CN111144479A - Traditional Chinese medicine face color recognition method based on image processing - Google Patents
Traditional Chinese medicine face color recognition method based on image processing Download PDFInfo
- Publication number
- CN111144479A CN111144479A CN201911358025.8A CN201911358025A CN111144479A CN 111144479 A CN111144479 A CN 111144479A CN 201911358025 A CN201911358025 A CN 201911358025A CN 111144479 A CN111144479 A CN 111144479A
- Authority
- CN
- China
- Prior art keywords
- face
- face color
- color
- chinese medicine
- image processing
- 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.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- 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
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- 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
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
Abstract
The invention discloses a traditional Chinese medicine complexion identification method based on image processing, which is used for quantifying traditional Chinese medicine complexion classification by using an image processing method. And (3) segmenting by using an elliptical skin color model of the YCbCr color space, setting the YCbCr color space to meet conditions, and segmenting the face from the complex background. The Dlib feature point positioning is used for accurately finding the position of the face interesting region, so that the feature point extraction is facilitated. And (3) effectively extracting the face color feature information by using an LBP feature extraction method. And classifying the extracted characteristic information by using an SVM classifier to obtain an accurate face color classification model. The face color identification method based on image processing needs not much face color data, which has good effect on the premise of difficult acquisition of face color data, and obtains the best classification result by the minimized data. The face color feature extraction and classification process in the face color identification method based on image processing can be manually regulated and controlled, and an optimal classification model can be obtained.
Description
Technical Field
The invention relates to medical signal processing, in particular to a traditional Chinese medicine face color identification method based on image processing.
Background
The four diagnostic methods of traditional Chinese medicine comprise inspection, auscultation, inquiry and palpation, wherein the related research on tongue diagnosis in inspection of traditional Chinese medicine progresses rapidly, and part of research results are applied to clinic, while the inspection for complexion in inspection is relatively less. In traditional Chinese medicine, it is considered that viscera, psychology, qi and blood and meridian changes can be shown in relevant areas of human face. When the inspection diagnosis is performed, the face color of the human face is observed for judgment. The complexion of traditional Chinese medicine can be divided into 5 types of green, red, yellow, white and black, wherein the green and black can indicate pain, the red and yellow can indicate heat syndrome, and the white can indicate cold syndrome.
At present, the traditional Chinese medicine complexion identification and diagnosis mainly depends on the experience accumulation of traditional Chinese medicine doctors to make judgment, the result is often limited by the personal experience of the doctors, and meanwhile, the traditional Chinese medicine complexion identification and diagnosis has larger subjectivity due to the influence of external light and the like. In addition, the complexion diagnosis in clinic at present still lacks of fixed evaluation standards, and quantitative analysis of the complexion diagnosis by adopting a computer technology can enable the traditional Chinese medicine complexion clinical diagnosis to be objective.
Disclosure of Invention
The invention aims to provide a traditional Chinese medicine face color identification method based on image processing.
The technical scheme adopted by the invention is as follows:
the traditional Chinese medicine face color identification method based on image processing comprises the following steps:
step 1, acquiring traditional Chinese medicine complexion image data;
step 2, performing color space conversion on the face color image, converting the face color image from an RGB space to a YCbCr color space, segmenting a face region of interest by adopting an elliptical skin color model of the YCbCr color space, and filtering the influence of other background factors; the specific steps of the step 2 are as follows:
step 2-1, firstly, an average filtering is carried out on the input face color image for noise reduction, and high-frequency information, such as boundary information, in the image is filtered.
Step 2-2, the values of Cb and Cr in the YCbCr color space are calculated, with the definitions Wcb-46.97, Wcr-38.76, WHCb-14, WHCr-10, WLCb-23, WLCr-20, Ymin-16, Ymax-255, Kl-128, Kh-188, WCb-0, Wcr-0, cbenter-0, CrCenter-0, satisfying the following:
and 2-3, segmenting the face color region of interest by using an elliptical skin color model in the YCbCr color space, obtaining Cb and Cr values based on calculation, wherein pixel points which meet the skin color brightness range of 133-173 Cb and 77-127 Cb in the YCbCr color space are set to be 1, and otherwise, the pixel points are set to be 0.
Step 3, carrying out Dlib facial feature point positioning on the segmented facial region of interest to obtain the positions of the forehead, the fundus part and the lower jaw of the face in the image;
step 4, extracting feature information of the forehead, the fundus part and the lower jaw of the face obtained by positioning by adopting an LBP feature extraction method;
step 5, extracting characteristic information, and performing face color classification by using an SVM classifier to obtain a face color classification model; the classification categories comprise 5 kinds of green, red, yellow, white and black;
and 6, inputting the face color image, and predicting the category of the face color image through a face color classification model.
By adopting the technical scheme, the invention quantifies the Chinese medicine complexion classification by using an image processing method, and is beneficial to the auxiliary diagnosis of doctors. And the YCbCr color space elliptical skin color model is used for segmentation, and the YCbCr color space is set to meet the conditions, so that the face of the human face is segmented from the complex background, and the segmentation effect is excellent. The Dlib feature point positioning is used for accurately finding the position of the face interesting region, so that the feature point extraction is facilitated. By using the LBP feature extraction method, the face color feature information can be effectively extracted. The extracted characteristic information can be classified by using an SVM classifier to obtain an accurate face color classification model, and the classification categories comprise 5 types of cyan, red, yellow, white and black. The face color identification method based on image processing needs not much face color data, which has good effect on the premise of difficult acquisition of face color data, and obtains the best classification result by the minimized data. The face color feature extraction and classification process in the face color identification method based on image processing can be manually regulated and controlled, and an optimal classification model can be obtained.
The invention utilizes the calculation technology to quantify 5 complexion categories of the traditional Chinese medicine complexion diagnosis, and adopts the related method of image processing to carry out processing and analysis, and finally obtains a stable model for analyzing the diagnosis result.
Drawings
The invention is described in further detail below with reference to the accompanying drawings and the detailed description;
fig. 1 is a schematic flow chart of the traditional Chinese medicine face color identification method based on image processing.
Detailed Description
The Chinese medicine complexion recognition occupies a very important position in the Chinese medicine inspection diagnosis, and plays a role in auxiliary diagnosis for the integral diagnosis of the Chinese medicine. The invention provides a Chinese medicine face color identification method by quantifying the face color of a face diagnosed by a Chinese doctor by utilizing image processing and computer technology.
As shown in fig. 1, the invention discloses a traditional Chinese medicine face color identification method based on image processing, which comprises the following steps:
step 1, acquiring traditional Chinese medicine complexion image data;
step 2, performing color space conversion on the face color image, converting the face color image from an RGB space to a YCbCr color space, segmenting a face region of interest by adopting an elliptical skin color model of the YCbCr color space, and filtering the influence of other background factors; the specific steps of the step 2 are as follows:
step 2-1, firstly, an average filtering is carried out on the input face color image for noise reduction, and high-frequency information, such as boundary information, in the image is filtered.
Step 2-2, the values of Cb and Cr in the YCbCr color space are calculated, with the definitions Wcb-46.97, Wcr-38.76, WHCb-14, WHCr-10, WLCb-23, WLCr-20, Ymin-16, Ymax-255, Kl-128, Kh-188, WCb-0, Wcr-0, cbenter-0, CrCenter-0, satisfying the following:
and 2-3, segmenting the face color region of interest by using an elliptical skin color model in the YCbCr color space, obtaining Cb and Cr values based on calculation, wherein pixel points which meet the skin color brightness range of 133-173 Cb and 77-127 Cb in the YCbCr color space are set to be 1, and otherwise, the pixel points are set to be 0.
Step 3, carrying out Dlib facial feature point positioning on the segmented facial region of interest to obtain the positions of the forehead, the fundus part and the lower jaw of the face in the image;
step 4, extracting feature information of the forehead, the fundus part and the lower jaw of the face obtained by positioning by adopting an LBP feature extraction method;
step 5, extracting characteristic information, and performing face color classification by using an SVM classifier to obtain a face color classification model; the classification categories include 5 kinds of green, red, yellow, white and black.
And 6, inputting the face color image, and predicting the category of the face color image through a face color classification model.
