CN107292894A - A kind of method and system for being handled tank spot characteristics of image - Google Patents
A kind of method and system for being handled tank spot characteristics of image Download PDFInfo
- Publication number
- CN107292894A CN107292894A CN201710508178.0A CN201710508178A CN107292894A CN 107292894 A CN107292894 A CN 107292894A CN 201710508178 A CN201710508178 A CN 201710508178A CN 107292894 A CN107292894 A CN 107292894A
- Authority
- CN
- China
- Prior art keywords
- spot image
- tank spot
- characteristic
- tank
- color
- 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
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- 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
- G06T2207/20021—Dividing image into blocks, subimages or windows
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of method for being handled tank spot characteristics of image, methods described includes：To tank spot image zooming-out characteristic information, the characteristic information is input to the good SVM classifier of training in advance, the characteristic information of the tank spot image handled by SVM classifier, the classification of the tank spot image is obtained；The characteristic information includes：The percentage and positional information of characteristic color in tank spot image, and characteristic area type and the positional information of characteristic area.Using technical scheme, user only needs to hand-held mobile terminal and taken pictures that analysis to tank spot image can be achieved, and obtains accurate analysis result, realize the analysis of tank spot theorize with it is scientific.
Description
Technical field
The present invention relates to traditional medicine digital Clinics field, it is used to enter tank spot characteristics of image more particularly, to one kind
The method and system of row processing.
Background technology
Tank spot or tank print are that cup is risen after tank, pull the normal appearance point sheet aubergine in position, petechia, ecchymosis, or with micro-
The normal therapeutic response of heat pain.Tank spot is also the external reflection of internal pathology, distinguishes that color is qualitative for tank spot, can be to a certain degree
The evil property of the upper prosperity and decline for inferring body qi and blood function and sense, so as to reflect the different conditions of body and the response to treatment of disease.
Tank spot is diagnosed a disease and is clinically widely used, and it is summed up according to mainly ancient Chinese medicine doctor by long-term clinical practice
The rule that the tank spot come is diagnosed a disease, it is the supplement of the diagnostic method of TCM, and the further investigation to it is the theoretical succession of centering medical tradition
And development.Diagnosed a disease using tank spot, it is both easy to use, it can find that body hides or still undiscovered illness at any time again, department is outer to carry interior
And internal function state is seen clearly, and cupping for treatment can be carried out simultaneously, integrate diagnosis and treatment.
Current Chinese society aging population is serious, the gesture of the threat for disease of waiting a moment spreads, the popularization of " preventiveing treatment of disease " concept with
Health of people theory has all been lifted, and the Chinese physical therapy method such as cup, massage, acupuncture is widely used in health care.In cup
In physiotherapy, the problem of body is hiding or not yet finds quickly is found by the easy analysis to tank spot, and can be pulled out simultaneously
Tank is treated, and integrates diagnosis and treatment.The subjective perception and empirical cumulative for relying primarily on doctor are distinguished tank spot at present, are not had still at present
The diagnostic analysis equipment large-scale application analyzed towards tank spot, in tank spot analysis method, due to being stopped for tank spot correlation theory
In the description of ancients' phenomenological language, the classification for tank spot lacks clear and definite standard, and then can not set up tank spot with health
Accurate corresponding relation between state, realizes the analysis of tank spot applied to the theorizing of physiotherapy, scientific.
Tank spot characteristics of image is carried out in the prior art diagnosis be by people's subjective judgement, may be because for same tank spot
For different people's subjective judgements, different results are drawn.And cause experience to be accumulated because the number of patients of everyone contact is limited
It is tired limited, influence the result judged.
Accordingly, it would be desirable to a kind of technology, to improve the accuracy to the processing of tank spot characteristics of image.
The content of the invention
The invention provides a kind of method and system for being handled tank spot characteristics of image, how to solve to tank
The problem of spot image is analyzed and processed.
