CN106725341A - A kind of enhanced lingual diagnosis system - Google Patents
A kind of enhanced lingual diagnosis system Download PDFInfo
- Publication number
- CN106725341A CN106725341A CN201710012778.8A CN201710012778A CN106725341A CN 106725341 A CN106725341 A CN 106725341A CN 201710012778 A CN201710012778 A CN 201710012778A CN 106725341 A CN106725341 A CN 106725341A
- Authority
- CN
- China
- Prior art keywords
- module
- face
- tongue
- illness
- enhanced
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0059—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence
- A61B5/0062—Arrangements for scanning
- A61B5/0064—Body surface scanning
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0033—Features or image-related aspects of imaging apparatus classified in A61B5/00, e.g. for MRI, optical tomography or impedance tomography apparatus; arrangements of imaging apparatus in a room
- A61B5/004—Features or image-related aspects of imaging apparatus classified in A61B5/00, e.g. for MRI, optical tomography or impedance tomography apparatus; arrangements of imaging apparatus in a room adapted for image acquisition of a particular organ or body part
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0059—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence
- A61B5/0077—Devices for viewing the surface of the body, e.g. camera, magnifying lens
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4854—Diagnosis based on concepts of traditional oriental medicine
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Pathology (AREA)
- Veterinary Medicine (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Public Health (AREA)
- Physics & Mathematics (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Alternative & Traditional Medicine (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
- Image Processing (AREA)
Abstract
The present invention provides a kind of enhanced lingual diagnosis system, including source illumination module, image capture module, host module, the source illumination module is area source, the facial zone of sufficient light patient, for described image acquisition module provides illumination, described image acquisition module includes high accuracy camera and ultra-low distortion camera lens, for carrying out IMAQ to patient, the host module includes facial recognition modules and illness grader, the facial recognition modules are communicated to connect with described image acquisition module, the image information transmitted by described image acquisition module can be received, and it is identified feedback, the illness grader communicates with the facial recognition modules, diagnosed with receiving the image information after identification.The present invention also provides the method that the tongue of the enhanced lingual diagnosis system carries out lingual diagnosis as acquisition methods and using the enhancing lingual diagnosis system.
Description
Technical field
The present invention relates to a kind of medical diagnosis technical field, in particular to a kind of enhanced lingual diagnosis system, using figure
As treatment, recognition of face and neural network algorithm carry out auxiliary diagnosis by face and tongue to patient.The invention further relates to upper
The face recognition of enhanced lingual diagnosis system and the method for facial feature localization are stated, the tongue of above-mentioned enhanced lingual diagnosis system is as acquisition methods.
Method the invention further relates to carry out lingual diagnosis using the enhanced lingual diagnosis system.
Background technology
The face and coating colour (tongue) for observing patient are the important objective index of tcm diagnosis disease.Face be divided into mass-tone,
Three kinds of objective color and sickly complexion, different face represent different health status, the face of people is exactly one side it can be seen that physiological health,
The mirror of psychological condition;Lingual diagnosis is the important method of tcm diagnosis disease, and the change for observing the color and luster, form of tongue is that auxiliary is examined
A simple effective method that is disconnected and differentiating, is one of observation key content, and tongue is the seedling of the heart, is waited outside spleen, and tongue is by stomach Qi
Give birth to.Internal organs are associated by passages through which vital energy circulates with tongue, the collateral of heart meridian system tongue sheet, and the arteries and veins of the few the moon of foot holds tongue sheet under the arm, the train of thought tongue of the moon of fainting enough
This, the lunar arteries and veins of foot connects tongue sheet, dissipates sublingual, thus internal organs lesion, can be reflected in tongue nature and tongue fur, tongue is mainly examined in lingual diagnosis
The form of matter and tongue fur, color and luster, moisturize, the property that judges disease with this, the shallow depth of patient's condition, the prosperity and decline of qi and blood, body fluid are full of
Thanks to and internal organs actual situation etc..
