CN110037651A - The method of quality control and device of eye fundus image - Google Patents
The method of quality control and device of eye fundus image Download PDFInfo
- Publication number
- CN110037651A CN110037651A CN201810036355.4A CN201810036355A CN110037651A CN 110037651 A CN110037651 A CN 110037651A CN 201810036355 A CN201810036355 A CN 201810036355A CN 110037651 A CN110037651 A CN 110037651A
- Authority
- CN
- China
- Prior art keywords
- fundus image
- eye fundus
- eye
- area
- image
- 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
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
- A61B3/0016—Operational features thereof
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
- A61B3/10—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions
- A61B3/14—Arrangements specially adapted for eye photography
Abstract
This application discloses a kind of method of quality control of eye fundus image and devices, and wherein method includes: reception eye fundus image；Calculate the values of ambiguity of the eye fundus image；Blood vessel is extracted from the eye fundus image, determines the avascular area domain in the eye fundus image, and calculates the area in the avascular area domain；According to the values of ambiguity and the area in the avascular area domain, judge whether the picture quality of the eye fundus image is qualified.The application is by comparing the values of ambiguity of eye fundus image and the area in avascular area domain, achieve the purpose that judge whether eye fundus image is qualified, to realize the technical effect for carrying out picture quality control to eye fundus image, and then solves the technical issues of rapidity and stability for the picture quality control for improving eye fundus image.
Description
Technical field
This application involves computer vision and technical field of image processing, in particular to a kind of matter of eye fundus image
Amount control method and device.
Background technique
Due to eye fundus image acquisition equipment all not no discrimination functions of picture quality, this results in a large amount of underproof low-qualitys
Amount eye fundus image enters later period doctor's diagosis link, causes low quality eye fundus image to occupy a large amount of resource, then reduces doctor
The raw analysis efficiency to eye fundus image, directly enhances doctor to the diagosis cost of patient.At the same time, because picture quality not
Good influence judgement, also improves the risk of doctor's erroneous judgement.
Mainly include in the related technology two ways to the control of picture quality: one is fully rely on operator's itself
Experience is differentiated；Another kind is judged by the analysis for carrying out clear zone or dark space to eye fundus image.
Former mode manually carries out judgement and leads to the control standard of picture quality and unstable because fully relying on, and latter
Kind of mode then causes the speed of service slow and unstable because computationally intensive, and function judgement is relatively single, it is difficult to meet actual
Clinical demand.
Summary of the invention
The main purpose of the application is to provide a kind of method of quality control of eye fundus image, with quick, stable judgement
Whether the picture quality of eye fundus image is qualified, avoids low-quality image and enters later period doctor's diagosis link, to receive doctor
Diagosis cost is finally reached the purpose for reducing erroneous judgement risk.
To achieve the goals above, according to the one aspect of the application, a kind of quality controlling party of eye fundus image is provided
Method.
Method of quality control according to the eye fundus image of the application includes:
Receive eye fundus image；
Calculate the values of ambiguity of the eye fundus image；
Blood vessel is extracted from the eye fundus image, determines the avascular area domain in the eye fundus image, and described in calculating
The area in avascular area domain；
According to the values of ambiguity and the area in the avascular area domain, judge the eye fundus image picture quality whether
It is qualified.
Further, described according to the values of ambiguity and the area in the avascular area domain, judge the eye fundus image
Whether picture quality is qualified, comprising:
Judge whether the area in the avascular area domain is greater than area threshold, and judges whether the values of ambiguity is greater than mould
Paste degree threshold value；
If the area in the avascular area domain is greater than area threshold, and the values of ambiguity is greater than the fuzziness threshold
Value, it is determined that the eye fundus image is not conform to table images.
Further, the method also includes:
After determining the eye fundus image not conform to table images, also judge whether the avascular area domain includes reflective clear zone
The dark space and/or；
If the avascular area domain includes reflective clear zone and/or dark space, the eye fundus image exposure has exception.
Further, the method also includes:
After determining the eye fundus image not conform to table images, also judge whether the avascular area domain covers entire eye
Bottom；
If covering entire eyeground, it is determined that the eye fundus image is global blurred picture.
Further, the method also includes:
After determining the eye fundus image for global blurred picture, the eye fundus image of relative users is reacquired, and judge
Whether the eye fundus image of reacquisition is global blurred picture；
If repeatedly judging the eye fundus image reacquired for global blurred picture, it is determined that there are eyes for the eye fundus image
Bottom lesion.
Further, the avascular area domain determined in the eye fundus image, comprising:
After extracting blood vessel in the eye fundus image, screening is carried out to the blood vessel of extraction, the blood vessel that removal is accidentally extracted obtains
To angiosomes；
The avascular area domain in the eye fundus image is determined according to the obtained angiosomes.
Further, the method also includes:
Before extracting blood vessel in the eye fundus image, the colored eye fundus image of eye fundus image progress is judged and preceding
Eye image judgement；
If eye image, rejects the eye fundus image before the eye fundus image is achromaticity eye fundus image or is.
Further, the method also includes:
If the eye fundus image is qualified images, eye is carried out to the eye fundus image and is not analyzed；
It Fen Xi be as a result, not carrying out the analysis of eyeground structure positioning to the eye fundus image according to eye；
Whether the position of the eyeground structure obtained according to positioning judges the eye fundus image centered on the eyeground structure；
If the eye fundus image centered on the eyeground structure, will not have been determined as the eyeground of picture quality qualification
Image is finally determined as that picture quality is unqualified.
To achieve the goals above, according to the another aspect of the application, a kind of quality control dress of eye fundus image is provided
It sets.
