CN110473199A - A kind of detection of color spot acne and health assessment method based on the segmentation of deep learning example - Google Patents
A kind of detection of color spot acne and health assessment method based on the segmentation of deep learning example Download PDFInfo
- Publication number
- CN110473199A CN110473199A CN201910774229.3A CN201910774229A CN110473199A CN 110473199 A CN110473199 A CN 110473199A CN 201910774229 A CN201910774229 A CN 201910774229A CN 110473199 A CN110473199 A CN 110473199A
- Authority
- CN
- China
- Prior art keywords
- acne
- color spot
- face
- area
- detection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- 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
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30088—Skin; Dermal
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
- G06T2207/30201—Face
Landscapes
- Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Health & Medical Sciences (AREA)
- Public Health (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Quality & Reliability (AREA)
- Biomedical Technology (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Radiology & Medical Imaging (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Image Analysis (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention discloses a kind of detection of color spot acne and health assessment method based on the segmentation of deep learning example, comprising: acquisition color spot acne face direct picture marks color spot and acne present in front face, forms blocky face health problem data set；Training color spot acne example parted pattern, using two sections of channel selecting pre-training methods, first from ImageNet image classification data collection DClsTraining core network, then the lesion boundary partitioned data set D announced from ISICLBSTraining example parted pattern；On blocky face health problem data set, optimal 3 channels of example parted pattern are selected to obtain, retraining obtains color spot acne example parted pattern；User's face direct picture is acquired, color spot distribution map, acne distribution map are divided to obtain by the detection of color spot acne example parted pattern, obtain face color spot, the position of acne and area data；Face health degree is evaluated according to skin problem quantity, locating face subregion and intensity.
Description
Technical field
The present invention relates to field of image processing more particularly to it is a kind of for face color spot acne identify with assess image at
Reason method.
Background technique
In medical cosmetology industry, color spot and acne are one of most important processing problems in face skin problem, at present
How quickly domestic face color spot and acne detection recognition method mainly use human subjective to judge, lack objective quantification,
Accurately identification face skin splash acne problems become urgent problem to be solved, existing patent refer to about acne and color
The detection method of spot relates generally to traditional images processing method, as CN106529429A provides a kind of color spot detection mould
Block, but it is based on watershed algorithm, accuracy is not high, and to environmental requirement height, applicability is poor；As CN106449300A proposes one
Kind color spot detection recognition method, but the method described does not identify color spot edges of regions, while cannot accurately screen color spot wheel
Recognition result that is wide and excluding non-color spot region；CN108876766A proposes the acne judgment method based on deep learning, but
The not specific detection of description, method of counting.
1), " a kind of skin of face analysis system based on image recognition ", patent No. CN106529429A.The disclosure of the invention
A kind of skin of face analysis system based on image recognition.A kind of face analysis system of image recognition is proposed, it can be right
The case where skin, makes a series of assessment and provides overall evaluation, mainly includes that Face datection divides module, the inspection of whitening degree
Survey module, roughness measurement module, color spot amount detection module and Comprehensive Evaluation module.Detection image is inputted first, passes through classification
Device carries out positive face detection, zone location and segmentation and then leads to the image progress bright dark and ruddy situation analysis of the colour of skin that segmentation obtains
Gray level co-occurrence matrixes are crossed, the roughness for obtaining characteristic index quantization detection zone uses method and the zoning in similar watershed
The sum of sectional area analyzes color spot amount.This method provides a kind of color spot detection modules, but are based on watershed algorithm, accurately
Property it is not high, to environmental requirement height, applicability is poor.