Traditional observation mode is typically doctor and detects by an unaided eye the feature such as face and tongue form and color and luster of patient, so
Diagnosed according to career in medicine experience afterwards.However, because the personal experience of doctor is different so that the subjectivity by doctor of lingual diagnosis
Property influence it is larger.Also, wait doctor to see that it is a cumbersome and time-consuming thing to examine, and consumes substantial amounts of energy.
In recent years, some lingual diagnosis systems are had been disclosed for, there is the lingual diagnosis system image capture module to carry out figure to tongue
As collection, then doctor can be diagnosed according to the image of collection, however, by the image information of this system acquisition often
Shoot not accurate enough, and there are aberration so that diagnosis also occurs deviation.
Accordingly, it is desired to provide one kind can like clockwork gather face and tongue as information, and can be to the letter of collection
The system that breath carries out automated diagnostic, to improve diagnostic accuracy and speed, reduces the stand-by period for spending.
The content of the invention
It is an object of the invention to provide one kind automation, intelligentized lingual diagnosis system, it can be carried out to face and tongue
Positioning is accurately identified, and the information of collection can be based on and automatically diagnosed.
The present invention is achieved through the following technical solutions above-mentioned purpose:A kind of enhanced lingual diagnosis system, including light source lighting
Module, image capture module, host module;The source illumination module is area source, and the facial zone of sufficient light patient is
Described image acquisition module provides illumination, and described image acquisition module includes camera and ultra-low distortion camera lens, for entering to patient
Row IMAQ, the host module includes facial recognition modules and illness grader, the facial recognition modules and the figure
As acquisition module communication connection, the image information transmitted by described image acquisition module can be received, and be identified feedback,
The illness grader communicates with the facial recognition modules, is diagnosed with receiving the image information after identification.
Preferably, the source illumination module is LED array, and with adjustable brightness of illumination and angle.
Preferably, described image acquisition module is WIFI industrial cameras, with USB storage, signal projector and signal
Receiver.
Preferably, the host module is also to include display module, and the display module communicates with the illness grader,
For showing diagnostic result.
According to another aspect of the present invention, the present invention also provides the tongue of enhanced lingual diagnosis system as acquisition methods, including:
1) adjustment source illumination module, image capture module make described image acquisition module clear relative to the position of face
The facial characteristics of face is gathered clearly;
2) information transfer for gathering described image acquisition module carries out facial knowledge to the facial recognition modules of host module
Not, if face recognition fails, return to step 1), if face recognition success, carries out next step;
3) measured human face ratio, carries out facial feature localization, if facial feature localization fails, return to step 1) and, if face are fixed
Position success, carries out next step;
4) face information is obtained, while limiting lip-region, points out to stretch out tongue, carry out tongue image;
5) tongue image is obtained, and the disposal of gentle filter is carried out to the tongue image.
According to another aspect of the present invention, the present invention also provides one kind and carries out lingual diagnosis using the enhanced lingual diagnosis system
Method, including:
By the tongue of the enhanced lingual diagnosis system as acquisition methods acquisition face and tongue are as information;
By the face and tongue as the illness grader of information input to the host module;
The illness grader, as information carries out computing, show that illness is classified to the face and tongue.
According to a preferred embodiment of the invention, also including illness classification results are included on display module.
According to a preferred embodiment of the invention, the illness grader is the diagnosis mould based on BP neural network
Type, it is further comprising the steps of:
By the face and tongue as information input to the input layer of the BP neural network;
Train the BP neural network diagnostic model;
Adjust the model parameter of the BP neural network;
Obtain optimal BP neural network diagnostic model.
The beneficial effects of the invention are as follows:It is of the invention that face is accurately recognized and positioned using image capture module, from
And face and tongue are accurately obtained as information, and above- mentioned information is input into illness grader, face and tongue are merged as information is carried out
Diagnosis so that diagnostic result is more accurate and reliable.