Quality control apparatus according to the eye fundus image of the application includes:
Image receiving unit, for receiving eye fundus image；
Fuzzy Calculation unit, for calculating the values of ambiguity of the eye fundus image；
Area extracting unit, for extracting blood vessel from the eye fundus image, determine in the eye fundus image without blood
Area under control domain, and calculate the area in the avascular area domain；
Quality estimation unit judges the eyeground for the area according to the values of ambiguity and the avascular area domain
Whether the picture quality of image is qualified.
Further, the Quality estimation unit includes:
Quality estimation module, for judging whether the area in the avascular area domain is greater than area threshold, and described in judgement
Whether values of ambiguity is greater than fuzziness threshold value；
Quality determination module, if the area for the avascular area domain is greater than area threshold, and the values of ambiguity
Greater than the fuzziness threshold value, it is determined that the eye fundus image is not conform to table images.
Further, described device further include:
Judging unit is exposed, for also judging the avascular area after determining the eye fundus image not conform to table images
Whether domain includes reflective clear zone and/or dark space；
Determination unit is exposed, if including reflective clear zone and/or dark space, the eyeground figure for the avascular area domain
As exposure has exception.
Further, described device further include:
Fuzzy Judgment unit, for also judging the avascular area after determining the eye fundus image not conform to table images
Whether domain covers entire eyeground；
Fuzzy determination unit, if for covering entire eyeground, it is determined that the eye fundus image is global blurred picture.
Further, described device further include:
Judging unit again, for reacquiring relative users after determining the eye fundus image for global blurred picture
Eye fundus image, and judge reacquire eye fundus image whether be global blurred picture；
Quality determination unit again, if the eye fundus image for repeatedly judge to reacquire for overall situation blurred picture, really
There are eyeground pathological changes for the fixed eye fundus image.
Further, the area extracting unit includes:
Blood vessel screening module, for carrying out screening to the blood vessel of extraction, going after extracting blood vessel in the eye fundus image
Except the blood vessel accidentally extracted, angiosomes are obtained；
Avascular area domain module, for according to obtained angiosomes determine in the eye fundus image without blood vessel
Region.
Further, described device further include:
Colored and preceding eye judging unit, for from the eye fundus image extract blood vessel before, to the eye fundus image into
Row colour eye fundus image judgement and the judgement of preceding eye image；
Colored and preceding eye processing unit, if eye image before being achromaticity eye fundus image or be for the eye fundus image,
Then reject the eye fundus image.
Further, described device further include:
The other analytical unit of eye, if being qualified images for the eye fundus image, it is other to carry out eye to the eye fundus image
Analysis；
Structure positioning unit, being used for Fen Xi be as a result, carries out the analysis of eyeground structure positioning to the eye fundus image according to eye；
Whether the position of structure decision unit, the eyeground structure for being obtained according to positioning judges the eye fundus image with institute
Centered on stating eyeground structure；
Determination unit will have determined that quality if being used for the eye fundus image not centered on the eyeground structure again
Finally it is determined as that picture quality is unqualified for the eye fundus image of picture quality qualification.
In the embodiment of the present application, by the way of calculating values of ambiguity and extracting avascular area domain, by comparing eye
The values of ambiguity of base map picture and the area in avascular area domain have achieved the purpose that judge whether eye fundus image is qualified, thus real
Show the technical effect for carrying out picture quality control to eye fundus image, and then solves the picture quality control for improving eye fundus image
Rapidity and stability the technical issues of.
Detailed description of the invention
The attached drawing constituted part of this application is used to provide further understanding of the present application, so that the application's is other
Feature, objects and advantages become more apparent upon.The illustrative examples attached drawing and its explanation of the application is for explaining the application, not
Constitute the improper restriction to the application.In the accompanying drawings:
Fig. 1 is the flow diagram of method of quality control one embodiment of eye fundus image described herein；
Fig. 2 is the process signal that the whether qualified one embodiment concrete operating principle of the eye fundus image is judged in the application
Figure；
Fig. 3 provides the flow diagram of avascular area domain one embodiment for the application；
Fig. 4 is the flow diagram of another embodiment of the method for quality control of herein described eye fundus image；
Fig. 5 is the process signal using method of quality control processing another embodiment of eye fundus image of the eye fundus image
Figure；
Fig. 6 is before filtering out using method of quality control processing another embodiment of eye fundus image of the eye fundus image
The result schematic diagram of eye image；
Fig. 7 is to filter out not using method of quality control processing another embodiment of eye fundus image of the eye fundus image
The result schematic diagram of qualified images；
Fig. 8 is the conjunction filtered out using method of quality control processing another embodiment of eye fundus image of the eye fundus image
The result schematic diagram of table images；
Fig. 9 is the structure of block diagram schematic diagram of quality control apparatus one embodiment of herein described eye fundus image；And
Figure 10 is the structure of block diagram schematic diagram of another embodiment of the quality control apparatus of herein described eye fundus image.
Specific embodiment
In order to make those skilled in the art more fully understand application scheme, below in conjunction in the embodiment of the present application
Attached drawing, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that described embodiment is only
The embodiment of the application a part, instead of all the embodiments.Based on the embodiment in the application, ordinary skill people
Member's every other embodiment obtained without making creative work, all should belong to the model of the application protection
It encloses.
It should be noted that term " includes " and " tool in the description and claims of this application and above-mentioned attached drawing
Have " and their any deformation, it is intended that cover it is non-exclusive include, for example, containing a series of steps or units
Process, method, system, product or equipment those of are not necessarily limited to be clearly listed step or unit, but may include without clear
Other step or units listing to Chu or intrinsic for these process, methods, product or equipment.