2), " a kind of color spot detection recognition method ", patent No. CN106449300A.The invention provides a kind of color spot detection knowledge
Other method, acquires face picture, and face picture includes the mutation picture of all color spot types and color spot and marks color spot area in advance
Domain and color spot type are read out by feature vector to each color spot and markup information and are iterated until obtaining
The color characteristic mean value and Wavelet Transform Feature mean value of each color spot, to the feature vector two and markup information of each color spot
It is read out and is iterated until obtaining color mean value, texture mean value and the blob features mean value of each color spot, iteration
Process is using the color spot information finally read in face picture to be detected, with the color characteristic mean value and Wavelet Transform Feature
Mean value carries out matching and obtains color spot region and color mean value, texture mean value and blob features mean value progress match cognization color spot type
Export the color spot region obtained and color spot information.This method provides a kind of color spot detection recognition methods, but can not be quasi-
Really screening color spot profile and the recognition result in the non-color spot region of exclusion.
3), " acne judgment method, terminal and storage medium based on face recognition ", patent No. CN108876766A.It should
A kind of acne judgment method based on face recognition of disclosure of the invention is applied to terminal, including obtaining the deep learning after training
Model loads the face-image and the training of the acquisition face-image of the deep learning model after the training according to acquisition
Deep learning model afterwards carries out image recognition and obtains image recognition result.This method is referred to a kind of Cuo based on deep learning
Sore recognition methods, but do not describe its specific detection method and count scoring model, the method does not have reproducibility.
Summary of the invention
In order to solve the above technical problems, the object of the present invention is to provide a kind of color spot Cuo based on the segmentation of deep learning example
Sore detection and health assessment method, to meet quick and precisely detection face skin splash acne and provide the requirement of quantizating index,
Improve the applicability of measurement efficiency and measurement method.
The purpose of the present invention is realized by technical solution below:
A kind of detection of color spot acne and health assessment method based on the segmentation of deep learning example, comprising: acquisition health is asked
Topic facial image simultaneously marks, trains color spot acne example parted pattern, detection segmentation color spot and acne, evaluation face health journey
Degree；It specifically includes:
A acquires color spot acne face direct picture Iface, mark color spot I present in front facestain, acne Iacne,
Form blocky face health problem data set Dlump；
B trains color spot acne example parted pattern, using two sections of channel selecting pre-training methods, first from ImageNet image
Categorized data set DClsTraining core network ηmain, then from ISIC (International Skin Imaging
Collaboration, ISIC) announce lesion boundary partitioned data set DLBS(Lesion Boundary Segmentation,
LBS) training example parted pattern mpretrained；In blocky face health problem data set DlumpOn, select example parted pattern
mpretrainedOptimal 3 channels obtain mpretrained_3, retraining obtains color spot acne example parted pattern mdetect；
C acquires user's face direct picture, by color spot acne example parted pattern ηdetectColor spot distribution map is divided to obtain in detectionAcne distribution mapObtain face color spot, the position of acne, area data；
D evaluates face health degree according to skin problem quantity, locating face subregion and intensity.
Compared with prior art, one or more embodiments of the invention can have following advantage: the method achieve
The detection of color spot acne and health assessment based on the segmentation of deep learning example, applicability is good, under the premise of guaranteeing real-time, mentions
It has supplied to accurately identify the index for detecting the color spot acne for representing face health degree.
Detailed description of the invention
Fig. 1 is the detection of color spot acne and health assessment method work flow diagram based on the segmentation of deep learning example；
Fig. 2 is the detection of color spot acne and health assessment method program frame diagram based on the segmentation of deep learning example；
Fig. 3 is 30 acne skin problematic amount N of embodiment stepacneProCalculate schematic diagram；
Fig. 4 is skin problem quantity N in forehead ridge area in embodiment step 40TPro, cheek area skin problem quantity
NcheekPro, eye socket area skin problem quantity NeyePro, chin area skin problem quantity NjawProCalculate schematic diagram.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with examples and drawings to this hair
It is bright to be described in further detail.