In order to solve the above problems, the invention provides a kind of method for being handled tank spot characteristics of image, institute
The method of stating includes：
To tank spot image zooming-out characteristic information, the characteristic information is input to the good SVM classifier of training in advance, passed through
SVM classifier is handled the characteristic information of the tank spot image, obtains the classification of the tank spot image；
The characteristic information includes：The percentage and positional information of characteristic color in tank spot image, and characteristic area class
The positional information of type and characteristic area.
Preferably, methods described is additionally included in before tank spot image zooming-out characteristic information, to including the original of tank spot image
Image is pre-processed, and the pretreatment includes carrying out dividing processing to tank spot image and background image, obtains tank spot image, institute
Stating pretreatment also includes the one or more in denoising, correction and blank level adjustment processing.
Preferably, wherein obtaining the percentage of characteristic color in tank spot image, when positional information includes：
The tank spot image is divided into multiple regions；
Obtain the characteristic color accounting in each region of tank spot image after the segmentation：
The percentage of characteristic color is characteristic color in acquired cut zone in tank spot image in the characteristic information
Account for the percentage of all characteristic colors, or characteristic color accounts for the percentage of all colours in cut zone in acquired cut zone
Than；
Obtain the centroid position in each region of tank spot image after the segmentation；It is special in tank spot image in the characteristic information
The corresponding positional information of color percentage is levied for the centroid position of the cut zone obtained；Or preset the volume of cut zone
Number, the numbering of the cut zone is the positional information.
Preferably, the location information acquisition method of the characteristic area type and characteristic area is：
Identification tank spot color of image value belongs to the pixel of default color value scope；
By morphological image process, the shape of the characteristic area in each region is determined；
The type of characteristic area is determined according to preparatory condition；The preparatory condition includes：Pre-set color scope, default feature
One or more in region area and default characteristic area shape information；
For meeting preparatory condition characteristic area, centroid position is calculated；
The characteristic area centroid position is the positional information of the characteristic area.
Preferably, the location information acquisition method of the characteristic area and characteristic area is：
The tank spot image is divided into multiple regions, and each region is numbered；
The region after the segmentation where the characteristic area centroid position is determined, the numbering of the cut zone is institute's rheme
Confidence ceases.
Preferably, it is described that dividing processing is carried out to tank spot image and background image, tank spot image is obtained, including：
Using one or more of RGB and/or hsv color value channel recognition tank spot image border, Threshold segmentation is utilized
Method splits the initial pictures, obtains tank spot image；Or
Using the channel B identification tank spot image border in RGB, split the initial pictures using threshold segmentation method, obtain
Take tank spot image.Count the inter-class variance of two classes inside and outside the tank spot image border, obtain variance it is maximum when threshold value as point
Threshold value is cut, the initial pictures are split.
Preferably, in addition to：The characteristic color in the tank spot image is specified, including：It is powder, red, blue or green, purple and white.Institute
Stating the color value of characteristic color includes the one or more in RGB, HSV or LAB in color value.
Preferably, it is described that the tank spot image is divided into multiple regions, including following splitting scheme：
The circumscribed rectangle of the tank spot image-region is obtained, the region of the rectangle is divided into multiple regions；Or
The circumscribed circle of the tank spot image-region is obtained, the circular region is divided into multiple regions；Or
The circumscribed circle of the tank spot image-region is obtained, the circular region is divided into multiple areas in radial direction
Domain.
Preferably, the cut zone in claim 3 and claim 4 is split by same splitting scheme,
Or different splitting scheme is split.
Based on another aspect of the present invention, the present invention provides a kind of be used for tank spot characteristics of image progress processing system, institute
Stating system includes photographing module, color correction module, data processing module, data transmission module, high in the clouds analysis module；
The photographing module, for obtaining tank spot image；
The color correction module is made up of default light source and/or default radical occlusion device, or by color with reference to module composition,
The photographing module obtains tank spot image and obtains the color simultaneously with reference to module map picture；
The data processing module, extracts the characteristic information of the tank spot image；
The data transmission module, for the characteristic information of the tank spot image to be uploaded into high in the clouds analysis module；
The high in the clouds analysis module, for the characteristic information received to be input into the good svm classifier of training in advance
Device, is handled the tank spot image by SVM classifier, obtains the classification of the tank spot image.