It should be noted that in the absence of conflict, the features in the embodiments and the embodiments of the present application can phase
Mutually combination.The application is described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
It is as shown in Figure 1 the flow diagram of method of quality control one embodiment of eye fundus image described herein.
The method includes S101~S104.
S101, eye fundus image is received.
S102, the values of ambiguity for calculating the eye fundus image.
Specifically, same process can be adopted and extract line on the enhancing figure of the eye fundus image, calculate the ladder in line region
Angle value.Specifically, extracting line using gaussian filtering on enhancing figure, lines are screened, and are expanded according to the length of line, gray value
For region；Gauss gradient magnitude is sought on enhancing figure simultaneously, calculates the mean value screened in lines overlay area on amplitude figure；To it
Taking logarithm and being normalized to 0~1 section is indicated as the metric of fog-level with FuzzFactor, wherein 0 indicates most mould
Paste, 1 indicates clearest.
S103, blood vessel is extracted from the eye fundus image, determines the avascular area domain in the eye fundus image, and calculate
The area in the avascular area domain.
S104, according to the values of ambiguity and the area in the avascular area domain, judge the image matter of the eye fundus image
Whether amount is qualified.
The reasons such as the illumination due to resolution ratio, the parameter of itself of the equipment of shooting eye fundus image, and when shooting, cause
There may be obscure, expose the eye fundus image that shooting obtains, it is therefore desirable to carry out image matter to the eye fundus image of shooting
Amount control, judges whether the eye fundus image is qualified.The method of quality control of the eye fundus image provided by the present application is applied to
Eye fundus image analysis field classifies to picture quality before especially carrying out lesion analysis to eye fundus image, to prevent not
Qualified eye fundus image enters lesion detection link, occupies a large amount of resource, to reduce the analysis efficiency to eye fundus image.
Method described herein can be applied in the equipment for carrying out post-processing to the eye fundus image of shooting, can also be answered
With with shooting eye fundus image equipment on.To eye fundus image carry out post-processing equipment include it is a variety of, such as smart phone, PC
Machine, medical treatment analysis meter etc. are classified by picture quality of the postprocessing device to eye fundus image, exclude underproof eyeground
Image carries out detection, guarantees that the eye fundus image for carrying out eyeground pathological changes detection is qualified high quality graphic.Specifically, described
Eye fundus image can be the equipment shooting of post-processing, be also possible to set using the professional shooting such as camera, fundus camera
Standby shooting obtains.Specifically, the equipment for carrying out post-processing to eye fundus image, such as PC machine can handle an eye fundus image,
Multiple eye fundus images can be handled simultaneously.
Fig. 2 is the process signal that the whether qualified one embodiment concrete operating principle of the eye fundus image is judged in the application
Figure.The method includes S201~S202.
S201, judge whether the area in the avascular area domain is greater than area threshold, and whether judge the values of ambiguity
Greater than fuzziness threshold value.
If the area of S202, the avascular area domain are greater than area threshold, and the values of ambiguity is greater than described obscure
Spend threshold value, it is determined that the eye fundus image is not conform to table images.
The purpose of the application is quickly to classify to eye fundus image, qualified eye fundus image is screened, to improve disease
Become analysis efficiency.It is detailed without carrying out eye fundus image as long as therefore judging whether the picture quality of eye fundus image is qualified
Disaggregated classification such as judges whether eye fundus image is bright or dark (because not making much sense to actual demand).Point of eyeground visible vessels
Cloth is extremely important for the judgement of eyeground pathological changes, therefore, when the area in avascular area domain in the eye fundus image is greater than area threshold
When value, show that eyeground visible vessels are considerably less, this eye fundus image is very unfavorable for carrying out lesion detection, and fuzziness is then straight
Connect influence eyeground structure extraction, such as blood vessel, when values of ambiguity be greater than fuzziness threshold value, and avascular area domain area be greater than area
Threshold value directly affects the judgement of late lesions.As long as therefore really according to the calculating of values of ambiguity and avascular area domain area
It is fixed come judge picture quality whether qualification, quickly the picture quality of eye fundus image can not only be judged, and
Manual sort compare for standard is unified, so that classification is more stable, more quickly.
This is arrived, the application has completed the figure to eye fundus image by the judgement of values of ambiguity and the area in avascular area domain
Image quality amount whether He Ge judgement.
In order to improve the picture quality progress exhaustive division to eye fundus image, can whether qualified to completion picture quality
After judgement, the underproof reason of eye fundus image can further be judged, the underproof reason of Accurate Analysis picture quality,
To solve in the related technology to the unicity problem of eye fundus image judging result.In order to solve this problem, the application is to eye
Base map has carried out further analysis as underproof reason.
There are a variety of, such as exposure exceptions for the underproof reason of eye fundus image, therefore in some embodiments, described in determination
Eye fundus image is after not conforming to table images, also to judge whether the avascular area domain includes reflective clear zone and/or dark space；If described
Avascular area domain includes reflective clear zone and/or dark space, then the eye fundus image exposure has exception.The application passes through the embodiment pair
Underproof eye fundus image further judges, judge the picture quality of eye fundus image it is unqualified whether be that there are exposure problems, such as
Fruit is because there are exposure problems, being judged by this to adjust when shooting eye fundus image, caused by reducing because of exposure not
Qualified images generate.
The underproof reason of eye fundus image is such as fuzzy there are a variety of, will directly affect the extraction of eyeground structure, therefore one
In a little embodiments, after determining the eye fundus image not conform to table images, also judge whether the avascular area domain covers entirely
Eyeground；If covering entire eyeground, it is determined that the eye fundus image is global blurred picture.