Brief description of the drawings
Fig. 1 is the structural representation of the enhanced lingual diagnosis system according to one embodiment of the invention;
Fig. 2 is the program flow diagram of enhanced lingual diagnosis system of the invention;
Fig. 3 is enhanced lingual diagnosis system face recognition flow chart of the invention;
Fig. 4 is the illness grader schematic diagram of enhanced lingual diagnosis system of the invention;
In figure:1-source illumination module;2-image capture module;3-host module.
Specific embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation is described, it is clear that described embodiment is only a part of embodiment of the invention, rather than whole embodiments.It is based on
Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under the premise of creative work is not made
Embodiment, belongs to the scope of protection of the invention.
Fig. 1 shows the enhanced lingual diagnosis system of an embodiment of the invention, including source illumination module 1,
Image capture module 2, host module 3.
The source illumination module 1 is annular light source, it is therefore preferable to LED array light source, and with adjustable brightness of illumination
And angle, the facial zone of sufficient light patient, it is that described image acquisition module 2 provides illumination.Source illumination module is preferably
Using high density LED array, high brightness and compact design are conducive to saving installing space, and source illumination module can basis
The brightness of illumination and angle of actual photographed environment adjustment special light sources, using the teaching of the invention it is possible to provide different irradiating angles, solve irradiation shade and ask
Topic, there is provided good, stabilization illumination condition, to reach optimal shooting effect.The source illumination module by interesting part and
The grey value difference of other parts is increased, and blanking as far as possible is lost interest in part, improves signal to noise ratio, it is ensured that stably obtain high-quality
Image.
Described image acquisition module 2 includes camera and ultra-low distortion camera lens, for carrying out IMAQ to patient.The figure
As acquisition module includes high accuracy camera, equipped with high-resolution, acutance high, short focus, large aperture camera lens, by adjusting mirror
Head focal length length and aperture size, makes camera obtain the clear, facial image of high-resolution, ultra-low distortion, and by the face figure
As being transferred to host module.Described image acquisition module 2 is preferably WIFI industrial cameras, launches with USB storage, signal
Device and signal receiver.Above-mentioned WIFI industrial cameras are used and can carried out the image information of collection with USB store functions
Storage, signal projector and signal receiver are realized using the WIFI of high stability, high-transmission ability and high anti-jamming capacity
Image information between industrial camera and host module is interacted.It is to be appreciated that can select wired or wireless according to site environment
Transmission means completes the interaction of information.
The host module 3 includes facial recognition modules and illness grader, the facial recognition modules and described image
Acquisition module is communicated to connect, and can receive the image information transmitted by described image acquisition module, and be identified feedback, institute
State illness grader to be communicated with the facial recognition modules, diagnosed with receiving the image information after identification.Wherein, the face
Portion's identification module includes that face recognition, positioning face and tongue image are obtained, and the facial recognition modules are received by IMAQ
The image information that module 2 sends, then facial recognition modules face recognition is carried out first to the image information for receiving, if facial
Recognize successfully, then position face, obtain face information, then obtain tongue image information;On the other hand, if face recognition is lost
Lose, feedback of the information to image capture module is re-started IMAQ by facial recognition modules.The work of the facial recognition modules
Be will be described below as flow.
Face information that the facial recognition modules will be obtained and tongue as information input to illness grader, through illness classification
Device exports diagnosis report after calculating.The algorithm of the illness grader will be described below.
Preferably, the host module 3 is also to include display module, and the display module is logical with the illness grader
Letter, for showing diagnostic result.The display module can be a display interface with programming, make doctor's Real Time Observation
It is that doctor uses system and the sick interaction platform for examining result of final display to diagnosing patient situation.
Fig. 2 shows the program flow diagram of enhanced lingual diagnosis system of the invention.System platform is built first, and light source is shone
Bright module, image capture module and patient standing place, host module are adjusted to optimum state;Camera and aperture are opened, it is main
Machine module gets the facial image of collection by wired or wireless way;Host module facial image is carried out face recognition and
Facial feature localization, reacquisition facial image is differentiated if failing;The face information and tongue that will be obtained are as information transfer
Give illness grader, the illness Multiple Classifier Fusion tongue carries out computing with face information, exports diagnosis report.It should be understood that will
It is a kind of preferred embodiment that face information and tongue all input to illness grader as both information, it is also possible to only by face and
Tongue carries out diagnosis computing as one of information inputs to illness grader.