By adopting the technical scheme, the invention quantifies the Chinese medicine complexion classification by using an image processing method, and is beneficial to the auxiliary diagnosis of doctors. And the YCbCr color space elliptical skin color model is used for segmentation, and the YCbCr color space is set to meet the conditions, so that the face of the human face is segmented from the complex background, and the segmentation effect is excellent. The Dlib feature point positioning is used for accurately finding the position of the face interesting region, so that the feature point extraction is facilitated. By using the LBP feature extraction method, the face color feature information can be effectively extracted. The extracted characteristic information can be classified by using an SVM classifier to obtain an accurate face color classification model, and the classification categories comprise 5 types of cyan, red, yellow, white and black. The face color identification method based on image processing needs not much face color data, which has good effect on the premise of difficult acquisition of face color data, and obtains the best classification result by the minimized data. The face color feature extraction and classification process in the face color identification method based on image processing can be manually regulated and controlled, and an optimal classification model can be obtained.
The invention utilizes the calculation technology to quantify 5 complexion categories of the traditional Chinese medicine complexion diagnosis, and adopts the related method of image processing to carry out processing and analysis, and finally obtains a stable model for analyzing the diagnosis result.
Claims (6)
1. The traditional Chinese medicine face color identification method based on image processing is characterized by comprising the following steps: which comprises the following steps:
step 1, acquiring complexion image data confirmed by traditional Chinese medicine;
step 2, performing color space conversion on the face color image, converting the face color image from an RGB space to a YCbCr color space, segmenting the face region of interest, and filtering the influence of other background factors;
step 3, carrying out Dlib facial feature point positioning on the segmented facial region of interest to obtain the positions of the forehead, the fundus part and the lower jaw of the face in the image;
step 4, extracting characteristic information of the forehead, the fundus part and the lower jaw of the face obtained by positioning by adopting an LBP (local binary pattern) characteristic extraction method;
step 5, extracting characteristic information, and performing face color classification by using an SVM classifier to obtain a face color classification model;
and 6, inputting the face color image, and predicting the category of the face color image through a face color classification model.
2. The method for traditional Chinese medicine face color recognition based on image processing according to claim 1, wherein: and 2, segmenting the face region of interest by adopting an elliptical skin color model of a YCbCr space.
3. The method for traditional Chinese medicine face color recognition based on image processing according to claim 1, wherein: the step 2 specifically comprises the following steps:
step 2-1, firstly, carrying out mean value filtering and noise reduction on an input face color image, and filtering high-frequency information in the image;
step 2-2, calculating Cb and Cr values of the acquired YCbCr color space,
and 2-3, segmenting the face color region of interest by using an elliptical skin color model in the YCbCr color space, and setting pixel points which meet the skin color brightness range of 133-173 and 77-127 in the YCbCr color space as 1 by utilizing the Cb and Cr values obtained by calculation in the step 2-2, otherwise, setting the pixel points as 0.
4. The method for traditional Chinese medicine face color recognition based on image processing according to claim 3, wherein: the high frequency information in step 2-1 includes boundary information.
5. The method for traditional Chinese medicine face color recognition based on image processing according to claim 3, wherein: the specific calculation method of the step 2-2 is as follows: cb. The value of Cr satisfies the following condition:
wherein Wcb-46.97, Wcr-38.76, WHCb-14, WHCr-10, WLCb-23, WLCr-20, Ymin-16, Ymax-255, Kl-128, Kh-188, WCb-0, Wcr-0, cbcnter-0, and CrCenter-0.