It should be noted that can both include to unqualified eye in the method for quality control of eye fundus image described herein
The abnormal judgement of the exposure of base map picture, and simultaneously include the fuzzy judgement of the overall situation to unqualified eye fundus image；Or only to two
A part judges in person, i.e., only abnormal to exposure to judge, or only obscures and judge to the overall situation.The application by pair
The further judgement for the reason of not conforming to table images, provides the diversity of judging result, so that user carries out according to judging result
Corresponding operating shoots the probability of unqualified eye fundus image so as to avoid the later period.
After the judgement for completing the unqualified reason of eye fundus image through the foregoing embodiment, since the overall situation obscures the possibility due to
It is that shooting causes, it is also possible to which eyeground pathological changes cause, therefore in some embodiments, are including the embodiment of global fuzzy Judgment
In, further, after determining the eye fundus image for global blurred picture, the eye fundus image of relative users is reacquired, and
Judge whether the eye fundus image reacquired is global blurred picture；If repeatedly judging the eye fundus image reacquired for the overall situation
Blurred picture, it is determined that there are eyeground pathological changes for the eye fundus image.
Fig. 3 provides the flow diagram of avascular area domain one embodiment for the application.
The method includes S301~S302.
S301, after extracting blood vessel in the eye fundus image, screening, the blood that removal is accidentally extracted are carried out to the blood vessel of extraction
Pipe, obtains angiosomes；
The angiosomes that S302, basis obtain determine the avascular area domain in the eye fundus image.
The application carries out screening by the blood vessel extracted, and removal misses the blood vessel extracted, to improve the precision of vessel extraction,
Improve the accuracy that avascular area domain determines.It specifically, can be by calculating the line lengths, right of all extracted blood vessels
Than degree and curvature, when the line length of a certain blood vessel is less than length threshold, contrast is less than contrast threshold, and curvature is small
When curvature threshold value, the blood vessel is determined accidentally to extract blood vessel, is deleted, being discharged caused by non-vascular part influences,
So as to improve avascular area domain area determination accuracy.
Colored eye fundus image has abundant information, therefore the colored eye fundus image that the application is directed to, and is the eye of non-preceding eye
Base map picture, therefore before extracting blood vessel in the eye fundus image, to the eye fundus image carry out colored eye fundus image judge with
And preceding eye image judgement；If eye image, rejects the eye fundus image before the eye fundus image is achromaticity eye fundus image or is.
Specifically, judge whether eye fundus image is that colored eye fundus image can be by judging the image channel number of eye fundus image
Judged judge whether image channel number is 1, if 1 by the eye fundus image be achromaticity eye fundus image.And at the moment
Judgement, then can be by selecting an image channel, with image channel channel B in order to be illustrated, in image channel B centre
In the circle of specified range, fixed threshold (230-255) extracts absolute clear zone, and screens the region of circularity > 0.8.Meet item if having
The region of part is considered eyeball center reflective areas, is determined as preceding eye.It should be noted that colored eye fundus image and preceding eye
The judgement of image is the necessary step of the application, to prevent achromaticity eye fundus image and non-preceding eye image from entering detection,
Reduce the accuracy of later period progress lesion detection.
Fig. 4 is the flow diagram of another embodiment of the method for quality control of herein described eye fundus image.
The method includes S401~S404.
If S401, the eye fundus image are qualified images, eye is carried out to the eye fundus image and is not analyzed, described in analysis
Eye fundus image is left eye or right eye.It specifically, can be according to optic disk and macula lutea under the conditions of optic disk and macular area are all existing
The position in area is judged that optic disk is located on the left of macula lutea, is left eye；Optic disk is located on the right side of macula lutea, is right eye.Or in optic disk or
In the case that person's macula lutea lacks, the shape encircleed according to main blood vessel is needed, judges the possible position of deletion construct, then carries out opposite position
Set differentiation.
S402, according to eye Fen Xi as a result, to the eye fundus image carry out the analysis of eyeground structure positioning, specifically, eyeground
Structure includes optic disk, macular area, main blood vessel arch etc., and the application is numerous to list herein.
S403, judged according to the position of eyeground structure that positioning obtains the eye fundus image whether with the eyeground structure be
Center.
According to geometric position of the optic disk in overall diagram, allowable error is set, determines image whether centered on optic disk.Or
Whether person, according only to the geometric position of macula lutea, can determine image centered on macula lutea in the image for navigating to macular region.
If S404, the eye fundus image not centered on the eyeground structure, will have been determined as picture quality qualification
Eye fundus image be finally determined as that picture quality is unqualified.
The application carries out eye to qualified eye fundus image and does not analyze, carries out eyeground structure positioning according to the result that eye is not analyzed
Afterwards, by judging the eye fundus image whether centered on the eyeground structure of positioning, to the eye fundus image for being determined as qualification
Secondary judgement is carried out, is judged by picture quality twice, the picture quality for improving eye fundus image judges precision.
Fig. 5 show the process of method of quality control processing another embodiment of eye fundus image using the eye fundus image
Schematic diagram.
Input first judges after one group of eye fundus image the eye fundus image, rejects the group eye in the present embodiment
Achromaticity figure in base map picture and with preceding eye image, preceding eye image is as shown in Figure 8.
It rejects achromaticity figure and fuzziness calculating is carried out to remaining eye fundus image with after preceding eye image, while calculating remaining eye
The area in the avascular area domain of base map picture, to be judged according to the area in the values of ambiguity of remaining eye fundus image and avascular area domain
Whether the picture quality of remaining eye fundus image is qualified, and through judging, remaining eye fundus image is classified as shown in Figure 7 by the present embodiment
The low-quality qualified images for not conforming to table images and high quality shown in Fig. 8.