Facial recognition modules are included in the host module of enhanced lingual diagnosis system.Fig. 3 shows enhancing of the invention
The face recognition flow chart of type lingual diagnosis system.The method is positioned based on Haar features, and Haar features are generally broadly divided
It is three classes:Point feature, line feature and central feature, essence are exactly the feature that a kind of rectangle by many stripeds is constituted.
First train face grader, can realize to face and it is non-face make a distinction, method is as follows:
Being compared based on several essential characteristics and different characteristic values can obtain Weak Classifier, in conjunction with certain condition
Select Weak Classifier that is wherein preferable, being best suitable for real work.
Weak Classifier mathematic(al) structure is as follows:
Wherein, h (x, f, p, θ) represents most initial Weak Classifier, and x is determined that f is haar features by subwindow image
Value, θ values are threshold value, and the effect of p is control sign of inequality direction.
Compare feature in the characteristic value and Weak Classifier of input picture, judge to be less than the threshold value when the characteristic value of input picture
When just determine that it is face.The process for training optimal Weak Classifier is actually to find suitable grader threshold value, making this
Grader is minimum to the parallax error of all samples.
Train the Weak Classifier for obtaining to get together once to be compared, then to according to current result of the comparison and pressing
The respective False Rate of Weak Classifier compared according to these participations is weighted, finally by the result after weighted sum and average ratio
The gap of relatively result is compared, and preferable strong classifier just can be obtained after so repeatedly being compared.Needs train multiple
The strong classifier of ad eundem accuracy rate, then allow these strong classifiers in layer arrange joint according to by letter to numerous order
Together, the cascade classifier that accuracy rate is high and efficiency is splendid, i.e. face point could finally be set up after carrying out such operation
Class device.
As shown in figure 3, several strong classifiers are arranged from simple to complex, multiple strong classifiers to facial image successively
Judge, when each layer of grader all differentiates successfully, be just determined as face.The method drops while improving strong classifier verification and measurement ratio
Low misclassification rate, detection time is significantly reduced while high detection rate is ensured with low misclassification rate again.Make each strong by training
Grader has verification and measurement ratio and relatively low misclassification rate higher, it is assumed that cascade classifier is made up of 20 strong classifiers, by instruction
The verification and measurement ratio of each strong classifier is 99% after white silk, and misclassification rate is all 50%, then the verification and measurement ratio of this cascade classifier is
0.9920≈ 98%, misclassification rate is respectively:0.520≈ 0.0001%.As can be seen here, the cascade classifier foot for being obtained by training
To meet actually detected demand.Facial feature localization is carried out if recognition of face success, to IMAQ if recognition of face failure
Information is fed back, and resurveys image.
After face classification device identifies face, face is regarded as into ROI region and facial feature localization is carried out.Based on showing for face
There is knowledge, face characteristic and their correlation are described using certain rule.For the face of standard proportional, from hairline
To open wiring, from open wiring to nose bottom line, from nose bottom line to neck bottom line, this three part is equal to line, and we typically pass through geisoma (eyebrow
Line) make a horizontal line, a horizontal line is remake by nose lower edge (nose bottom line), if the distance between two parallel lines is W, also
Eye center point in vertical direction is considered as W to the distance of face central point.From in terms of front, face place most wide is five eyes
Width, if width for eye is L, take two lines of central point, its distance in the horizontal direction is 2L.This is people
Face is long with face general standard ratio wide, is outlined according to this ratio Distribution and localization face and with appropriately sized rectangle.Recognizing
Go out on the basis of face, be distributed by this ratio, with the face leftmost side and the intersection point of hair line, the i.e. upper left of standard proportional face
Angle can describe to orient face position as the origin of coordinates.If facial feature localization fails, fed back to image capture module, weight
Newly carry out IMAQ;If facial feature localization success, obtains face information, and suggestion device by voice device etc. is carried
Show that patient carries out tongue image acquisition.Face position is oriented in facial feature localization, because the area of lip is maximum, at corrosion
Lip region will not be eliminated in binary image after reason, be a base with two corners of the mouth lines so as to orient Hp position
Standard, the distance of vertical drop-down W does rectangle, retains image in this rectangle, i.e. tongue picture part.Using detection tongue picture part of differentiating
The place that gray level or structure have mutation obtains tongue edge, because edge and noise are all gray scale discontinuity poinies, in frequency
Domain is high fdrequency component, is directly difficult to the influence for overcoming noise using differentiating, therefore detects right before edge with differential operator
Image carries out smothing filtering, finally gives pure tongue as region.