As shown in Figure 1, for the detection of color spot acne and health assessment method workflow divided based on deep learning example,
Include the following steps:
Step 10 acquires color spot acne face direct picture Iface, mark color spot I present in front facestain, acne
Iacne, form blocky face health problem data set Dlump；
Step 20 trains color spot acne example parted pattern, using two sections of channel selecting pre-training methods, first from
ImageNet image classification data collection DClsTraining core network ηmain, then from ISIC (International Skin Imaging
Collaboration, ISIC) announce lesion boundary partitioned data set DLBS(Lesion Boundary Segmentation,
LBS) training example parted pattern mpretrained；In blocky face health problem data set DlumpOn, select example parted pattern
mpretrainedOptimal 3 channels obtain mpretrained_3, retraining obtains color spot acne example parted pattern ηdetect；
In blocky face health problem data set DlumpOn, select example parted pattern mpretrainedChannel, and delete it is extra
Channel obtains mpretrained_n, comprising:
If example partitioned data set DInstanceShared NclsKind identification object (containing background), if input face direct picture
IfaceResolution is Xin×Yin, example parted pattern mpretrainedOutput is Xin×Yin×NclsRecognition result matrix Ioutput。
In blocky face health problem data set DlumpOn, then there are the kinds such as color spot, acne, background to identify object, then mdetect
Output is Xin×Yin× 3 recognition result matrix Idetect.Thus according to the estimated performance of final convolutional layer, it may be selected best 3
A channel obtains mpretrained_3。
If mpretrainedAs a result identification outputN-th
Channel output probability seal is Ioutput_n, then it is most suitable for detecting the channel n of segmentation for color spotstainAre as follows:
In formula, fpixnum(I) it is counting function, represents the had pixel value of I as 1 quantity.
Similarly, it is most suitable for detecting the channel n of segmentation for acneacneAre as follows:
Similarly, it is most suitable for the channel n for background segmentBGAre as follows:
Step 30 acquires user's face direct picture, by color spot acne example parted pattern ηdetectColor spot is divided to obtain in detection
Distribution mapAcne distribution mapObtain face color spot, the position of acne, area data；It specifically includes:
User's face direct picture is acquired, by color spot acne example parted pattern ηdetectColor spot distribution map is divided to obtain in detectionColor spot quantity N can be obtainedstain, and have i-th of color spot area Astain_i, position (xstain_i,ystain_i)。
By color spot acne example parted pattern ηdetectAcne distribution map is divided to obtain in detectionAcne count can be obtained
Nacne, and have i-th of acne area Aacne_i, position (xacne_i,yacne_i)。
Further screen color spot area Astain_iOverstep the extreme limit TstainColor spot skin problem quantity NstainPro, area
Aacne_iOverstep the extreme limit TacneAcne skin problematic amount.
Then area oversteps the extreme limit TstainColor spot skin problem quantity NstainProCalculation formula are as follows:
H (x) is jump function in formula, is had
Similarly area oversteps the extreme limit TacneAcne skin problematic amount NacneProCalculation formula are as follows:
Such as Fig. 3, acne 8 are detected altogether on face, wherein 5 have been more than that area oversteps the extreme limit Tacne, therefore acne skin
Problematic amount NacnePro=5.
Step 40 evaluates face health degree according to skin problem quantity, locating face subregion and intensity.Specific packet
It includes:
Face is divided into forehead ridge area AT(area T), cheek area Acheek, eye socket area Aeye, chin area Ajaw, in the area Ze Ge
Problematic amount can be used for measuring concentration, such as forehead ridge area ATSkin problem quantity N in (area T)TProAre as follows:
Similarly cheek area skin problem quantity NcheekPro, eye socket area skin problem quantity NeyePro, chin area skin problem number
Measure NjawProIt is respectively as follows:
As shown in figure 4, forehead ridge area skin problem quantity NTPro=3, cheek area skin problem quantity NcheekPro=4,
Eye socket area skin problem quantity NeyePro=1, chin area skin problem quantity NjawPro=0.
Face health degree is evaluated according to skin problem quantity, locating face subregion and intensity.Particular by people
Color spot skin problem quantity N on the facestainPro, acne skin problematic amount NacnePro, forehead ridge area skin problem quantity
NTPro, cheek area skin problem quantity NcheekPro, eye socket area skin problem quantity NeyePro, chin area skin problem quantity NjawPro
Intelligent automatic Evaluation is carried out Deng 6 indexs.