And eye fundus image is classified as it is as shown in Figure 7 do not conform to table images after, the application can also be to not conforming to table images
Judged judge whether underproof reason is that exposure is abnormal again；Or judge whether underproof reason is global mould
Paste.On the basis of being judged as global fuzzy, the fuzzy eye fundus image of the overall situation can also be judged again, so that it is determined that global mould
The reason of paste.
The application avoids the eye fundus image for being judged as qualification if the qualified images in Fig. 8 can also be screened again
Because eye fundus image is not centered on the eyeground structure, to influence the lesion judgement essence in later period in the qualified images of Fig. 8
Degree.
The application analyzes judging result by this again, increases the processing mode of judging result, not only increases
The diversification of judging result provides help for eye fundus image shooting, while judging greatly to meet to the preliminary of eye fundus image
Actual clinical demand.
It should be noted that step shown in the flowchart of the accompanying drawings can be in such as a group of computer-executable instructions
It is executed in computer system, although also, logical order is shown in flow charts, and it in some cases, can be with not
The sequence being same as herein executes shown or described step.
According to embodiments of the present invention, it additionally provides a kind of for implementing the dress of the method for quality control of above-mentioned eye fundus image
It sets, as shown in figure 9, the device includes:
Image receiving unit 10, for receiving eye fundus image；
Fuzzy Calculation unit 20, for calculating the values of ambiguity of the eye fundus image；
Area extracting unit 30 determines the nothing in the eye fundus image for extracting blood vessel from the eye fundus image
Angiosomes, and calculate the area in the avascular area domain；
Quality estimation unit 40 judges the eye for the area according to the values of ambiguity and the avascular area domain
Whether the picture quality of base map picture is qualified.
The quality control apparatus of eye fundus image described herein can be applied in the post facility of processing eye fundus image,
It can be applied in shooting eye fundus image equipment, the control to picture quality completed on capture apparatus.To eye fundus image into
Row postprocessing device can be PC machine.Specifically, the PC machine is connect with fundus camera by cable, and fundus camera has been shot
After eye fundus image, the eye fundus image of shooting is sent to PC machine, picture quality control is carried out to the eye fundus image by PC machine, is sentenced
Whether the eye fundus image that breaks is qualified images.
Further, the Quality estimation unit includes:
Quality estimation module 401 for judging whether the area in the avascular area domain is greater than area threshold, and judges institute
State whether values of ambiguity is greater than fuzziness threshold value；
Quality determination module 402, if the area for the avascular area domain is greater than area threshold, and the fuzziness
Value is greater than the fuzziness threshold value, it is determined that the eye fundus image is not conform to table images.
Further, described device further include:
Judging unit 50 is exposed, for also judging the no blood vessel after determining the eye fundus image not conform to table images
Whether region includes reflective clear zone and/or dark space；
Determination unit 60 is exposed, if including reflective clear zone and/or dark space, the eyeground for the avascular area domain
Image exposure has exception.
Further, described device further include:
Fuzzy Judgment unit 70, for also judging the no blood vessel after determining the eye fundus image not conform to table images
Whether region covers entire eyeground；
Fuzzy determination unit 80, if for covering entire eyeground, it is determined that the eye fundus image is global blurred picture.
It is as shown in Figure 10 the structure of block diagram schematic diagram of herein described another embodiment of device, described device is also wrapped
It includes:
Judging unit 90 again, for after determining the eye fundus image for global blurred picture, reacquisition mutually to be applied
The eye fundus image at family, and judge whether the eye fundus image reacquired is global blurred picture；
Quality determination unit 100 again, if the eye fundus image for repeatedly judge to reacquire for overall situation blurred picture,
Determine that there are eyeground pathological changes for the eye fundus image.
Further, the area extracting unit includes:
Blood vessel screening module, for carrying out screening to the blood vessel of extraction, going after extracting blood vessel in the eye fundus image
Except the blood vessel accidentally extracted, angiosomes are obtained；
Avascular area domain module, for according to obtained angiosomes determine in the eye fundus image without blood vessel
Region.
Further, described device further include:
Colored and preceding eye judging unit, for from the eye fundus image extract blood vessel before, to the eye fundus image into
Row colour eye fundus image judgement and the judgement of preceding eye image；
Colored and preceding eye processing unit, if eye image before being achromaticity eye fundus image or be for the eye fundus image,
Then reject the eye fundus image.
Further, described device further include:
The other analytical unit of eye, if being qualified images for the eye fundus image, it is other to carry out eye to the eye fundus image
Analysis；
Structure positioning unit, being used for Fen Xi be as a result, carries out the analysis of eyeground structure positioning to the eye fundus image according to eye；
Whether the position of structure decision unit, the eyeground structure for being obtained according to positioning judges the eye fundus image with institute
Centered on stating eyeground structure；
Determination unit will have determined that quality if being used for the eye fundus image not centered on the eyeground structure again
Finally it is determined as that picture quality is unqualified for the eye fundus image of picture quality qualification.Specifically, image receiving unit, Fuzzy Calculation
Unit area extraction unit and Quality estimation unit can be applied on eye fundus image capture apparatus, and the eye does not analyze list
Determination unit can then be applied to the eye fundus image progress later period again for member, structure positioning unit, structure decision single domain and quality
In the equipment of processing；Or said units can be applied in the equipment for carrying out post-processing to eye fundus image.
The application judges the eye fundus image twice, for the first time according to the values of ambiguity of eye fundus image and avascular area domain
Area judge, after being judged as qualified images, eyeground structure positioning is carried out further according to the result that eye is not analyzed, thus again
It is whether qualified to eye fundus image to carry out secondary interpretation, improve eye fundus image whether He Ge judgement precision, to improve eyeground figure
As the effect of quality classification.