Illness grader is further comprises in the host module of enhanced lingual diagnosis system.Fig. 4 is enhanced tongue of the invention
Examine the illness grader schematic diagram of system.As shown in figure 4, in the present embodiment, the illness grader is based on BP nerves
The diagnostic model of network, by face and tongue as information input illness grader, is diagnosed, output illness classification.Using illness
When grader is diagnosed, first have to by being input into face and tongue as information is trained to illness grader.Below to being based on
The structure principle of the illness grader of BP neural network is described.BP neural network is also referred to as error backward propagation method,
The Multi-layered Feedforward Networks that it is made up of non-linear conversion unit, are typically made up of input layer, output layer and the part of hidden layer three,
By several neurons by certain contact come work.Wherein hidden layer can be one or more layers, in the present embodiment
In, from the point of view of simple and practical, from one layer of hidden layer, this is enough to complete arbitrary n dimensions to the mapping of m dimensions.Input
The nodes of layer can select according to the number of input quantity, in the present embodiment, input layer be face and tongue as information, therefore
Input layer number 2.The selection of node in hidden layer purpose is a relatively complicated problem, and it is more with input-output unit
Rare direct relation.The performance of network may all be influenceed very little too much.Can cause learning time long too much, and then make very little
Fault freedom into network is poor, for the Some features of hidden layer structure, on the basis of abundant experimental results, provides a warp
Formula is tested as reference:
Wherein, NhiddenIt is node in hidden layer, NIn, Nout, NclassThe respectively input layer number of neutral net, defeated
Go out the target classification number of node layer number and required point.Max is the function of maximizing.In the present embodiment, target classification number
It is disorder class number.
The transfer function of neuron selects nonlinear function, and commonly used is Logsig types and Tansig
Type, is expressed as follows respectively:
The model of face and tongue diagnostic method, i.e. illness grader are constructed with above-mentioned principle, simple analog people's
Brain cranial nerve function realizes the mapping model of input-output, then by training illness grader, realize by face and
Tongue is diagnosed.
In BP neural network, the output valve computing formula of neuron is as follows:
Wherein, xiIt is the input of neuron, wiIt is the connection weight between neuron, b is the threshold value of neuron, Huo Chengwei
Biasing, f is the transfer function of neuron or is excitation function, and O is the output of neuron.
In the present embodiment, input layer is the entrance that face and tongue enter illness grader as characteristic information, through implicit
After layer each neuron treatment, it is transmitted to output layer and obtains the classification of each illness.Row information is entered to the total data for storing first
Forward-propagating, in information forward-propagating, sick database data is incoming from input layer, after being processed through hidden layer each neuron, is transmitted to
Output layer obtains each illness grader.If the reality output disorder class number of output layer has error and this mistake with desired output
Difference more than target error when, neutral net enter the error back propagation stage, in this stage, the operation being substantially carried out be amendment
Each neuron weights, four parameters are corrected according to error gradient descent method successively:The weighed value adjusting amount △ w of output layerki, output
The adjusting thresholds amount △ a of layerk, the weighed value adjusting amount △ w of hidden layerij, the adjusting thresholds amount △ θ of hidden layeri, adjustment formula is such as
Under: Wherein, η
It is Learning Step parameter, E is overall error.