Although disclosed herein embodiment it is as above, the content is only to facilitate understanding the present invention and adopting
Embodiment is not intended to limit the invention.Any those skilled in the art to which this invention pertains are not departing from this
Under the premise of the disclosed spirit and scope of invention, any modification and change can be made in the implementing form and in details,
But scope of patent protection of the invention, still should be subject to the scope of the claims as defined in the appended claims.
Claims (6)
1. a kind of detection of color spot acne and health assessment method based on the segmentation of deep learning example, which is characterized in that the side
Method include acquisition health problem facial image and mark, train color spot acne example parted pattern, detection segmentation color spot and acne,
Evaluate face health degree；The specific method is as follows:
A acquires color spot acne face direct picture Iface, mark color spot I present in front facestainWith acne Iacne, formed
Blocky face health problem data set Dlump；
B trains color spot acne example parted pattern, using two sections of channel selecting pre-training methods, first from ImageNet image classification
Data set DClsTraining core network ηmain, then the lesion boundary partitioned data set D announced from ISICLBSTraining example parted pattern
mpretrained；In blocky face health problem data set DlumpOn, select example parted pattern mpretrainedOptimal 3 channels obtain
To mpretrained_3, retraining obtains color spot acne example parted pattern ηdetect；
C acquires user's face direct picture, by color spot acne example parted pattern ηdetectColor spot distribution map is divided to obtain in detectionAcne distribution mapObtain face color spot, the position of acne and area data；
D evaluates face health degree according to skin problem quantity, locating face subregion and intensity.
2. the detection of color spot acne and health assessment method based on the segmentation of deep learning example as described in claim 1, special
Sign is, in blocky face health problem data set D in the step BlumpOn, select example parted pattern mpretrainedChannel,
And it deletes extra channel and obtains mpretrained_n, comprising:
If example partitioned data set DInstanceShared NclsKind identification object, if input face direct picture IfaceResolution is Xin
×Yin, example parted pattern mpretrainedOutput is Xin×Yin×NclsRecognition result matrix Ioutput；
In blocky face health problem data set DlumpOn, then there are the kinds such as color spot, acne, background to identify object, then mdetectOutput
For Xin×Yin× 3 recognition result matrix Idetect；Thus according to the estimated performance of final convolutional layer, it may be selected best 3 and lead to
Road obtains mpretrained_3；
If mpretrainedAs a result identification outputN-th of channel
Output probability seal is Ioutput_n, then it is most suitable for detecting the channel n of segmentation for color spotstainAre as follows:
In formula, fpixnum(I) it is counting function, represents the had pixel value of I as 1 quantity；
Similarly, it is most suitable for detecting the channel n of segmentation for acneacneAre as follows:
Similarly, it is most suitable for the channel n for background segmentBGAre as follows:
3. the detection of color spot acne and health assessment method based on the segmentation of deep learning example as described in claim 1, special
Sign is, user's face direct picture is acquired in the step C, by color spot acne example parted pattern ηdetectDetection is divided
Color spot distribution mapColor spot quantity N can be obtainedstain, and have i-th of color spot area Astain_i, position (xstain_i,
ystain_i)；
By color spot acne example parted pattern ηdetectAcne distribution map is divided to obtain in detectionAcne count N can be obtainedacne,
And there is i-th of acne area Aacne_i, position (xacne_i,yacne_i)。
4. the detection of color spot acne and health assessment method based on the segmentation of deep learning example as described in claim 1, special
Sign is, according to skin problem quantity circular in the step D are as follows:
Further screen color spot area Astain_iOverstep the extreme limit TstainColor spot skin problem quantity NstainPro, area Aacne_iIt is super
Cross limit TacneAcne skin problematic amount；