The technical scheme that the present invention is provided is by the way that to tank spot image zooming-out characteristic information, advance instruction is input to by characteristic information
The SVM classifier perfected, is handled the characteristic information of tank spot image by SVM classifier, obtains the classification of tank spot image.
Technical scheme is realized to tank spot image science, objective analysis, can utilized by the feature of extractor spot image
Intelligent analysis method, tank that is accurate, intelligently analyzing needs examines conclusion.Using technical scheme, user only needs to
Hand-held mobile terminal, which is taken pictures, can be achieved analysis to tank spot image, obtains accurate analysis result, realizes the reason of tank spot analysis
By change and it is scientific.
Brief description of the drawings
By reference to the following drawings, the illustrative embodiments of the present invention can be more fully understood by：
Fig. 1 is a kind of method flow diagram for being handled tank spot characteristics of image according to embodiment of the present invention；
Fig. 2 is a kind of schematic diagram for being pre-processed to tank spot characteristics of image according to embodiment of the present invention；
Fig. 3 is a kind of circumscribed rectangle splitting scheme structural representation according to embodiment of the present invention；
Fig. 4 is a kind of circumscribed fan-segmentation scenario-frame schematic diagram according to embodiment of the present invention；
Fig. 5 is a kind of circumscribed annulus splitting scheme structural representation according to embodiment of the present invention；And
Fig. 6 is to be used to carry out processing system structure chart to tank spot characteristics of image according to one kind of embodiment of the present invention.
Embodiment
The illustrative embodiments of the present invention are introduced with reference now to accompanying drawing, however, the present invention can use many different shapes
Formula is implemented, and it is to disclose at large and fully there is provided these embodiments to be not limited to embodiment described herein
The present invention, and fully pass on the scope of the present invention to person of ordinary skill in the field.For showing for being illustrated in the accompanying drawings
Term in example property embodiment is not limitation of the invention.In the accompanying drawings, identical cells/elements are attached using identical
Icon is remembered.
Unless otherwise indicated, term (including scientific and technical terminology) used herein has to person of ordinary skill in the field
It is common to understand implication.Further it will be understood that the term limited with usually used dictionary, is appreciated that and it
The linguistic context of association area has consistent implication, and is not construed as Utopian or excessively formal meaning.
Fig. 1 is a kind of method flow diagram for being handled tank spot characteristics of image according to embodiment of the present invention.
Embodiment of the present invention provide a kind of method for being handled tank spot characteristics of image, as shown in figure 1, method 100 from
Step 101 starts：
Preferably, before step 101, before to tank spot image zooming-out characteristic information, to including the original of tank spot image
Image is pre-processed, and pretreatment includes carrying out dividing processing to tank spot image and background image, obtains tank spot image, pretreatment
Also include the one or more in denoising, correction and blank level adjustment processing.As shown in Fig. 2 picture 201 is to include tank spot image
With the original image of background image, picture 202 is the tank spot image obtained after being pre-processed to original image.The reality of the present invention
Apply in mode, the pretreatment to image, including：Denoising, denoising is carried out using common Denoising Algorithm to picture, is preferably profit
Denoising is carried out to tank spot image with Gauss algorithm.
Preferably, dividing processing is carried out to tank spot image and background image, obtains tank spot image, including：Using RGB and/
Or one or more of hsv color value channel recognition tank spot image border, split initial pictures using threshold segmentation method, obtain
Take tank spot image.Embodiments of the present invention to tank spot image and background image when carrying out dividing processing, including from original graph
Tank spot image is extracted as in, using one or more of color values such as RGB, HSV channel recognition tank spot edge, threshold is then used
It is worth the method segmentation figure picture of segmentation.
Rgb color pattern is a kind of color standard of industrial quarters, is by leading to red (R), green (G), blue (B) three colors
The change in road and their superpositions each other obtain miscellaneous color, and RGB is to represent red, green, blue three
The color of passage, this standard almost includes all colours that human eyesight can perceive, and is at present with most wide color
One of system.
Hsv color model (Hue, Saturation, Value) is to be existed according to the intuitive nature of color by A.R.Smith
A kind of color space created for 1978, also referred to as hexagonal pyramid model (Hexcone Model).The parameter of color in HSV models
It is respectively：Tone (H), saturation degree (S), lightness (V).
Or using the channel B identification tank spot image border in RGB, split initial pictures using threshold segmentation method, obtain
Take tank spot image.The inter-class variance of two classes inside and outside tank spot image border is counted, threshold value when acquisition variance is maximum is used as segmentation threshold
Initial pictures are split by value.Embodiments of the present invention, tank spot edge identification, statistics are carried out by the channel B in RGB
The inter-class variance of two classes inside and outside edge, the threshold value when variance is maximum is segmentation threshold.
Preferably, the original image including tank spot image is pre-processed also includes correction processing, is corrected using algorithm
Because of the form variations of camera angles formation.
Preferably, in step 101：To tank spot image zooming-out characteristic information.
Characteristic information includes：The percentage and positional information of characteristic color in tank spot image, and characteristic area type and
The positional information of characteristic area.For example, characteristic information accounts for the percentage of all colours, and the red position of characteristic color for red
Confidence ceases.
Preferably, the characteristic color in tank spot image is specified, including：It is powder, red, blue or green, purple and white.The color of characteristic color
Value includes the one or more in RGB, HSV or LAB in color value.Preferably, designated color is preferably：Powder, it is red, it is blue or green, it is purple,
In vain, the colour of designated color is a scope, and non-determined value.
Lab colour models are by brightness (L) and a about color, tri- key element compositions of b.L represents brightness
(Luminosity), a represents the scope from carmetta to green, and b represents the scope from yellow to blueness.
Preferably, tank spot image is divided into multiple regions, including following splitting scheme：
The circumscribed rectangle of tank spot image-region is obtained, the region of rectangle is divided into multiple regions；Or
The circumscribed circle of tank spot image-region is obtained, circular region is divided into multiple regions；Or
The circumscribed circle of tank spot image-region is obtained, circular region is divided into multiple regions in radial direction.This hair
Bright embodiment can be split by same splitting scheme, or different splitting schemes is split.Such as Fig. 3
It is shown, it is the circumscribed rectangle splitting scheme of embodiment of the present invention, the region of rectangle is divided into multiple regions.As shown in figure 4,
Will be circumscribed circular by the fan-shaped schematic diagram split for embodiment of the present invention.As shown in figure 5, being embodiment of the present invention
The schematic diagram for being split circumscribed circle by annulus.
Preferably, wherein obtaining the percentage of characteristic color in tank spot image, when positional information includes：By tank spot image point
Multiple regions are cut into, the characteristic color accounting in each region of tank spot image after segmentation is obtained, including：
The percentage of characteristic color accounts for institute for characteristic color in acquired cut zone in tank spot image in characteristic information
There is the percentage of characteristic color, or characteristic color accounts for the percentage of all colours in cut zone in acquired cut zone.
For example, obtaining a certain characteristic color, such as purple accounts for the percentage of all characteristic colors in affiliated cut zone, or obtains a certain spy
Color is levied, such as purple accounts for the percentage of all colours in affiliated cut zone.
Obtain the centroid position in each region of tank spot image after segmentation；Characteristic color hundred in tank spot image in characteristic information
It is the centroid position of the cut zone obtained to divide than corresponding positional information；Or the numbering of cut zone is preset, split
The numbering in region is positional information.For example, a certain characteristic color is obtained, the correspondence position information of the corresponding cut zone of such as purple
For the numbering of the centroid position of violet region, or cut zone, for example numbering 1,2,3 ... 9 are positional information.
Preferably, the location information acquisition method of characteristic area type and characteristic area is：Recognize tank spot color of image
Value belongs to the pixel of default color value scope；By morphological image process, it is determined that characteristic area in each region
Shape.For example, characteristic area can be purple red color spot, petechia, ecchymosis.
The type of characteristic area is determined according to preparatory condition；Preparatory condition includes：Pre-set color scope, default characteristic area
One or more in area and default characteristic area shape information.For example, characteristic area area is more than a preset value, or shape
For rule circular or ellipse, or pre-set color in the range of etc..
For meeting preparatory condition characteristic area, centroid position is calculated；Characteristic area centroid position is characterized region
Positional information.
Preferably, the location information acquisition method of characteristic area and characteristic area is：Tank spot image is divided into multiple
Region, and each region is numbered.For example, being 1,2,3 ... 9 etc. by zone number.
The region after the segmentation where characteristic area centroid position is determined, the numbering of cut zone is positional information.For example
By the numbering A01 of cut zone as positional information.
Preferably, in step 102：Characteristic information is input to the good SVM classifier of training in advance.The embodiment party of the present invention
Formula, be to the method that SVM classifier is trained：First with specifying mobile terminal to gather big measuring tank spot picture photo, extract above
Characteristic information write-in training file, and mark the affiliated tank spot classification of tank spot picture, analyze the probability distribution of sample data with
Just suitable grader is chosen, SVM instrument training environments is built, gaussian kernel function is used by sample data probability distribution, matched somebody with somebody
SVM training tools have been put to be trained according to the sample data put in order.The sample of embodiment of the present invention is constituted：Tank spot figure
Picture, and tank spot image classification.
Preferably, in step 103：The characteristic information of tank spot image is handled by SVM classifier, tank spot figure is obtained
The classification of picture.Embodiments of the present invention, by the way that the characteristic information of tank spot image is input into the good svm classifier of training in advance
Device, using the SVM classifier trained, tank spot classification is carried out to the characteristic information of the tank spot image of input.
The tank spot image classification of embodiment of the present invention is that the characteristic information come out by big measuring tank spot image zooming-out is carried out
The optimal tank spot image classification result that machine learning is obtained, eliminating the influence of subjective factor makes tank spot image classification result more
Precisely so that same class tank spot image can obtain uniformity classification.Fig. 6 be according to one kind of embodiment of the present invention be used for pair
Tank spot characteristics of image carries out processing system structure chart.As shown in fig. 6, system 600 includes：
Photographing module 601, color correction module 602, data processing module 603, data transmission module 604, high in the clouds analysis
Module 605.Photographing module 601, for obtaining tank spot image.
For the color value of tank spot image can be influenceed due to different illumination conditions, subsequent analysis is influenceed, the present invention sets face
Color correction module is corrected to picture color.
Color correction module 602 is made up of default light source and/or default radical occlusion device, and embodiments of the present invention pass through
Default light source can ensure that light source situation is consistent when shooting tank spot image, eliminates ambient light to shooting with default radical occlusion device
Influence.
Color correction module 602 can also be by color with reference to module composition, and photographing module obtains tank spot image and obtained simultaneously
Color is with reference to module map picture.Color can be colour atla, blank sheet of paper etc. with reference to module, and photographing module obtains tank spot image and obtained simultaneously
Color can be corrected with reference to module map picture by specific color object of reference to color.
Data processing module 603, according to the characteristic information of method extractor spot image above.
Data transmission module 604, for the characteristic information of tank spot image to be uploaded into high in the clouds analysis module.
High in the clouds analysis module 605, for the characteristic information received to be input into the good SVM classifier of training in advance, leads to
Cross SVM classifier realization to handle tank spot image, obtain the classification of tank spot image.Embodiments of the present invention, by inciting somebody to action
The characteristic information of tank spot image is input to the good SVM classifier of training in advance, using the SVM classifier trained, to input
The characteristic information of tank spot image carries out tank spot classification.Analysis module is arranged at high in the clouds, can effectively ensure the safety of algorithm
Property, while the accuracy for updating offer system to algorithm can be passed through.
One kind of embodiment of the present invention is used to carry out processing system 600 and embodiment of the present invention to tank spot characteristics of image
One kind be used for tank spot characteristics of image carry out processing method 100 it is corresponding, no longer repeated herein.
The present invention is described by reference to a small amount of embodiment.However, it is known in those skilled in the art, as
What subsidiary Patent right requirement was limited, except the present invention other embodiments disclosed above equally fall the present invention's
In the range of.
Normally, all terms used in the claims are all solved according to them in the usual implication of technical field
Release, unless clearly defined in addition wherein.All references " one/described/be somebody's turn to do [device, component etc.] " are all opened ground
At least one example in described device, component etc. is construed to, unless otherwise expressly specified.Any method disclosed herein
Step need not all be run with disclosed accurate order, unless explicitly stated otherwise.
Claims (10)
1. a kind of method for being handled tank spot characteristics of image, methods described includes：
To tank spot image zooming-out characteristic information, the characteristic information is input to the good SVM classifier of training in advance, passes through SVM
Grader is handled the characteristic information of the tank spot image, obtains the classification of the tank spot image；
The characteristic information includes：The percentage and positional information of characteristic color in tank spot image, and characteristic area type and
The positional information of characteristic area.
2. according to the method described in claim 1, methods described is additionally included in before tank spot image zooming-out characteristic information, to including
The original image of tank spot image is pre-processed, and the pretreatment includes carrying out dividing processing to tank spot image and background image,
Tank spot image is obtained, the pretreatment also includes the one or more in denoising, correction and blank level adjustment processing.
3. according to the method described in claim 1, wherein obtaining the percentage when positional information bag of characteristic color in tank spot image
Include：
The tank spot image is divided into multiple regions；
Obtain the characteristic color accounting in each region of tank spot image after the segmentation：
The percentage of characteristic color accounts for institute for characteristic color in acquired cut zone in tank spot image in the characteristic information
There is the percentage of characteristic color, or characteristic color accounts for the percentage of all colours in cut zone in acquired cut zone；
Obtain the centroid position in each region of tank spot image after the segmentation；Feature face in tank spot image in the characteristic information
The corresponding positional information of color percentage is the centroid position of the cut zone obtained；Or preset the numbering of cut zone,
The numbering of the cut zone is the positional information.
4. according to the method described in claim 1, the location information acquisition method of the characteristic area type and characteristic area
For：
Identification tank spot color of image value belongs to the pixel of default color value scope；
By morphological image process, the shape of the characteristic area in each region is determined；
The type of characteristic area is determined according to preparatory condition；The preparatory condition includes：Pre-set color scope, default characteristic area
One or more in area and default characteristic area shape information；
For meeting preparatory condition characteristic area, centroid position is calculated；
The characteristic area centroid position is the positional information of the characteristic area.
5. method according to claim 4, the location information acquisition method of the characteristic area and characteristic area is：
The tank spot image is divided into multiple regions, and each region is numbered；
The region after the segmentation where the characteristic area centroid position is determined, the numbering of the cut zone is the position letter
Breath.
6. method according to claim 2, described to carry out dividing processing to tank spot image and background image, tank spot figure is obtained
Picture, including：
Using one or more of RGB and/or hsv color value channel recognition tank spot image border, threshold segmentation method is utilized
Split the initial pictures, obtain tank spot image；Or
Using the channel B identification tank spot image border in RGB, split the initial pictures using threshold segmentation method, obtain tank
Spot image.The inter-class variance of two classes inside and outside the tank spot image border is counted, threshold value when acquisition variance is maximum is used as segmentation threshold
The initial pictures are split by value.
7. according to the method in any one of claims 1 to 3, in addition to：The characteristic color in the tank spot image is specified,
Including：It is powder, red, blue or green, purple and white.The color value of the characteristic color include one kind in RGB, HSV or LAB in color value or
It is a variety of.
8. the method according to any one of claim 3 or 4, described that the tank spot image is divided into multiple regions, bag
Include following splitting scheme：
The circumscribed rectangle of the tank spot image-region is obtained, the region of the rectangle is divided into multiple regions；Or
The circumscribed circle of the tank spot image-region is obtained, the circular region is divided into multiple regions；Or
The circumscribed circle of the tank spot image-region is obtained, the circular region is divided into multiple regions in radial direction.
9. the cut zone in method according to claim 8, claim 3 and claim 4 is by same point
What the scheme of cutting was split, or different splitting scheme split.
10. one kind be used for tank spot characteristics of image carry out processing system, the system include photographing module, color correction module,
Data processing module, data transmission module, high in the clouds analysis module；
The photographing module, for obtaining tank spot image；
The color correction module is made up of default light source and/or default radical occlusion device, or by color with reference to module composition, it is described
Photographing module obtains tank spot image and obtains the color simultaneously with reference to module map picture；
The data processing module, extracts the characteristic information of the tank spot image；
The data transmission module, for the characteristic information of the tank spot image to be uploaded into high in the clouds analysis module；
The high in the clouds analysis module, for the characteristic information received to be input into the good SVM classifier of training in advance, leads to
Cross SVM classifier to handle the tank spot image, obtain the classification of the tank spot image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710508178.0A CN107292894B (en) | 2017-06-28 | 2017-06-28 | Method and system for processing pot spot image characteristics |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710508178.0A CN107292894B (en) | 2017-06-28 | 2017-06-28 | Method and system for processing pot spot image characteristics |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107292894A true CN107292894A (en) | 2017-10-24 |
| CN107292894B CN107292894B (en) | 2021-03-09 |
Family
ID=60098844
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710508178.0A Active CN107292894B (en) | 2017-06-28 | 2017-06-28 | Method and system for processing pot spot image characteristics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107292894B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110464308A (en) * | 2019-08-22 | 2019-11-19 | 南京医尔仁医疗科技有限公司 | Intelligent tank print detector and its detection method |
| CN110739047A (en) * | 2018-07-19 | 2020-01-31 | 脉景（杭州）健康管理有限公司 | Patient image color restoration method, device, equipment and readable storage medium |
| CN111797830A (en) * | 2020-07-07 | 2020-10-20 | 因凡科技(北京)有限公司 | Rapid red seal detection method, system and device for bill image |
| CN113436171A (en) * | 2021-06-28 | 2021-09-24 | 博奥生物集团有限公司 | Processing method and device for canned image |
| CN115381690A (en) * | 2021-05-22 | 2022-11-25 | 左点实业（湖北）有限公司 | Scraping and cupping device suitable for home physiotherapy |
| CN115661178A (en) * | 2022-11-17 | 2023-01-31 | 博奥生物集团有限公司 | Method and apparatus for segmenting an imprinted image |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102138860A (en) * | 2011-01-10 | 2011-08-03 | 西安交通大学 | Intelligentized rehabilitation training equipment for hand functions of patients suffering from cerebral injury |
| CN103473550A (en) * | 2013-09-23 | 2013-12-25 | 广州中医药大学 | Plant blade image segmentation method based on Lab space and local area dynamic threshold |
| WO2014116181A2 (en) * | 2013-01-22 | 2014-07-31 | Agency For Science, Technology And Research | Cost-sensitive linear reconstruction based optic cup localization |
| CN106650253A (en) * | 2016-12-16 | 2017-05-10 | 新绎健康科技有限公司 | Dialectical system of traditional Chinese medicine |
| CN106725314A (en) * | 2016-12-05 | 2017-05-31 | 牛欣 | The traditional Chinese medical science based on Quick Response Code, topological code quantifies to distinguish color Diagnostic technique and procedure |
-
2017
- 2017-06-28 CN CN201710508178.0A patent/CN107292894B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102138860A (en) * | 2011-01-10 | 2011-08-03 | 西安交通大学 | Intelligentized rehabilitation training equipment for hand functions of patients suffering from cerebral injury |
| WO2014116181A2 (en) * | 2013-01-22 | 2014-07-31 | Agency For Science, Technology And Research | Cost-sensitive linear reconstruction based optic cup localization |
| CN103473550A (en) * | 2013-09-23 | 2013-12-25 | 广州中医药大学 | Plant blade image segmentation method based on Lab space and local area dynamic threshold |
| CN106725314A (en) * | 2016-12-05 | 2017-05-31 | 牛欣 | The traditional Chinese medical science based on Quick Response Code, topological code quantifies to distinguish color Diagnostic technique and procedure |
| CN106650253A (en) * | 2016-12-16 | 2017-05-10 | 新绎健康科技有限公司 | Dialectical system of traditional Chinese medicine |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110739047A (en) * | 2018-07-19 | 2020-01-31 | 脉景（杭州）健康管理有限公司 | Patient image color restoration method, device, equipment and readable storage medium |
| CN110464308A (en) * | 2019-08-22 | 2019-11-19 | 南京医尔仁医疗科技有限公司 | Intelligent tank print detector and its detection method |
| CN111797830A (en) * | 2020-07-07 | 2020-10-20 | 因凡科技(北京)有限公司 | Rapid red seal detection method, system and device for bill image |
| CN115381690A (en) * | 2021-05-22 | 2022-11-25 | 左点实业（湖北）有限公司 | Scraping and cupping device suitable for home physiotherapy |
| CN113436171A (en) * | 2021-06-28 | 2021-09-24 | 博奥生物集团有限公司 | Processing method and device for canned image |
| CN113436171B (en) * | 2021-06-28 | 2024-02-09 | 博奥生物集团有限公司 | Processing method and device for can printing image |
| CN115661178A (en) * | 2022-11-17 | 2023-01-31 | 博奥生物集团有限公司 | Method and apparatus for segmenting an imprinted image |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107292894B (en) | 2021-03-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107292894A (en) | A kind of method and system for being handled tank spot characteristics of image | |
| Zhou et al. | Color retinal image enhancement based on luminosity and contrast adjustment | |
| Alwazzan et al. | A hybrid algorithm to enhance colour retinal fundus images using a Wiener filter and CLAHE | |
| WO2020151307A1 (en) | Automatic lesion recognition method and device, and computer-readable storage medium | |
| CN103914699B (en) | A kind of method of the image enhaucament of the automatic lip gloss based on color space | |
| Zhao et al. | Image processing based recognition of images with a limited number of pixels using simulated prosthetic vision | |
| CN107590445A (en) | Aesthetic images quality evaluating method based on EEG signals | |
| EP3201826A1 (en) | Method and data-processing device for computer-assisted hair colouring guidance | |
| CN106778785A (en) | Build the method for image characteristics extraction model and method, the device of image recognition | |
| CN109961426A (en) | A kind of detection method of face skin skin quality | |
| CN105913463B (en) | A kind of texture based on location-prior-color characteristic overall situation conspicuousness detection method | |
| CN104392211A (en) | Skin recognition method based on saliency detection | |
| Nayak et al. | Automatic identification of diabetic maculopathy stages using fundus images | |
| CN106821324A (en) | A kind of lingual diagnosis auxiliary medical system based on lingual surface and sublingual comprehensive analysis | |
| CN109242792B (en) | White balance correction method based on white object | |
| Morillas et al. | A design framework to model retinas | |
| Zhang et al. | A fundus image enhancer based on illumination-guided attention and optic disc perception GAN | |
| CN112070737A (en) | Traditional Chinese medicine tongue observation and syndrome differentiation intelligent identification method based on tongue picture image processing | |
| CN106846422A (en) | The recognition methods of antitan agent situation | |
| CN109711306B (en) | Method and equipment for obtaining facial features based on deep convolutional neural network | |
| CN106897711A (en) | Method, the equipment of health status monitoring | |
| Josh et al. | Psychophysics testing of bionic vision image processing algorithms using an FPGA Hatpack | |
| CN110288588A (en) | Retinal images blood vessel segmentation method and system based on gray variance and standard deviation | |
| CN104050455B (en) | A kind of skin color detection method and system | |
| CN110490844A (en) | A kind of recognition methods, system, device and the therapeutic equipment of electromagnetic therapeutic apparatus tank print |
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 |