System is to the total error criteria function of n training sample:
Wherein, TkRepresent the desired output of k-th neuron of output layer, OkRepresent that the reality of k-th neuron of output layer is defeated
Go out.
Forward-propagating and the two processes of the backpropagation of error of information are repeated, until error is less than target error
When, preferable illness grader is obtained, while illness classifier training is completed, next can just utilize various faces and tongue picture
Information diagnose.
The invention provides a kind of enhanced lingual diagnosis system, enhanced lingual diagnosis system tongue is as acquisition methods and utilizes tongue
Method as carrying out lingual diagnosis so that the diagnosis based on tongue picture is more convenient and reliable.
Preferably, enhanced lingual diagnosis system of the invention also includes a big data platform so that according to face and tongue picture
After the classification of information acquisition illness, a therapeutic scheme can be automatically obtained according to the big data platform.
After enhanced lingual diagnosis system diagnostics goes out tongue illness and differentiates face, both comprehensive analysis diagnostic result evaluates disease
People's entirety body and psychological condition, provide coordinating program.
Moreover, it will be appreciated that although the present specification is described in terms of embodiments, not each implementation method is only wrapped
Containing an independent technical scheme, this narrating mode of specification is only that for clarity, those skilled in the art should
Specification an as entirety, the technical scheme in each embodiment can also be formed into those skilled in the art through appropriately combined
May be appreciated other embodiment.
Claims (8)
1. a kind of enhanced lingual diagnosis system, including source illumination module, image capture module, host module, it is characterised in that institute
Source illumination module is stated for area source, the facial zone of sufficient light patient is described for described image acquisition module provides illumination
Image capture module includes high accuracy camera and ultra-low distortion camera lens, for carrying out IMAQ, the host module to patient
Including facial recognition modules and illness grader, the facial recognition modules are communicated to connect with described image acquisition module, can
The image information that reception is transmitted by described image acquisition module, and it is identified feedback, the illness grader and the face
Portion's identification module communication, is diagnosed with receiving the image information after identification.
2. enhanced lingual diagnosis system according to claim 1, it is characterised in that the source illumination module is LED array,
And with adjustable brightness of illumination and angle.
3. enhanced lingual diagnosis system according to claim 1, it is characterised in that described image acquisition module is WIFI industry
Camera, with USB storage, signal projector and signal receiver.
4. according to the enhanced lingual diagnosis system that any one of claim 1-3 is described, it is characterised in that the host module is
Also include display module, the display module communicates with the illness grader, for showing diagnostic result.
5. the tongue of the enhanced lingual diagnosis system described in claim 1-4 is as acquisition methods, it is characterised in that comprise the following steps:
1) adjustment source illumination module, image capture module make described image acquisition module clearly relative to the position of face
Gather the facial characteristics of face;
2) information transfer for gathering described image acquisition module carries out face recognition to the facial recognition modules of host module,
If face recognition fails, return to step 1), if face recognition success, carries out next step;
3) measured human face ratio, carries out facial feature localization, if facial feature localization fail, return to step 1), if facial feature localization into
Work(, carries out next step;
4) face information is obtained, while limiting lip-region, points out to stretch out tongue, carry out tongue image;
5) tongue image is obtained, and the disposal of gentle filter is carried out to the tongue image.
6. a kind of method that enhanced lingual diagnosis system using described in claim 1-4 carries out lingual diagnosis, it is characterised in that including:
By the tongue of the enhanced lingual diagnosis system described in claim 5 as acquisition methods acquisition face and tongue are as information;
By the face and tongue as the illness grader of information input to the host module;
The illness grader, as information carries out computing, show that illness is classified to the face and tongue.
7. method according to claim 6, it is characterised in that also including illness classification results are included in display module
On.
