convex hull, making a straight line parallel to the x axis through the point to determine the lower boundary or the upper boundary of the circumscribed rectangle, finding a left side point with the minimum x value and a right side point with the maximum x value on the left vein convex hull, respectively making two straight lines perpendicular to the x axis through the left side point and the right side point to determine the left boundary and the right boundary of the circumscribed rectangle, thus obtaining the circumscribed rectangle, and calculating and storing the endpoint coordinates of the AB edge, the length, the width and the area of the circumscribed rectangle;
(4) sequentially selecting the next edge BC on the left vein convex hull, and repeating the steps (2) to (3) to search for the next external rectangle until all edges on the left vein convex hull are traversed;
(5) comparing the areas of all the external rectangles, and finding out the external rectangle with the smallest area as the smallest external rectangle of the left vein;
the calculation method for extracting the length and width of the right vein is the same as the calculation method for extracting the length and width of the left vein;
the calculation method for extracting the left vein-tongue length ratio comprises the following steps: the ratio of the input tongue length divided by the length of the sublingual left vein is recorded as the tongue length ratio of the left vein;
the calculation method for extracting the length ratio of the right vein tongue comprises the following steps: the ratio of the input tongue length divided by the length of the sublingual right vein was recorded as the tongue length ratio of the right vein.
On the other hand, the invention also provides a method for extracting the characteristics of the sublingual veins, which comprises the following steps: inputting different tongue surface sample images through an input unit to construct a plurality of positive and negative samples; processing a plurality of positive and negative samples by using an opencv _ createsamples program in an opencv open source library to generate a training data set; training the training data set by using an opencv _ traincacade program in an opencv open source library to generate a tongue ventral surface detector; detecting the ventral and lingual surfaces containing lips from the lingual surface image to be detected by using a ventral and lingual surface detector; determining the position information of the lingual surface and the ventral surface based on the lingual surface image to be detected, and intercepting the lingual surface and the ventral surface containing lips according to the position information of the lingual surface and the ventral surface; performing threshold segmentation processing on the intercepted tongue ventral surface to obtain a shadow area and a tooth area of the tongue ventral surface; creating a tongue ventral surface segmentation template by using a shadow area and a tooth area of the tongue ventral surface; inputting a tongue surface image to be detected and a tongue ventral surface segmentation template into a grabCut function in an opencv open source library to segment the tongue ventral surface image; classifying the tongue ventral surface images by using a kmeans clustering algorithm in an opencv open source library to obtain classified result images; separating the left and right sublingual veins from the classification result image, and performing morphological treatment on the left and right sublingual veins to obtain left and right sublingual vein templates; calculating a sublingual vein image by using the sublingual left and right vein template, and extracting sublingual vein features from the sublingual vein image.
Preferably, the step of creating a lingual ventral surface segmentation template by using the dark shadow area and the tooth area of the lingual ventral surface comprises: extracting a contour line of a tongue ventral surface shadow area by using a canny edge detection algorithm, extracting all n coordinate points of the contour line, and pairing all the n coordinate points to form n (n-1)/2 edges; for each edge, checking whether the remaining (n-2) points are on the same side of the edge; if all the points are on one side of the edge, adding the edge into the convex hull set until all the edges are traversed, and taking the convex hull set as the contour line of the tongue ventral surface shadow area; creating a single-channel template image with the size of the tongue surface sample image, and mapping the contour line of the tongue ventral surface shadow area to the corresponding position of the single-channel template image; setting all pixel values of non-white areas inside the contour line as 1, setting all pixel values of white areas inside the contour line as 3, setting all pixel values of areas outside the contour line as 0, and setting the pixel value of the corresponding position of the single-channel template image as 0 according to the lingual tooth area to obtain the lingual segmentation template.
Preferably, the step of classifying the tongue ventral surface image by using a kmeans clustering algorithm in an opencv open source library to obtain a classification result image comprises the following steps: converting the RGB tongue ventral surface image into an Lab color space, and classifying the tongue ventral surface image by using a kmeans clustering algorithm in an opencv open source library according to the pixel values of a color channel a and a color channel b of the tongue ventral surface image in the Lab color space to obtain a classification result image.
Preferably, the step of calculating the sublingual vein image by using the sublingual left and right vein template comprises: and respectively performing AND operation on the sublingual left and right vein templates and each color channel of the tongue ventral surface image, and combining operation results to obtain the sublingual vein image.
Preferably, the sublingual vein features include R, G, B color values and H, S, V color values of the sublingual vein, a length and a width of the left vein, a length and a width of the right vein, a tongue length ratio of the left vein, and a tongue length ratio of the right vein, wherein:
the calculation method for extracting the R color value of the sublingual vein comprises the following steps: dividing the sum of all pixel values of an R channel of the sublingual vein image by the number of sublingual vein pixels of the R channel to obtain an R color value; G. b, H, S, V color values are calculated according to the method of extracting R color values;
the calculation method for extracting the length and width of the left vein comprises the following steps: calculating the minimum circumscribed rectangle of the left vein in the sublingual vein image, wherein the length and the width of the rectangle are respectively the length and the width of the left vein under the tongue, and the calculation step of the minimum circumscribed rectangle of the left vein comprises the following steps:
(1) extracting a left vein contour line and calculating a left vein convex hull according to the contour line;
(2) randomly selecting an edge AB on the left vein convex hull as an initial edge, wherein A and B are a left end point and a right end point, rotating the AB by an angle theta by taking the end point A as a center, and enabling the edge AB to be parallel to an x axis of a coordinate transverse axis, so that all points of the left vein convex hull rotate by the angle theta around the point A;
(3) taking the AB edge as the upper boundary or the lower boundary of the circumscribed rectangle, finding a point with the minimum y value or the maximum y value on the left vein convex hull, making a straight line parallel to the x axis through the point to determine the lower boundary or the upper boundary of the circumscribed rectangle, finding a left side point with the minimum x value and a right side point with the maximum x value on the left vein convex hull, respectively making two straight lines perpendicular to the x axis through the left side point and the right side point to determine the left boundary and the right boundary of the circumscribed rectangle, thus obtaining the circumscribed rectangle, and calculating and storing the endpoint coordinates of the AB edge, the length, the width and the area of the circumscribed rectangle;
(4) sequentially selecting the next edge BC on the left vein convex hull, and repeating the steps (2) to (3) to search for the next external rectangle until all edges on the left vein convex hull are traversed;
(5) comparing the areas of all the external rectangles, and finding out the external rectangle with the smallest area as the smallest external rectangle of the left vein;
the calculation method for extracting the length and width of the right vein is the same as the calculation method for extracting the length and width of the left vein;
the calculation method for extracting the left vein-tongue length ratio comprises the following steps: the ratio of the input tongue length divided by the length of the sublingual left vein is recorded as the tongue length ratio of the left vein;
the calculation method for extracting the length ratio of the right vein tongue comprises the following steps: the ratio of the input tongue length divided by the length of the sublingual right vein was recorded as the tongue length ratio of the right vein.
Compared with the prior art, the sublingual vein feature extraction device and the sublingual vein feature extraction method have the advantages that the ventral and lingual surfaces containing lips are effectively detected by the ventral and lingual surface detector obtained through the training of a large number of lingual surface sample images, and the accuracy of the ventral and lingual surface segmentation is improved; the sublingual left and right veins are effectively segmented by carrying out color classification treatment on the tongue ventral surface image to create a sublingual left and right vein template, so that the sublingual left and right veins segmented in the vein segmentation aspect are more complete, the accuracy of characteristic extraction of the sublingual veins is improved, and a reference is provided for traditional Chinese medical students to the sublingual vein pulse diagnosis of the traditional Chinese medical science, so that the accuracy of the judgment result of the traditional Chinese medical students to the sublingual vein pulse diagnosis of the traditional Chinese medical science is assisted.
Drawings
FIG. 1 is a functional block schematic diagram of a preferred embodiment of the sublingual vein feature extraction device of the present invention;
FIG. 2 is a flow chart of a method of a preferred embodiment of the sublingual vein feature extraction method of the present invention;
FIG. 3 is a schematic diagram of a tongue ventral image segmented from an original tongue facial image;
FIG. 4 is a schematic diagram of a sublingual vein image segmented from a tongue ventral image;
fig. 5 is a schematic diagram of the extraction of sublingual vein features from sublingual vein images.
The objects, features and advantages of the present invention will be further described with reference to the following embodiments, which are illustrated in the accompanying drawings.
Detailed Description
To further explain the technical means and effects of the present invention adopted to achieve the predetermined objects, the following detailed description of the embodiments, structures, features and effects of the present invention will be given with reference to the accompanying drawings and preferred embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Referring to fig. 1, fig. 1 is a functional module schematic diagram of a sublingual vein feature extraction device according to a preferred embodiment of the invention. In this embodiment, the sublingual vein feature extraction device 1 may be a computer device having a data processing function and an image processing function, such as a personal computer, a workstation computer, a traditional Chinese medicine face imaging apparatus, and a traditional Chinese medicine four-diagnosis apparatus, which are equipped with the sublingual vein feature extraction system 10. In the present embodiment, the sublingual vein feature extraction device 1 comprises, but is not limited to, a sublingual vein feature extraction system 10, an input unit 11, a memory 12 adapted to store a plurality of computer program instructions, a processor 13 executing various computer program instructions, and an output unit 14. The input unit 11 may be an image input device such as a high-definition camera, and is configured to capture a tongue surface image and input the tongue surface image into the sublingual vein feature extraction device 1; the input unit 11 may also be an image reading device for reading a tongue surface image from a database in which the tongue surface image is stored and inputting the tongue surface image into the sublingual vein feature extraction apparatus 1. The memory 12 may be a read only memory ROM, a random access memory RAM, an electrically erasable programmable memory EEPROM, a FLASH memory FLASH, a magnetic or optical disk, or the like. The processor 13 is a Central Processing Unit (CPU), a Microcontroller (MCU), a data processing chip, or an information processing unit having a data processing function. The output unit 14 may be a display or a printer, and the like, and can output the extracted sublingual vein features to the display or the printer for printing, so that the doctor can provide clinical reference for the doctor to perform the tongue diagnosis, thereby assisting the doctor in determining the accuracy of the tongue diagnosis result.