Then area oversteps the extreme limit TstainColor spot skin problem quantity NstainProCalculation formula are as follows:
H (x) is jump function in formula, is had
Similarly area oversteps the extreme limit TacneAcne skin problematic amount NacneProCalculation formula are as follows:
5. the detection of color spot acne and health assessment method based on the segmentation of deep learning example as described in claim 1, special
Sign is, face subregion and intensity circular in the step D are as follows:
Face is divided into forehead ridge area AT, cheek area Acheek, eye socket area Aeye, chin area Ajaw, the problems in area Ze Ge quantity
It can be used for measuring concentration, such as forehead ridge area ATSkin problem quantity N in (area T)TProAre as follows:
Similarly cheek area skin problem quantity NcheekPro, eye socket area skin problem quantity NeyePro, chin area skin problem quantity
NjawProIt is respectively as follows:
6. the detection of color spot acne and health assessment method based on the segmentation of deep learning example as described in claim 1, special
Sign is, evaluates face health degree according to skin problem quantity, locating face subregion and intensity in the step D；Tool
Body is by the color spot skin problem quantity N on facestainPro, acne skin problematic amount NacnePro, forehead ridge area skin
Problematic amount NTPro, cheek area skin problem quantity NcheekPro, eye socket area skin problem quantity NeyeProWith chin area skin problem
Quantity NjawPro6 indexs carry out intelligent automatic Evaluation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910774229.3A CN110473199B (en) | 2019-08-21 | 2019-08-21 | Mottle acne detection and health evaluation method based on deep learning example segmentation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910774229.3A CN110473199B (en) | 2019-08-21 | 2019-08-21 | Mottle acne detection and health evaluation method based on deep learning example segmentation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110473199A true CN110473199A (en) | 2019-11-19 |
| CN110473199B CN110473199B (en) | 2022-09-27 |
Family
ID=68513181
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910774229.3A Active CN110473199B (en) | 2019-08-21 | 2019-08-21 | Mottle acne detection and health evaluation method based on deep learning example segmentation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110473199B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111862118A (en) * | 2020-07-20 | 2020-10-30 | 四川大学 | Pressure sore staging training method, staging method and staging system |
| CN112464885A (en) * | 2020-12-14 | 2021-03-09 | 上海交通大学 | Image processing system for future change of facial color spots based on machine learning |
| CN112509688A (en) * | 2020-09-25 | 2021-03-16 | 卫宁健康科技集团股份有限公司 | Automatic analysis system, method, equipment and medium for pressure sore picture |
| CN112967285A (en) * | 2021-05-18 | 2021-06-15 | 中国医学科学院皮肤病医院(中国医学科学院皮肤病研究所) | Chloasma image recognition method, system and device based on deep learning |
| CN113128373A (en) * | 2021-04-02 | 2021-07-16 | 西安融智芙科技有限责任公司 | Color spot scoring method based on image processing, color spot scoring device and terminal equipment |
| CN113128375A (en) * | 2021-04-02 | 2021-07-16 | 西安融智芙科技有限责任公司 | Image recognition method, electronic device and computer-readable storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10032281B1 (en) * | 2017-05-03 | 2018-07-24 | Siemens Healthcare Gmbh | Multi-scale deep reinforcement machine learning for N-dimensional segmentation in medical imaging |
| CN108876766A (en) * | 2018-05-23 | 2018-11-23 | 平安科技（深圳）有限公司 | Acne judgment method, terminal and storage medium based on face recognition |
| CN110097034A (en) * | 2019-05-15 | 2019-08-06 | 广州纳丽生物科技有限公司 | A kind of identification and appraisal procedure of Intelligent human-face health degree |
-
2019
- 2019-08-21 CN CN201910774229.3A patent/CN110473199B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10032281B1 (en) * | 2017-05-03 | 2018-07-24 | Siemens Healthcare Gmbh | Multi-scale deep reinforcement machine learning for N-dimensional segmentation in medical imaging |
| CN108876766A (en) * | 2018-05-23 | 2018-11-23 | 平安科技（深圳）有限公司 | Acne judgment method, terminal and storage medium based on face recognition |