8. method according to claim 6, it is characterised in that the illness grader is the diagnosis based on BP neural network
Model, it is further comprising the steps of:
By the face and tongue as information input to the input layer of the BP neural network;
Train the BP neural network diagnostic model;
Adjust the model parameter of the BP neural network;
Obtain optimal BP neural network diagnostic model.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710012778.8A CN106725341A (en) | 2017-01-09 | 2017-01-09 | A kind of enhanced lingual diagnosis system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710012778.8A CN106725341A (en) | 2017-01-09 | 2017-01-09 | A kind of enhanced lingual diagnosis system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106725341A true CN106725341A (en) | 2017-05-31 |
Family
ID=58950637
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710012778.8A Pending CN106725341A (en) | 2017-01-09 | 2017-01-09 | A kind of enhanced lingual diagnosis system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106725341A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107811619A (en) * | 2017-12-08 | 2018-03-20 | 西安科技大学 | Portable pulse-taking instrument and its analysis method |
CN108553081A (en) * | 2018-01-03 | 2018-09-21 | 京东方科技集团股份有限公司 | A kind of diagnostic system based on tongue fur image |
CN109350011A (en) * | 2018-11-19 | 2019-02-19 | 陶磊 | A kind of lingual diagnosis device based on mobile intelligent terminal |
CN109995977A (en) * | 2019-03-31 | 2019-07-09 | 郑州大学 | A kind of tongue image collecting device |
CN111225614A (en) * | 2017-10-13 | 2020-06-02 | 佳能株式会社 | Diagnosis support device, information processing method, diagnosis support system, and program |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1731417A (en) * | 2005-08-19 | 2006-02-08 | 清华大学 | Method of robust human face detection in complicated background image |
US20060247502A1 (en) * | 2005-04-28 | 2006-11-02 | Eastman Kodak Company | Method for diagnosing disease from tongue image |
CN101406389A (en) * | 2007-10-10 | 2009-04-15 | 厦门大学 | Auxiliary Chinese patent medicine using system of traditional Chinese medicine |
CN201481396U (en) * | 2009-09-01 | 2010-05-26 | 上海道生医疗科技有限公司 | Observation detection device of traditional Chinese medical science |
CN103310179A (en) * | 2012-03-06 | 2013-09-18 | 上海骏聿数码科技有限公司 | Method and system for optimal attitude detection based on face recognition technology |
CN104510447A (en) * | 2015-01-13 | 2015-04-15 | 哈尔滨工业大学 | Visible light and near-infrared light sublingual image acquisition system |
CN105011903A (en) * | 2014-04-30 | 2015-11-04 | 上海华博信息服务有限公司 | Intelligent health diagnosis system |
CN105354565A (en) * | 2015-12-23 | 2016-02-24 | 北京市商汤科技开发有限公司 | Full convolution network based facial feature positioning and distinguishing method and system |
CN105877709A (en) * | 2016-04-07 | 2016-08-24 | 新绎健康科技有限公司 | Health condition data acquisition device, health condition data acquisition analysis system and method |
-
2017
- 2017-01-09 CN CN201710012778.8A patent/CN106725341A/en active Pending
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20060247502A1 (en) * | 2005-04-28 | 2006-11-02 | Eastman Kodak Company | Method for diagnosing disease from tongue image |
CN1731417A (en) * | 2005-08-19 | 2006-02-08 | 清华大学 | Method of robust human face detection in complicated background image |
CN101406389A (en) * | 2007-10-10 | 2009-04-15 | 厦门大学 | Auxiliary Chinese patent medicine using system of traditional Chinese medicine |
CN201481396U (en) * | 2009-09-01 | 2010-05-26 | 上海道生医疗科技有限公司 | Observation detection device of traditional Chinese medical science |
CN103310179A (en) * | 2012-03-06 | 2013-09-18 | 上海骏聿数码科技有限公司 | Method and system for optimal attitude detection based on face recognition technology |
CN105011903A (en) * | 2014-04-30 | 2015-11-04 | 上海华博信息服务有限公司 | Intelligent health diagnosis system |
CN104510447A (en) * | 2015-01-13 | 2015-04-15 | 哈尔滨工业大学 | Visible light and near-infrared light sublingual image acquisition system |