In the present embodiment, the sublingual vein feature extraction system 10 is composed of program modules composed of a plurality of computer program instructions, including but not limited to a lingual training module 101, a lingual surface detection module 102, a lingual surface segmentation module 103, a sublingual vein segmentation module 104, and a sublingual vein feature extraction module 105. The module referred to in the present invention refers to a series of computer program instruction segments capable of being executed by the processor 13 of the sublingual vein feature extraction device 1 and performing a fixed function, which are stored in the memory 12, and the specific function of each module is described in detail below in conjunction with fig. 2.
Referring to fig. 2, it is a flow chart of the preferred embodiment of the sublingual vein feature extraction method of the invention. In this embodiment, the various method steps of the sublingual vein feature extraction method are implemented by a computer software program stored in a computer-readable storage medium (e.g., the memory 12) in the form of computer program instructions, which may include: read-only memory, random access memory, magnetic or optical disk, etc., which can be loaded by a processor (e.g., the processor 13) and which performs the following steps S21 through S32.
Step S21, inputting different tongue surface sample images to construct a plurality of positive and negative samples; in the present embodiment, the input unit 11 captures a large number of different tongue surface sample images by a high-definition camera or reads a large number of different tongue surface sample images from an external database, and inputs the images into the sublingual vein feature extraction system 10. The tongue ventral surface training module 101 constructs a plurality of positive and negative samples according to different tongue surface sample images, wherein the positive and negative samples comprise a plurality of positive samples and a plurality of negative samples, for example, 200 positive samples and 300 negative samples, one positive sample comprises image data of a tongue ventral surface area in a tongue surface sample image, and one negative sample comprises image data of a non-tongue ventral surface area in a tongue surface sample image. The tongue surface sample image input by the input unit 11 is sent to the tongue ventral surface training module 101 for training to obtain the tongue ventral surface detector. The invention adopts a large number of different tongue surface sample images to train a tongue ventral surface detector for identifying the tongue surface image, and the tongue ventral surface image of the lip can be detected as long as a doctor inputs the tongue surface image to be detected to the tongue ventral surface detector.
Step S22, processing a plurality of positive and negative samples by using an opencv _ createsamples program in an opencv open source library to generate a training data set; in this embodiment, the tongue ventral training module 101 processes a plurality of positive and negative samples by using an opencv _ createsamples program in an opencv open source library to generate a training data set. The opencv _ createsamples program is a general program for creating samples in an opencv open-source library, and a person skilled in the art can process a plurality of positive and negative samples to generate a training data set through the existing opencv _ createsamples program.
Step S23, training a training data set by using an opencv _ traincascade program in an opencv open source library to generate a tongue ventral surface detector; in this embodiment, the lingual ventral training module 101 trains the training data set to generate the lingual ventral detector using the opencv _ traincacade program in the opencv open source library. The opencv _ traincascade program is a general program for training classifiers in an opencv open-source library, and a person skilled in the art can train the training data set to generate the tongue ventral surface detector through the existing opencv _ traincascade program.
Step S24, detecting the ventral surface of the tongue including lips from the tongue image to be detected by using a ventral surface detector; specifically, when detecting and recognizing the lingual surface, the lingual surface detection module 102 first receives an image of the lingual surface to be detected from the input unit 11, for example, the image of the lingual surface to be detected is shown in fig. 3(a), and then detects the lingual surface including the lips from the image of the lingual surface to be detected by using the trained lingual surface detector, for example, the lingual surface including the lips is shown in fig. 3 (b).
Step S25, determining the position information of the lingual surface and the ventral surface based on the lingual surface image to be detected, and intercepting the lingual surface and the ventral surface containing lips according to the position information of the lingual surface and the ventral surface; specifically, the lingual ventral surface detection module 102 determines positional information rect (x, y, l, w) of the lingual ventral surface based on the position of the lingual ventral surface to be detected in the lingual surface image, and then intercepts the lingual ventral surface including the lips according to the positional information of the lingual ventral surface, wherein x and y represent coordinates of a top left corner vertex of the square frame in fig. 3(a), and l and w represent the length and width of the square frame.
Step S26, performing threshold segmentation processing on the intercepted tongue ventral surface to obtain a shadow area and a tooth area of the tongue ventral surface; in this embodiment, the ventral tongue surface segmentation module 103 performs threshold segmentation on the intercepted ventral tongue surface (including lips), and performs morphological transformation on the segmentation result to remove some small spot impurities in the ventral tongue surface image, so as to obtain a dark shadow region and a tooth region (a dark shadow region and a tooth region located in the lips) of the ventral tongue surface, respectively, as a white region in fig. 3(c) represents the dark shadow region of the ventral tongue surface, and a white region in fig. 3(d) represents the tooth region of the ventral tongue surface.
Step S27, creating a tongue ventral surface segmentation template by using the shadow area and the tooth area of the tongue ventral surface; in this embodiment, the tongue ventral surface segmentation module 103 performs convex hull calculation according to the tongue ventral surface shadow area to generate a contour line of the tongue ventral surface shadow area, and the generation result is a white contour line as shown in fig. 3 (e). The method comprises the following specific steps: the tongue ventral surface segmentation module 103 extracts the contour line of the tongue ventral surface shadow area by using a canny edge detection algorithm, then extracts all coordinate points (assuming that n points are provided in total) of the contour line, and pairwise pairs of all n extracted coordinate points form n x (n-1)/2 edges; for each edge, check again whether the remaining (n-2) points are on the same side of the edge; if all the points are on one side of the edge, adding the edge into a convex hull set to be used as a contour line of a tongue ventral surface shadow area until all the formed edges are traversed; creating a single-channel template image with the size of a tongue sample image, mapping the contour line of a tongue ventral dark shadow area to the corresponding position of the single-channel template image, setting all pixel values of non-white areas in the contour line to be 1, setting all pixel values of white areas in the contour line to be 3, setting all pixel values of outer areas of the contour line to be 0, and setting the pixel values of the corresponding position of the single-channel template image to be 0 according to the tongue ventral tooth area, thereby obtaining the tongue ventral segmentation template. Where 0 represents a determined background pixel value (not the ventral tongue), 1 represents a determined foreground pixel value (the ventral tongue), 2 represents an uncertain background pixel value (possibly the non-ventral tongue), and 3 represents an uncertain foreground pixel value (possibly the ventral tongue).
Step S28, inputting the tongue surface image and the tongue ventral surface segmentation template to be detected into a grabCut function of an opencv open source library to segment the tongue ventral surface image; in the present embodiment, the lingual surface segmentation module 103 inputs the lingual surface image and the lingual surface segmentation template into the grabCut function of the opencv open source library to segment the lingual surface image. The grabCut function of the opencv open source library is an existing tongue and ventral surface segmentation function, the grabCut function can be used for image segmentation, and a person skilled in the art inputs a tongue surface image and a tongue and ventral surface segmentation template into the grabCut function, and the tongue and ventral surface image can be segmented by using the grabCut function. As shown in fig. 3, for example, the lingual surface image (a) and the lingual surface segmentation template (e) to be detected are input into the grabCut function to segment the lingual surface image (f).
Step S29, classifying the tongue ventral surface images by using a kmeans clustering algorithm in an opencv open source library to obtain classified result images; in this embodiment, the tongue ventral surface segmentation module 103 converts the RGB tongue ventral surface image (e.g., (f) image in fig. 4) into the Lab color space, classifies the tongue ventral surface image according to the a color channel and the b color channel of the Lab color space (the a color channel represents the range from magenta to green, and the b color channel represents the range from yellow to blue), and then uses the kmeans clustering algorithm in the opencv open source library to obtain a classification result image, for example, the classification result image is classified into 3 classes, the classification result image is (g) in fig. 4, and each color represents one class, for example, red, green, and blue.
Step S30, separating the sublingual left and right veins from the classification result image, and carrying out morphological treatment on the sublingual left and right veins to obtain a sublingual left and right vein template; in this embodiment, the sublingual vein segmentation module 104 selects two positions of the sublingual left and right veins from the image of the sublingual ventral surface, the position coordinate points are left (x, y) and right (x, y), the two position coordinate points are mapped onto the classification result image as seed points, the sublingual left and right veins are sequentially segmented from the classification result image according to a region growing algorithm, and the sublingual left and right veins are morphologically processed to obtain a sublingual left and right vein template, where (h) in fig. 4 represents the sublingual left vein template, and (i) in fig. 4 represents the sublingual right vein template.
Step S31, calculating a sublingual vein image by using the sublingual left and right vein template; in this embodiment, the sublingual vein feature extraction module 105 calculates the sublingual vein image by using the sublingual left and right vein templates, and the calculation method is as follows: and respectively performing an and operation on the sublingual left and right vein templates and each color channel of the tongue ventral surface image, and combining operation results to obtain a sublingual vein image, wherein (j) in fig. 4 represents the sublingual vein image.