It can be understood from the foregoing, whether the application is by fuzziness and avascular area domain area qualified to eye fundus image
Judgement not only reduces the calculation amount for carrying out qualified judgement to image, also improves the arithmetic speed of system.The application can not only
Enough to eye fundus image, whether qualification judges, moreover it is possible to whether underproof reason judges, so that user is not according to conforming to
Lattice reason re-starts shooting, by the judgement of this unqualified reason, improves the experience property of user, while in underproof base
The preliminary judgement that lesion can be completed on plinth is led to the judgment mode of lesion by this unqualified reason, greatly reduced
The speed of lesion judgement, is convenient for clinical application.Meanwhile the application judges the use of qualified images twice, facilitates access for link
Eye fundus image feature extraction, further improve the treatment effeciency of eye fundus image.
Obviously, those skilled in the art should be understood that each module of the above invention or each step can be with general
Computing device realize that they can be concentrated on a single computing device, or be distributed in multiple computing devices and formed
Network on, optionally, they can be realized with the program code that computing device can perform, it is thus possible to which they are stored
Be performed by computing device in the storage device, perhaps they are fabricated to each integrated circuit modules or by they
In multiple modules or step be fabricated to single integrated circuit module to realize.In this way, the present invention is not limited to any specific
Hardware and software combines.
The foregoing is merely preferred embodiment of the present application, are not intended to limit this application, for the skill of this field
For art personnel, various changes and changes are possible in this application.Within the spirit and principles of this application, made any to repair
Change, equivalent replacement, improvement etc., should be included within the scope of protection of this application.
Claims (10)
1. a kind of method of quality control of eye fundus image characterized by comprising
Receive eye fundus image；
Calculate the values of ambiguity of the eye fundus image；
Blood vessel is extracted from the eye fundus image, determines the avascular area domain in the eye fundus image, and calculates the no blood
The area in area under control domain；
According to the values of ambiguity and the area in the avascular area domain, judge whether the picture quality of the eye fundus image closes
Lattice.
2. the method according to claim 1, wherein described according to the values of ambiguity and the avascular area domain
Area, judge whether the picture quality of the eye fundus image qualified, comprising:
Judge whether the area in the avascular area domain is greater than area threshold, and judges whether the values of ambiguity is greater than fuzziness
Threshold value；
If the area in the avascular area domain is greater than area threshold, and the values of ambiguity is greater than the fuzziness threshold value, then
Determine that the eye fundus image is not conform to table images.
3. method according to claim 1 or 2, which is characterized in that the method also includes:
Determine the eye fundus image be do not conform to table images after, also judge the avascular area domain whether include reflective clear zone and/
Or dark space；
If the avascular area domain includes reflective clear zone and/or dark space, the eye fundus image exposure has exception.
4. the method according to claim 1, wherein the method also includes:
After determining the eye fundus image not conform to table images, also judge whether the avascular area domain covers entire eyeground；
If covering entire eyeground, it is determined that the eye fundus image is global blurred picture.
5. according to the method described in claim 4, it is characterized in that, the method also includes:
After determining the eye fundus image for global blurred picture, the eye fundus image of relative users is reacquired, and judge again
Whether the eye fundus image of acquisition is global blurred picture；
If repeatedly judging the eye fundus image reacquired for global blurred picture, it is determined that there are retinopathies for the eye fundus image
Become.
6. the method according to claim 1, wherein the avascular area determined in the eye fundus image
Domain, comprising:
After extracting blood vessel in the eye fundus image, screening is carried out to the blood vessel of extraction, the blood vessel that removal is accidentally extracted obtains blood
Area under control domain；
The avascular area domain in the eye fundus image is determined according to the obtained angiosomes.
7. the method according to claim 1, wherein the method also includes:
Before extracting blood vessel in the eye fundus image, colored eye fundus image judgement and preceding eye figure are carried out to the eye fundus image
As judgement；
If eye image, rejects the eye fundus image before the eye fundus image is achromaticity eye fundus image or is.
8. the method according to claim 1, wherein the method also includes:
If the eye fundus image is qualified images, eye is carried out to the eye fundus image and is not analyzed；
It Fen Xi be as a result, not carrying out the analysis of eyeground structure positioning to the eye fundus image according to eye；
Whether the position of the eyeground structure obtained according to positioning judges the eye fundus image centered on the eyeground structure；
If the eye fundus image centered on the eyeground structure, will not have been determined as the eye fundus image of picture quality qualification
Finally it is determined as that picture quality is unqualified.
9. a kind of quality control apparatus of eye fundus image characterized by comprising
Image receiving unit, for receiving eye fundus image；
Fuzzy Calculation unit, for calculating the values of ambiguity of the eye fundus image；
Area extracting unit determines the avascular area in the eye fundus image for extracting blood vessel from the eye fundus image
Domain, and calculate the area in the avascular area domain；
Quality estimation unit judges the eye fundus image for the area according to the values of ambiguity and the avascular area domain
Picture quality it is whether qualified.