CN105354565A (en) * | 2015-12-23 | 2016-02-24 | 北京市商汤科技开发有限公司 | Full convolution network based facial feature positioning and distinguishing method and system |
CN105877709A (en) * | 2016-04-07 | 2016-08-24 | 新绎健康科技有限公司 | Health condition data acquisition device, health condition data acquisition analysis system and method |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111225614A (en) * | 2017-10-13 | 2020-06-02 | 佳能株式会社 | Diagnosis support device, information processing method, diagnosis support system, and program |
CN107811619A (en) * | 2017-12-08 | 2018-03-20 | 西安科技大学 | Portable pulse-taking instrument and its analysis method |
CN108553081A (en) * | 2018-01-03 | 2018-09-21 | 京东方科技集团股份有限公司 | A kind of diagnostic system based on tongue fur image |
US11471095B2 (en) | 2018-01-03 | 2022-10-18 | Beijing Boe Technology Development Co., Ltd. | Tongue-image-based diagnostic system and diagnostic method |
CN108553081B (en) * | 2018-01-03 | 2023-02-21 | 京东方科技集团股份有限公司 | Diagnosis system based on tongue fur image |
CN109350011A (en) * | 2018-11-19 | 2019-02-19 | 陶磊 | A kind of lingual diagnosis device based on mobile intelligent terminal |
CN109995977A (en) * | 2019-03-31 | 2019-07-09 | 郑州大学 | A kind of tongue image collecting device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106725341A (en) | A kind of enhanced lingual diagnosis system | |
CN107437092B (en) | The classification method of retina OCT image based on Three dimensional convolution neural network | |
CN110532887A (en) | A kind of method for detecting fatigue driving and system based on facial characteristics fusion | |
CN110175501B (en) | Face recognition-based multi-person scene concentration degree recognition method | |
CN109858540B (en) | Medical image recognition system and method based on multi-mode fusion | |
CN110169770A (en) | The fine granularity visualization system and method for mood brain electricity | |
CN108664947A (en) | A kind of fatigue driving method for early warning based on Expression Recognition | |
CN100511266C (en) | Organism eye determining method and organism eye determining device | |
CN109284738A (en) | Irregular face antidote and system | |
CN110287792A (en) | A kind of classroom Middle school students ' learning state real-time analysis method in nature teaching environment | |
CN107680683A (en) | A kind of AI eye healths appraisal procedure | |
CN110119672A (en) | A kind of embedded fatigue state detection system and method | |
CN109431523A (en) | Autism primary screening apparatus based on asocial's sonic stimulation behavior normal form | |
CN109948719A (en) | A kind of eye fundus image quality automatic classification method based on the intensive module network structure of residual error | |
CN109948671B (en) | Image classification method, device, storage medium and endoscopic imaging equipment | |
CN113159227A (en) | Acne image recognition method, system and device based on neural network | |
CN111461218B (en) | Sample data labeling system for fundus image of diabetes mellitus | |
CN110334649A (en) | A kind of five dirty situation of artificial vision's intelligence Chinese medicine facial diagnosis examines survey method and device | |
CN109902616A (en) | Face three-dimensional feature point detecting method and system based on deep learning | |
CN115281602B (en) | Dynamic analysis system for studying pupil light reflex obstacle for glaucoma | |
CN110495888A (en) | It is a kind of based on Chinese medicine tongue, the standard color card of face image and its application | |
CN109998497A (en) | System and plane of illumination illumination testing apparatus are sentenced in inspection of falling asleep in luminous environment | |
CN110288574A (en) | A kind of adjuvant Ultrasonographic Diagnosis hepatoncus system and method | |
CN110929570B (en) | Iris rapid positioning device and positioning method thereof | |
CN113974627B (en) | Emotion recognition method based on brain-computer generated confrontation |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170531 |