Step S32, extracting sublingual vein features from the sublingual vein image; in this embodiment, the sublingual vein feature extraction module 105 extracts a total of 12 sublingual vein features from the sublingual vein image, including R, G, B color values and H, S, V color values of the sublingual veins, a length and a width of the left vein, a length and a width of the right vein, a tongue length ratio of the left vein, and a tongue length ratio of the right vein, respectively, wherein:
the calculation method for extracting the R color value of the sublingual vein by the sublingual vein feature extraction module 105 is as follows: dividing the sum of all pixel values of an R channel of the sublingual vein image by the number of sublingual vein pixels of the R channel to obtain an R color value; G. b, H, S, V color values are calculated according to the method of extracting R color values;
referring to fig. 5, the calculation method for the sublingual vein feature extraction module 105 to extract the length and width of the left vein from the sublingual vein image is as follows: calculating the minimum circumscribed rectangle of the left vein in the sublingual vein image, wherein the length and the width of the rectangle are respectively the length and the width of the left vein under the tongue, and the calculation step of the minimum circumscribed rectangle of the left vein comprises the following steps:
(1) extracting a left vein contour line and calculating a left vein convex hull according to the contour line;
(2) randomly selecting an edge AB on the left vein convex hull as an initial edge, wherein A and B are a left end point and a right end point, rotating the AB by an angle theta by taking the end point A as a center, enabling the edge AB to be parallel to an x axis of a coordinate transverse axis, and rotating all points on the left vein convex hull by the angle theta around the point A;
(3) taking the AB edge as the upper boundary or the lower boundary of the circumscribed rectangle, finding a point with the minimum y value or the maximum y value on the left vein convex hull, making a straight line parallel to the x axis through the point to determine the lower boundary or the upper boundary of the circumscribed rectangle, finding a left side point with the minimum x value and a right side point with the maximum x value on the left vein convex hull, respectively making two straight lines perpendicular to the x axis through the left side point and the right side point to determine the left boundary and the right boundary of the circumscribed rectangle, thus obtaining the circumscribed rectangle, and calculating and storing the endpoint coordinates of the AB edge, the length, the width and the area of the circumscribed rectangle;
(4) sequentially selecting the next edge BC on the left vein convex hull, and repeating the steps (2) to (3) to search for the next external rectangle until all edges on the left vein convex hull are traversed;
(5) comparing the areas of all the external rectangles, and finding out the external rectangle with the smallest area as the smallest external rectangle of the left vein;
the calculation method for extracting the length and width of the right vein from the sublingual vein image by the sublingual vein feature extraction module 105 is the same as the calculation method for extracting the length and width of the left vein.
The calculation method for extracting the left vein-tongue length ratio from the sublingual vein image by the sublingual vein feature extraction module 105 is as follows: the ratio of the length of the tongue divided by the length of the left sublingual vein was recorded as the ratio of the length of the left sublingual vein.
The calculation method for extracting the length ratio of the right vein to the tongue from the sublingual vein image by the sublingual vein feature extraction module 105 is as follows: the ratio of the input tongue length divided by the length of the sublingual right vein was recorded as the tongue length ratio of the right vein.
As a preferred embodiment, the inferior vein feature extraction module 105 outputs the extracted 12 sublingual vein features to an externally connected display through the output unit 14 for display or printing on a printer, so that the traditional Chinese doctor can provide reference for the sublingual vein pulse diagnosis of the traditional Chinese medicine, thereby assisting the traditional Chinese doctor in the accuracy of the judgment result of the sublingual vein pulse diagnosis of the traditional Chinese medicine.
The present invention also provides a computer readable storage medium storing a plurality of computer program instructions for being loaded by a processor of a computer apparatus and for performing the method steps of the sublingual vein feature extraction method according to the present invention. Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments may be implemented by related program instructions, and the program may be stored in a computer-readable storage medium, which may include: read-only memory, random access memory, magnetic or optical disk, and the like.
According to the sublingual vein feature extraction device and method, the ventral-lingual surface detector is obtained through training of a large number of lingual surface sample images to effectively detect the ventral-lingual surface containing lips, so that the accuracy of the ventral-lingual surface segmentation is improved; the sublingual left and right veins are effectively segmented by carrying out color classification treatment on the tongue ventral surface image to create a sublingual left and right vein template, so that the sublingual left and right veins segmented in the vein segmentation aspect are more complete, and the accuracy of sublingual vein feature extraction is improved.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. A sublingual vein feature extraction device comprising a processor adapted to implement various computer program instructions and a memory adapted to store a plurality of computer program instructions, characterized in that said computer program instructions are loaded by the processor and execute the steps of:
inputting different tongue surface sample images through an input unit to construct a plurality of positive and negative samples;
processing a plurality of positive and negative samples by using an opencv _ createsamples program in an opencv open source library to generate a training data set;
training the training data set by using an opencv _ traincacade program in an opencv open source library to generate a tongue ventral surface detector;
detecting the ventral and lingual surfaces containing lips from the lingual surface image to be detected by using a ventral and lingual surface detector;
determining the position information of the lingual surface and the ventral surface based on the lingual surface image to be detected, and intercepting the lingual surface and the ventral surface containing lips according to the position information of the lingual surface and the ventral surface;
performing threshold segmentation processing on the intercepted tongue ventral surface to obtain a shadow area and a tooth area of the tongue ventral surface;
creating a tongue ventral surface segmentation template by using a shadow area and a tooth area of the tongue ventral surface;
inputting a tongue surface image to be detected and a tongue ventral surface segmentation template into a grabCut function in an opencv open source library to segment the tongue ventral surface image;
classifying the tongue ventral surface images by using a kmeans clustering algorithm in an opencv open source library to obtain classified result images;
separating the left and right sublingual veins from the classification result image, and performing morphological treatment on the left and right sublingual veins to obtain left and right sublingual vein templates;
calculating a sublingual vein image by using the sublingual left and right vein template, and extracting sublingual vein features from the sublingual vein image.
2. The sublingual vein feature extraction device according to claim 1, wherein the step of creating a lingual ventral segmentation template using the dark shadow area and the tooth area of the lingual ventral surface comprises:
extracting a contour line of a tongue ventral surface shadow area by using a canny edge detection algorithm, extracting all n coordinate points of the contour line, and pairing all the n coordinate points to form n (n-1)/2 edges;
for each edge, checking whether the remaining (n-2) points are on the same side of the edge;
if all the points are on one side of the edge, adding the edge into the convex hull set until all the edges are traversed, and taking the convex hull set as the contour line of the tongue ventral surface shadow area;
creating a single-channel template image with the size of the tongue surface sample image, and mapping the contour line of the tongue ventral surface shadow area to the corresponding position of the single-channel template image;
setting all pixel values of the non-white area inside the contour line as 1, setting all pixel values of the white area inside the contour line as 3, setting all pixel values of the area outside the contour line as 0, and setting the pixel value of the corresponding position of the single-channel template image as 0 according to the lingual tooth area to obtain the lingual segmentation template.
3. The sublingual vein feature extraction device according to claim 1, wherein the step of classifying the tongue ventral surface image by using a kmeans clustering algorithm in an opencv open source library to obtain a classification result image comprises:
converting the RGB tongue ventral surface image into a Lab color space, and classifying the tongue ventral surface image by using a kmeans clustering algorithm in an opencv open source library according to the pixel values of a color channel a and a color channel b of the tongue ventral surface image in the Lab color space to obtain a classification result image.
4. The sublingual vein feature extraction device according to claim 1, wherein the step of calculating the sublingual vein image using the sublingual left and right vein template comprises:
and respectively performing AND operation on the sublingual left and right vein templates and each color channel of the tongue ventral surface image, and combining operation results to obtain the sublingual vein image.
5. The sublingual vein feature extraction device of claim 1, wherein the sublingual vein features comprise R, G, B color values and H, S, V color values of the sublingual vein, a length and a width of the left vein, a length and a width of the right vein, a tongue length ratio of the left vein, and a tongue length ratio of the right vein, wherein:
the calculation method for extracting the R color value of the sublingual vein comprises the following steps: dividing the sum of all pixel values of an R channel of the sublingual vein image by the number of sublingual vein pixels of the R channel to obtain an R color value; G. b, H, S, V color values are calculated according to the method of extracting R color values;
the calculation method for extracting the length and width of the left vein comprises the following steps: calculating the minimum circumscribed rectangle of the left vein in the sublingual vein image, wherein the length and the width of the rectangle are respectively the length and the width of the left vein under the tongue, and the calculation step of the minimum circumscribed rectangle of the left vein comprises the following steps:
(1) extracting a left vein contour line and calculating a left vein convex hull according to the contour line;
(2) randomly selecting an edge AB on the left vein convex hull as an initial edge, wherein A and B are a left end point and a right end point, rotating the AB by an angle theta by taking the end point A as a center, and enabling the edge AB to be parallel to an x axis of a coordinate transverse axis, so that all points of the left vein convex hull rotate by the angle theta around the point A;
(3) taking the AB edge as the upper boundary or the lower boundary of the circumscribed rectangle, finding a point with the minimum y value or the maximum y value on the left vein convex hull, making a straight line parallel to the x axis through the point to determine the lower boundary or the upper boundary of the circumscribed rectangle, finding a left side point with the minimum x value and a right side point with the maximum x value on the left vein convex hull, respectively making two straight lines perpendicular to the x axis through the left side point and the right side point to determine the left boundary and the right boundary of the circumscribed rectangle, thus obtaining the circumscribed rectangle, and calculating and storing the endpoint coordinates of the AB edge, the length, the width and the area of the circumscribed rectangle;
(4) sequentially selecting the next edge BC on the left vein convex hull, and repeating the steps (2) to (3) to search for the next external rectangle until all edges on the left vein convex hull are traversed;
(5) comparing the areas of all the external rectangles, and finding out the external rectangle with the smallest area as the smallest external rectangle of the left vein;
the calculation method for extracting the length and width of the right vein is the same as the calculation method for extracting the length and width of the left vein;
the calculation method for extracting the left vein-tongue length ratio comprises the following steps: the ratio of the input tongue length divided by the length of the sublingual left vein is recorded as the tongue length ratio of the left vein;
the calculation method for extracting the length ratio of the right vein tongue comprises the following steps: the ratio of the input tongue length divided by the length of the sublingual right vein was recorded as the tongue length ratio of the right vein.
6. A sublingual vein feature extraction method is characterized by comprising the following steps:
inputting different tongue surface sample images through an input unit to construct a plurality of positive and negative samples;
processing a plurality of positive and negative samples by using an opencv _ createsamples program in an opencv open source library to generate a training data set;
training the training data set by using an opencv _ traincacade program in an opencv open source library to generate a tongue ventral surface detector;
detecting the ventral and lingual surfaces containing lips from the lingual surface image to be detected by using a ventral and lingual surface detector;
determining the position information of the lingual surface and the ventral surface based on the lingual surface image to be detected, and intercepting the lingual surface and the ventral surface containing lips according to the position information of the lingual surface and the ventral surface;
performing threshold segmentation processing on the intercepted tongue ventral surface to obtain a shadow area and a tooth area of the tongue ventral surface;
creating a tongue ventral surface segmentation template by using a shadow area and a tooth area of the tongue ventral surface;
inputting a tongue surface image to be detected and a tongue ventral surface segmentation template into a grabCut function in an opencv open source library to segment the tongue ventral surface image;
classifying the tongue ventral surface images by using a kmeans clustering algorithm in an opencv open source library to obtain classified result images;
separating the left and right sublingual veins from the classification result image, and performing morphological treatment on the left and right sublingual veins to obtain left and right sublingual vein templates;
calculating a sublingual vein image by using the sublingual left and right vein template, and extracting sublingual vein features from the sublingual vein image.
7. The sublingual vein feature extraction method of claim 6, wherein the step of creating a lingual ventral segmentation template using the dark shadow area and the tooth area of the lingual ventral surface comprises:
extracting a contour line of a tongue ventral surface shadow area by using a canny edge detection algorithm, extracting all n coordinate points of the contour line, and pairing all the n coordinate points to form n (n-1)/2 edges;
for each edge, checking whether the remaining (n-2) points are on the same side of the edge;
if all the points are on one side of the edge, adding the edge into the convex hull set until all the edges are traversed, and taking the convex hull set as the contour line of the tongue ventral surface shadow area;
creating a single-channel template image with the size of the tongue surface sample image, and mapping the contour line of the tongue ventral surface shadow area to the corresponding position of the single-channel template image;
setting all pixel values of the non-white area inside the contour line as 1, setting all pixel values of the white area inside the contour line as 3, setting all pixel values of the area outside the contour line as 0, and setting the pixel value of the corresponding position of the single-channel template image as 0 according to the lingual tooth area to obtain the lingual segmentation template.
8. The sublingual vein feature extraction method of claim 6, wherein the step of classifying the tongue ventral surface image by using a kmeans clustering algorithm in an opencv open source library to obtain a classification result image comprises the following steps:
converting the RGB tongue ventral surface image into a Lab color space, and classifying the tongue ventral surface image by using a kmeans clustering algorithm in an opencv open source library according to the pixel values of a color channel a and a color channel b of the tongue ventral surface image in the Lab color space to obtain a classification result image.
9. The sublingual vein feature extraction method of claim 6, wherein the step of calculating the sublingual vein image using the sublingual left and right vein template comprises:
and respectively performing AND operation on the sublingual left and right vein templates and each color channel of the tongue ventral surface image, and combining operation results to obtain the sublingual vein image.
10. The sublingual vein feature extraction method of claim 6, wherein the sublingual vein features comprise R, G, B color values and H, S, V color values of the sublingual vein, a length and a width of the left vein, a length and a width of the right vein, a tongue length ratio of the left vein, and a tongue length ratio of the right vein, wherein:
the calculation method for extracting the R color value of the sublingual vein comprises the following steps: dividing the sum of all pixel values of an R channel of the sublingual vein image by the number of sublingual vein pixels of the R channel to obtain an R color value; G. b, H, S, V color values are calculated according to the method of extracting R color values;
the calculation method for extracting the length and width of the left vein comprises the following steps: calculating the minimum circumscribed rectangle of the left vein in the sublingual vein image, wherein the length and the width of the rectangle are respectively the length and the width of the left vein under the tongue, and the calculation step of the minimum circumscribed rectangle of the left vein comprises the following steps:
(1) extracting a left vein contour line and calculating a left vein convex hull according to the contour line;
(2) randomly selecting an edge AB on the left vein convex hull as an initial edge, wherein A and B are a left end point and a right end point, rotating the AB by an angle theta by taking the end point A as a center, and enabling the edge AB to be parallel to an x axis of a coordinate transverse axis, so that all points of the left vein convex hull rotate by the angle theta around the point A;
(3) taking the AB edge as the upper boundary or the lower boundary of the circumscribed rectangle, finding a point with the minimum y value or the maximum y value on the left vein convex hull, making a straight line parallel to the x axis through the point to determine the lower boundary or the upper boundary of the circumscribed rectangle, finding a left side point with the minimum x value and a right side point with the maximum x value on the left vein convex hull, respectively making two straight lines perpendicular to the x axis through the left side point and the right side point to determine the left boundary and the right boundary of the circumscribed rectangle, thus obtaining the circumscribed rectangle, and calculating and storing the endpoint coordinates of the AB edge, the length, the width and the area of the circumscribed rectangle;
(4) sequentially selecting the next edge BC on the left vein convex hull, and repeating the steps (2) to (3) to search for the next external rectangle until all edges on the left vein convex hull are traversed;
(5) comparing the areas of all the external rectangles, and finding out the external rectangle with the smallest area as the smallest external rectangle of the left vein;
the calculation method for extracting the length and width of the right vein is the same as the calculation method for extracting the length and width of the left vein;
the calculation method for extracting the left vein-tongue length ratio comprises the following steps: the ratio of the input tongue length divided by the length of the sublingual left vein is recorded as the tongue length ratio of the left vein;
the calculation method for extracting the length ratio of the right vein tongue comprises the following steps: the ratio of the input tongue length divided by the length of the sublingual right vein was recorded as the tongue length ratio of the right vein.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811419050.8A CN111222371A (en) | 2018-11-26 | 2018-11-26 | Sublingual vein feature extraction device and method |
| PCT/CN2019/120645 WO2020108437A1 (en) | 2018-11-26 | 2019-11-25 | Sublingual vein feature extraction apparatus and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811419050.8A CN111222371A (en) | 2018-11-26 | 2018-11-26 | Sublingual vein feature extraction device and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111222371A true CN111222371A (en) | 2020-06-02 |
Family
ID=70825673
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811419050.8A Pending CN111222371A (en) | 2018-11-26 | 2018-11-26 | Sublingual vein feature extraction device and method |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN111222371A (en) |
| WO (1) | WO2020108437A1 (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112329587A (en) * | 2020-10-30 | 2021-02-05 | 苏州中科先进技术研究院有限公司 | Beverage bottle classification method and device and electronic equipment |
| CN113409304A (en) * | 2021-07-15 | 2021-09-17 | 深圳市圆道妙医科技有限公司 | Holographic-based multidimensional tongue image analysis method, system, equipment and storage medium |
| CN112329587B (en) * | 2020-10-30 | 2024-05-24 | 苏州中科先进技术研究院有限公司 | Beverage bottle classification method and device and electronic equipment |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112163452B (en) * | 2020-08-25 | 2022-11-18 | 同济大学 | Three-dimensional reconstruction method of binocular near-infrared limb vein image based on deep learning |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101576913B (en) * | 2009-06-12 | 2011-09-21 | 中国科学技术大学 | Automatic clustering, visual and retrieval system for tongue picture based on self-organizing map neural network |
| CN104298983B (en) * | 2013-07-15 | 2018-09-11 | 清华大学 | Tongue fur image with distributed user terminal obtains and analysis system |
| CN104537379A (en) * | 2014-12-26 | 2015-04-22 | 上海大学 | High-precision automatic tongue partition method |
-
2018
- 2018-11-26 CN CN201811419050.8A patent/CN111222371A/en active Pending
-
2019
- 2019-11-25 WO PCT/CN2019/120645 patent/WO2020108437A1/en active Application Filing
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112329587A (en) * | 2020-10-30 | 2021-02-05 | 苏州中科先进技术研究院有限公司 | Beverage bottle classification method and device and electronic equipment |
| CN112329587B (en) * | 2020-10-30 | 2024-05-24 | 苏州中科先进技术研究院有限公司 | Beverage bottle classification method and device and electronic equipment |
| CN113409304A (en) * | 2021-07-15 | 2021-09-17 | 深圳市圆道妙医科技有限公司 | Holographic-based multidimensional tongue image analysis method, system, equipment and storage medium |
| CN113409304B (en) * | 2021-07-15 | 2022-05-20 | 深圳市圆道妙医科技有限公司 | Holographic-based multidimensional tongue image analysis method, system, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2020108437A1 (en) | 2020-06-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108615239B (en) | Tongue image segmentation method based on threshold technology and gray level projection | |
| EP2847738B1 (en) | Method and apparatus for image scoring and analysis | |
| CN110866932A (en) | Multi-channel tongue edge detection device and method and storage medium | |
| WO2022001571A1 (en) | Computing method based on super-pixel image similarity | |
| US9489561B2 (en) | Method and system for estimating fingerprint pose | |
| US9852354B2 (en) | Method and apparatus for image scoring and analysis | |
| Mirzaalian et al. | Automatic globally-optimal pictorial structures with random decision forest based likelihoods for cephalometric x-ray landmark detection | |
| Yang et al. | Real-time traffic sign detection via color probability model and integral channel features | |
| Han et al. | Robust anatomical landmark detection with application to MR brain image registration | |
| CN104766316A (en) | Novel lip segmentation algorithm for traditional Chinese medical inspection diagnosis | |
| CN111223117A (en) | Tongue surface image segmentation device, method and computer storage medium | |
| CN111222371A (en) | Sublingual vein feature extraction device and method | |
| KR20130067658A (en) | Apparatus and method of diagnosing health by separating region of tongue coating from region of tongue substance | |
| Rehman et al. | Human identification using dental biometric analysis | |
| CN111340052A (en) | Tongue tip red detection device and method for tongue diagnosis in traditional Chinese medicine and computer storage medium | |
| CN104933723A (en) | Tongue image segmentation method based on sparse representation | |
| CN111311602A (en) | Lip image segmentation device and method for traditional Chinese medicine facial diagnosis | |
| CN110648336A (en) | Method and device for dividing tongue texture and tongue coating | |
| CN112258536B (en) | Integrated positioning and segmentation method for calluses and cerebellum earthworm parts | |
| CN106372593B (en) | Optic disk area positioning method based on vascular convergence | |
| US20190333212A1 (en) | Visual cardiomyocyte analysis | |
| CN108629780B (en) | Tongue image segmentation method based on color decomposition and threshold technology | |
| CN106780470A (en) | CT image nipple automated detection methods | |
| CN113989269B (en) | Traditional Chinese medicine tongue image tooth trace automatic detection method based on convolutional neural network multi-scale feature fusion | |
| Gu et al. | Automatic Tongue Image Segmentation Based on Thresholding and an Improved Level Set Model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200602 |