10. device according to claim 9, which is characterized in that the Quality estimation unit includes:
Quality estimation module for judging whether the area in the avascular area domain is greater than area threshold, and judges described fuzzy
Whether angle value is greater than fuzziness threshold value；
Quality determination module, if the area for the avascular area domain is greater than area threshold, and the values of ambiguity is greater than
The fuzziness threshold value, it is determined that the eye fundus image is not conform to table images.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810036355.4A CN110037651B (en) | 2018-01-15 | 2018-01-15 | Method and device for controlling quality of fundus image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810036355.4A CN110037651B (en) | 2018-01-15 | 2018-01-15 | Method and device for controlling quality of fundus image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110037651A true CN110037651A (en) | 2019-07-23 |
| CN110037651B CN110037651B (en) | 2022-03-25 |
Family
ID=67273340
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810036355.4A Active CN110037651B (en) | 2018-01-15 | 2018-01-15 | Method and device for controlling quality of fundus image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110037651B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110428410A (en) * | 2019-07-31 | 2019-11-08 | 腾讯医疗健康(深圳)有限公司 | Eyeground medical image processing method, device, equipment and storage medium |
| CN110619273A (en) * | 2019-08-14 | 2019-12-27 | 中山市奥珀金属制品有限公司 | Efficient iris recognition method and device |
| CN112508898A (en) * | 2020-11-30 | 2021-03-16 | 北京百度网讯科技有限公司 | Method and device for detecting fundus image and electronic equipment |
| CN113744254A (en) * | 2021-09-08 | 2021-12-03 | 中山大学中山眼科中心 | Fundus image analysis method and system, storage medium and computer equipment |
Citations (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007007044A (en) * | 2005-06-29 | 2007-01-18 | Canon Inc | Ophtalmic photographing apparatus |
| CN102479385A (en) * | 2010-11-19 | 2012-05-30 | 佳能株式会社 | Image processing apparatus and image processing method |
| US20120249959A1 (en) * | 2011-03-31 | 2012-10-04 | The Hong Kong Polytechnic University | Apparatus and method for non-invasive diabetic retinopathy detection and monitoring |
| CN102908120A (en) * | 2012-10-09 | 2013-02-06 | 北京大恒图像视觉有限公司 | Eye fundus image registration method, eye fundus image optic disk nerve and vessel measuring method and eye fundus image matching method |
| CN103810492A (en) * | 2014-01-17 | 2014-05-21 | 北京大恒图像视觉有限公司 | Ambiguity analytical method of eye fundus image |
| CN204698510U (en) * | 2015-06-02 | 2015-10-14 | 福州大学 | The diabetic retinopathy optical fundus examination photographic means that picture quality ensures |
| CN105395163A (en) * | 2014-09-05 | 2016-03-16 | 佳能株式会社 | Ophthalmologic Apparatus And Ophthalmologic Apparatus Control Method |
| CN105411525A (en) * | 2015-11-10 | 2016-03-23 | 广州河谷互动医疗科技有限公司 | Eye ground picture and image intelligent obtaining and recognizing system |
| CN105787928A (en) * | 2016-02-14 | 2016-07-20 | 浙江大学 | Fuzzy fundus image automatic detection and screening method based on visual fuzziness |
| CN106446805A (en) * | 2016-09-08 | 2017-02-22 | 北京化工大学 | Segmentation method and system for optic cup in eye ground photo |
| CN106530295A (en) * | 2016-11-07 | 2017-03-22 | 首都医科大学 | Fundus image classification method and device of retinopathy |
| CN106650596A (en) * | 2016-10-10 | 2017-05-10 | 北京新皓然软件技术有限责任公司 | Fundus image analysis method, device and system |
| CN106651888A (en) * | 2016-09-28 | 2017-05-10 | 天津工业大学 | Color fundus image optic cup segmentation method based on multi-feature fusion |
| CN107045720A (en) * | 2017-05-04 | 2017-08-15 | 深圳硅基智能科技有限公司 | Artificial neural network and system for recognizing eye fundus image lesion |
| CN107123124A (en) * | 2017-05-04 | 2017-09-01 | 季鑫 | Retina image analysis method and device and computing equipment |
| CN107133932A (en) * | 2017-05-04 | 2017-09-05 | 季鑫 | Retina image preprocessing method and device and computing equipment |
| CN107209933A (en) * | 2014-08-25 | 2017-09-26 | 新加坡科技研究局 | For assessing retinal images and the method and system of information being obtained from retinal images |
| CN107358612A (en) * | 2017-07-07 | 2017-11-17 | 东北大学 | A kind of retinal vessel segmenting system combined based on fractal dimension with gaussian filtering and method |
| CN107451998A (en) * | 2017-08-08 | 2017-12-08 | 北京大恒普信医疗技术有限公司 | A kind of eye fundus image method of quality control |
| CN104899876B (en) * | 2015-05-18 | 2018-04-06 | 天津工业大学 | A kind of eye fundus image blood vessel segmentation method based on adaptive Gauss difference |
| CN108272434A (en) * | 2017-12-07 | 2018-07-13 | 江威 | The method and device that eye fundus image is handled |
-
2018
- 2018-01-15 CN CN201810036355.4A patent/CN110037651B/en active Active
Patent Citations (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007007044A (en) * | 2005-06-29 | 2007-01-18 | Canon Inc | Ophtalmic photographing apparatus |
| CN102479385A (en) * | 2010-11-19 | 2012-05-30 | 佳能株式会社 | Image processing apparatus and image processing method |
| US20120249959A1 (en) * | 2011-03-31 | 2012-10-04 | The Hong Kong Polytechnic University | Apparatus and method for non-invasive diabetic retinopathy detection and monitoring |
| CN102908120A (en) * | 2012-10-09 | 2013-02-06 | 北京大恒图像视觉有限公司 | Eye fundus image registration method, eye fundus image optic disk nerve and vessel measuring method and eye fundus image matching method |
| CN103810492A (en) * | 2014-01-17 | 2014-05-21 | 北京大恒图像视觉有限公司 | Ambiguity analytical method of eye fundus image |
| CN107209933A (en) * | 2014-08-25 | 2017-09-26 | 新加坡科技研究局 | For assessing retinal images and the method and system of information being obtained from retinal images |
| CN105395163A (en) * | 2014-09-05 | 2016-03-16 | 佳能株式会社 | Ophthalmologic Apparatus And Ophthalmologic Apparatus Control Method |
| CN104899876B (en) * | 2015-05-18 | 2018-04-06 | 天津工业大学 | A kind of eye fundus image blood vessel segmentation method based on adaptive Gauss difference |
| CN204698510U (en) * | 2015-06-02 | 2015-10-14 | 福州大学 | The diabetic retinopathy optical fundus examination photographic means that picture quality ensures |
| CN105411525A (en) * | 2015-11-10 | 2016-03-23 | 广州河谷互动医疗科技有限公司 | Eye ground picture and image intelligent obtaining and recognizing system |
| CN105787928A (en) * | 2016-02-14 | 2016-07-20 | 浙江大学 | Fuzzy fundus image automatic detection and screening method based on visual fuzziness |
| CN106446805A (en) * | 2016-09-08 | 2017-02-22 | 北京化工大学 | Segmentation method and system for optic cup in eye ground photo |
| CN106651888A (en) * | 2016-09-28 | 2017-05-10 | 天津工业大学 | Color fundus image optic cup segmentation method based on multi-feature fusion |
| CN106650596A (en) * | 2016-10-10 | 2017-05-10 | 北京新皓然软件技术有限责任公司 | Fundus image analysis method, device and system |
| CN106530295A (en) * | 2016-11-07 | 2017-03-22 | 首都医科大学 | Fundus image classification method and device of retinopathy |
| CN107133932A (en) * | 2017-05-04 | 2017-09-05 | 季鑫 | Retina image preprocessing method and device and computing equipment |
| CN107123124A (en) * | 2017-05-04 | 2017-09-01 | 季鑫 | Retina image analysis method and device and computing equipment |
| CN107045720A (en) * | 2017-05-04 | 2017-08-15 | 深圳硅基智能科技有限公司 | Artificial neural network and system for recognizing eye fundus image lesion |
| CN107358612A (en) * | 2017-07-07 | 2017-11-17 | 东北大学 | A kind of retinal vessel segmenting system combined based on fractal dimension with gaussian filtering and method |
| CN107451998A (en) * | 2017-08-08 | 2017-12-08 | 北京大恒普信医疗技术有限公司 | A kind of eye fundus image method of quality control |
| CN108272434A (en) * | 2017-12-07 | 2018-07-13 | 江威 | The method and device that eye fundus image is handled |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110428410A (en) * | 2019-07-31 | 2019-11-08 | 腾讯医疗健康(深圳)有限公司 | Eyeground medical image processing method, device, equipment and storage medium |
| CN110428410B (en) * | 2019-07-31 | 2024-02-27 | 腾讯医疗健康(深圳)有限公司 | Fundus medical image processing method, device, equipment and storage medium |
| CN110619273A (en) * | 2019-08-14 | 2019-12-27 | 中山市奥珀金属制品有限公司 | Efficient iris recognition method and device |
| CN110619273B (en) * | 2019-08-14 | 2023-10-31 | 张杰辉 | Efficient iris recognition method and recognition device |
| CN112508898A (en) * | 2020-11-30 | 2021-03-16 | 北京百度网讯科技有限公司 | Method and device for detecting fundus image and electronic equipment |
| CN113744254A (en) * | 2021-09-08 | 2021-12-03 | 中山大学中山眼科中心 | Fundus image analysis method and system, storage medium and computer equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110037651B (en) | 2022-03-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110037651A (en) | The method of quality control and device of eye fundus image | |
| CN107451998B (en) | Fundus image quality control method | |
| CN106022209B (en) | A kind of method and device of range estimation and processing based on Face datection | |
| US20140285768A1 (en) | System and Method for Identifying Eye Conditions | |
| CN109684981B (en) | Identification method and equipment of cyan eye image and screening system | |
| EP2881891B1 (en) | Image processing device and image processing method | |
| CN103942539B (en) | A kind of oval accurate high efficiency extraction of head part and masking method for detecting human face | |
| CN111932535A (en) | Method, apparatus, device and storage medium for processing image | |
| CN104021382A (en) | Eye image collection method and system | |
| CN109583364A (en) | Image-recognizing method and equipment | |
| CN104574285A (en) | Method for automatically removing image black eyes | |
| EP1779320A1 (en) | Image processing apparatus and method, image sensig apparatus, and program | |
| KR20180009303A (en) | Method and apparatus for iris recognition | |
| CN110533654A (en) | The method for detecting abnormality and device of components | |
| CN108629378A (en) | Image-recognizing method and equipment | |
| Harms et al. | Segmentation of stained blood cell images measured at high scanning density with high magnification and high numerical aperture optics | |
| CN112115886A (en) | Image detection method and related device, equipment and storage medium | |
| CN110399822A (en) | Action identification method of raising one's hand, device and storage medium based on deep learning | |
| CN112489060A (en) | System and method for pneumonia focus segmentation | |
| Reddy et al. | A robust scheme for iris segmentation in mobile environment | |
| CN113536267A (en) | Human body gait verification method based on artificial intelligence and cloud server | |
| CN111311565A (en) | Eye OCT image-based detection method and device for positioning points of optic cups and optic discs | |
| CN111951214B (en) | Method and device for dividing readable area in image, electronic equipment and storage medium | |
| CN103577791B (en) | A kind of red-eye detecting method and system | |
| CN115829980A (en) | Image recognition method, device, equipment and storage medium for fundus picture |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220412 Address after: Collective of Shangdi branch of Haidian District Talent Service Center, No. 19, Shangdi Information Road, Haidian District, Beijing 100089 Patentee after: Ke Xin Address before: 432507 group 4, Huoju village, xiaxindian Town, Yunmeng County, Xiaogan City, Hubei Province Patentee before: Jiang Wei |
|
| TR01 | Transfer of patent right |