6. The method for traditional Chinese medicine face color recognition based on image processing according to claim 1, wherein: and 5, classifying the face color classification model obtained in the step 5 into 5 classes including cyan, red, yellow, white and black.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911358025.8A CN111144479A (en) | 2019-12-25 | 2019-12-25 | Traditional Chinese medicine face color recognition method based on image processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911358025.8A CN111144479A (en) | 2019-12-25 | 2019-12-25 | Traditional Chinese medicine face color recognition method based on image processing |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111144479A true CN111144479A (en) | 2020-05-12 |
Family
ID=70520081
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911358025.8A Withdrawn CN111144479A (en) | 2019-12-25 | 2019-12-25 | Traditional Chinese medicine face color recognition method based on image processing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111144479A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113140309A (en) * | 2021-04-14 | 2021-07-20 | 五邑大学 | Traditional Chinese medicine complexion diagnosis method and device |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170231550A1 (en) * | 2014-08-25 | 2017-08-17 | Singapore University Of Technology And Design | Method and device for analysing an image |
| CN109063542A (en) * | 2018-06-11 | 2018-12-21 | 平安科技（深圳）有限公司 | Image identification method, device, computer equipment and storage medium |
-
2019
- 2019-12-25 CN CN201911358025.8A patent/CN111144479A/en not_active Withdrawn
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170231550A1 (en) * | 2014-08-25 | 2017-08-17 | Singapore University Of Technology And Design | Method and device for analysing an image |
| CN109063542A (en) * | 2018-06-11 | 2018-12-21 | 平安科技（深圳）有限公司 | Image identification method, device, computer equipment and storage medium |
| WO2019237548A1 (en) * | 2018-06-11 | 2019-12-19 | 平安科技（深圳）有限公司 | Picture recognition method and device, computer device and storage medium |
Non-Patent Citations (4)
| Title |
|---|
| 李燕等: "一种基于椭圆肤色模型的人脸检测方法", 《计算机测量与控制》 * |
| 胡章芳 等: "《MATLAB仿真及其在光学课程中的应用 第2版》", 30 April 2018, 北京航空航天大学出版社 * |
| 裔隽 等: "《Python机器学习实战》", 28 February 2018, 科学技术文献出版社 * |
| 陈梦竹等: "基于图像处理的望诊面色自动识别研究", 《中国中医药信息杂志》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113140309A (en) * | 2021-04-14 | 2021-07-20 | 五邑大学 | Traditional Chinese medicine complexion diagnosis method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109472781B (en) | Diabetic retinopathy detection system based on serial structure segmentation | |
| Senan et al. | Analysis of dermoscopy images by using ABCD rule for early detection of skin cancer | |
| WO2019196268A1 (en) | Diabetic retina image classification method and system based on deep learning | |
| CN108695000B (en) | Ultrasonic image-based intelligent diagnosis system for diffuse thyroid diseases | |
| Vocaturo et al. | Features for melanoma lesions characterization in computer vision systems | |
| CN108670297B (en) | Multi-mode transcranial ultrasound-based Parkinson's disease auxiliary analysis system and method | |
| CN104299242A (en) | Fluorescence angiography fundus image extraction method based on NGC-ACM | |
| CN111839492B (en) | Heart rate non-contact type measuring method based on face video sequence | |
| CN112927187A (en) | Method for automatically identifying and positioning focal cortical dysplasia epileptic focus | |
| Abbes et al. | High-level features for automatic skin lesions neural network based classification | |
| CN113130066A (en) | Tongue diagnosis image identification method based on artificial intelligence | |
| CN105869151B (en) | Tongue segmentation and tongue fur tongue nature separation method | |
| CN114332910A (en) | Human body part segmentation method for similar feature calculation of far infrared image | |
| CN108230341B (en) | Eyeground image blood vessel segmentation method based on layered matting algorithm | |
| CN111144479A (en) | Traditional Chinese medicine face color recognition method based on image processing | |
| CN111222371A (en) | Sublingual vein feature extraction device and method | |
| CN112560911A (en) | Tongue image classification method and system in traditional Chinese medicine | |
| CN110390678B (en) | Tissue type segmentation method of colorectal cancer IHC staining image | |
| CN109711306B (en) | Method and equipment for obtaining facial features based on deep convolutional neural network | |
| Zhou et al. | Automated detection of red lesions using superpixel multichannel multifeature | |
| CN108629780B (en) | Tongue image segmentation method based on color decomposition and threshold technology | |
| Hameed et al. | Hybrid technique for skin pimples image detection and classification | |
| Dey et al. | Red-plane asymmetry analysis of breast thermograms for cancer detection | |
| Liu et al. | Hepatitis diagnosis using facial color image | |
| Kekal et al. | Optimization of Melanoma Skin Cancer Detection with the Convolutional Neural Network |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20200512 |
|
| WW01 | Invention patent application withdrawn after publication |