| CN110097034A (en) * | 2019-05-15 | 2019-08-06 | 广州纳丽生物科技有限公司 | A kind of identification and appraisal procedure of Intelligent human-face health degree |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111862118A (en) * | 2020-07-20 | 2020-10-30 | 四川大学 | Pressure sore staging training method, staging method and staging system |
| CN112509688A (en) * | 2020-09-25 | 2021-03-16 | 卫宁健康科技集团股份有限公司 | Automatic analysis system, method, equipment and medium for pressure sore picture |
| CN112509688B (en) * | 2020-09-25 | 2024-06-11 | 卫宁健康科技集团股份有限公司 | Automatic analysis system, method, equipment and medium for pressure sore picture |
| CN112464885A (en) * | 2020-12-14 | 2021-03-09 | 上海交通大学 | Image processing system for future change of facial color spots based on machine learning |
| CN113128373A (en) * | 2021-04-02 | 2021-07-16 | 西安融智芙科技有限责任公司 | Color spot scoring method based on image processing, color spot scoring device and terminal equipment |
| CN113128375A (en) * | 2021-04-02 | 2021-07-16 | 西安融智芙科技有限责任公司 | Image recognition method, electronic device and computer-readable storage medium |
| CN113128373B (en) * | 2021-04-02 | 2024-04-09 | 西安融智芙科技有限责任公司 | Image processing-based color spot scoring method, color spot scoring device and terminal equipment |
| CN113128375B (en) * | 2021-04-02 | 2024-05-10 | 西安融智芙科技有限责任公司 | Image recognition method, electronic device, and computer-readable storage medium |
| CN112967285A (en) * | 2021-05-18 | 2021-06-15 | 中国医学科学院皮肤病医院(中国医学科学院皮肤病研究所) | Chloasma image recognition method, system and device based on deep learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110473199B (en) | 2022-09-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110473199A (en) | A kind of detection of color spot acne and health assessment method based on the segmentation of deep learning example | |
| CN110097034A (en) | A kind of identification and appraisal procedure of Intelligent human-face health degree | |
| CN109961426B (en) | Method for detecting skin of human face | |
| CN105139004B (en) | Facial expression recognizing method based on video sequence | |
| CN109636772A (en) | The defect inspection method on the irregular shape intermetallic composite coating surface based on deep learning | |
| CN109635875A (en) | A kind of end-to-end network interface detection method based on deep learning | |
| CN107169953A (en) | Bridge concrete surface crack detection method based on HOG features | |
| CN110210448B (en) | Intelligent face skin aging degree identification and evaluation method | |
| CN104299011A (en) | Skin type and skin problem identification and detection method based on facial image identification | |
| CN104992449A (en) | Information identification and surface defect on-line detection method based on machine visual sense | |
| JP2021504816A (en) | Bone age evaluation and height prediction model, its system and its prediction method | |
| CN104715238A (en) | Pedestrian detection method based on multi-feature fusion | |
| CN109840483B (en) | Landslide crack detection and identification method and device | |
| CN111967363B (en) | Emotion prediction method based on micro-expression recognition and eye movement tracking | |
| CN104636755A (en) | Face beauty evaluation method based on deep learning | |
| CN112818775B (en) | Forest road rapid identification method and system based on regional boundary pixel exchange | |
| CN111563452A (en) | Multi-human body posture detection and state discrimination method based on example segmentation | |
| CN112907519A (en) | Metal curved surface defect analysis system and method based on deep learning | |
| Chen et al. | CUFuse: Camera and ultrasound data fusion for rail defect detection | |
| CN103034838A (en) | Special vehicle instrument type identification and calibration method based on image characteristics | |
| CN101726498B (en) | Intelligent detector and method of copper strip surface quality on basis of vision bionics | |
| TW202004776A (en) | Establishing method of bone age assessment and height prediction model, bone age assessment and height prediction system, and bone age assessment and height prediction method | |
| CN108921004A (en) | Safety cap wears recognition methods, electronic equipment, storage medium and system | |
| CN113240623B (en) | Pavement disease detection method and device | |
| CN104778684A (en) | Method and system thereof for automatically measuring, representing and classifying heterogeneous defects on surface of steel |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |