CN110503639A - The method and apparatus for handling eye fundus image - Google Patents
The method and apparatus for handling eye fundus image Download PDFInfo
- Publication number
- CN110503639A CN110503639A CN201910754948.9A CN201910754948A CN110503639A CN 110503639 A CN110503639 A CN 110503639A CN 201910754948 A CN201910754948 A CN 201910754948A CN 110503639 A CN110503639 A CN 110503639A
- Authority
- CN
- China
- Prior art keywords
- eye fundus
- effective coverage
- fundus image
- image
- gray
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 86
- 238000000605 extraction Methods 0.000 claims abstract description 19
- 230000000694 effects Effects 0.000 claims abstract description 12
- 230000003902 lesion Effects 0.000 claims description 37
- 230000008569 process Effects 0.000 claims description 34
- 238000012545 processing Methods 0.000 claims description 19
- 239000000284 extract Substances 0.000 claims description 11
- 238000000691 measurement method Methods 0.000 claims description 3
- 201000010099 disease Diseases 0.000 claims description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims description 2
- 210000001508 eye Anatomy 0.000 description 174
- 210000004204 blood vessel Anatomy 0.000 description 16
- 238000012216 screening Methods 0.000 description 14
- 238000004422 calculation algorithm Methods 0.000 description 12
- 230000009466 transformation Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 4
- 210000002189 macula lutea Anatomy 0.000 description 4
- 210000003733 optic disk Anatomy 0.000 description 4
- 238000007781 pre-processing Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000000717 retained effect Effects 0.000 description 3
- 238000007619 statistical method Methods 0.000 description 3
- 239000000654 additive Substances 0.000 description 2
- 230000000996 additive effect Effects 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 210000001525 retina Anatomy 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 210000005252 bulbus oculi Anatomy 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 208000030533 eye disease Diseases 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 230000001976 improved effect Effects 0.000 description 1
- 230000035800 maturation Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 231100000915 pathological change Toxicity 0.000 description 1
- 230000036285 pathological change Effects 0.000 description 1
- 230000035479 physiological effects, processes and functions Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000035807 sensation Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 230000002311 subsequent effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30041—Eye; Retina; Ophthalmic
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Eye Examination Apparatus (AREA)
- Image Analysis (AREA)
Abstract
The present invention provides a kind of method and apparatus for handling eye fundus image, which comprises determines the effective coverage of the eye fundus image;The essential information of the eye fundus image is obtained according to the effective coverage;And before carrying out feature extraction to the eye fundus image, the eye fundus image is pre-processed based on the essential information, realize the essential information that eye fundus image is obtained in effective coverage, basic reference information is provided for the pretreatment of eye fundus image, to improve effect when the subsequent progress feature extraction to eye fundus image.
Description
Technical field
The present invention relates to technical field of image processing, and in particular to a kind of method and apparatus for handling eye fundus image.
Background technique
In recent years by Computer Automatic Recognition eye fundus image carry out auxiliary diagnosis technology be developed, by observation and
Identify eye fundus image it can be found that patient body many illnesss.The shooting of eye fundus image, not only by the shadow of fundus camera type
Ring, while also being influenced by shooting environmental, different types of fundus camera for same eyeground shoot come eye fundus image
Information be not quite similar;The information for the eye fundus image that same type fundus camera is shot under various circumstances is not quite similar.These letters
The otherness of breath greatly interferes the extraction of late feature.Therefore before being extracted to eye fundus image feature, it is necessary first to
Eye fundus image is pre-processed, to achieve the purpose that denoising, information normalization and enhancing.
But existing preprocess method is still unsatisfactory to the treatment effect of ophthalmoscopic image at present.
Summary of the invention
In view of this, the embodiment of the present invention is dedicated to providing a kind of method and apparatus for handling eye fundus image, can be improved
Effect when the subsequent progress feature extraction to eye fundus image.
According to a first aspect of the embodiments of the present invention, a kind of method for handling eye fundus image is provided, comprising: determine the eye
The effective coverage of base map picture;The essential information of the eye fundus image is obtained according to the effective coverage;And to the eyeground
Before image carries out feature extraction, the eye fundus image is pre-processed based on the essential information.
In one embodiment, the effective coverage of the determination eye fundus image, comprising: determine the eye fundus image
First gray threshold;And according to first gray threshold, the pixel of the eye fundus image is screened described in acquisition
Effective coverage.
In one embodiment, described according to first gray threshold, the pixel of the eye fundus image is screened
It include: to screen, obtain to the pixel of the eye fundus image according to first gray threshold to obtain the effective coverage
Initial virtual zone;Calculate the average gray and standard deviation of the eye fundus image;According to the average gray and standard deviation,
Determine the second gray value threshold value of the eye fundus image;And according to the second gray value threshold value, to the initial effective district
Domain is further screened, and the effective coverage is obtained.
In one embodiment, the essential information includes the effective radius and effective area of the effective coverage, described
The essential information of the eye fundus image is obtained according to the effective coverage, comprising: determine the minimum circumscribed circle of the effective coverage;
Pixel number shared by the radius of the minimum circumscribed circle is calculated, using the pixel number as the size of the effective radius;And
The effective area is calculated according to the effective radius.
In one embodiment, the essential information includes average gray, is used for the conduct in subsequent image treatment process
Calculate the underlying parameter value of contrast, the essential information that the eye fundus image is obtained according to the effective coverage, comprising: will
The effective coverage of the eye fundus image is converted to gray level image, using the average gray of the gray level image as the effective district
The average gray in domain.
In one embodiment, the essential information includes average tone, is used for the conduct in subsequent image treatment process
Calculate the underlying parameter value of contrast, the essential information that the eye fundus image is obtained according to the effective coverage, comprising: will
The eye fundus image is converted to HIS color space by RGB color;And have described in being calculated in the HIS color space
The average gray for imitating the H band in region, using the average gray of the H band as the average tone of the effective coverage.
In one embodiment, the essential information includes noise, for deciding whether during subsequent image denoises
It needs to carry out denoising, select which kind of denoising method and in subsequent image treatment process as the basis for calculating contrast
Parameter value, the essential information that the eye fundus image is obtained according to the effective coverage, comprising: extract the effective coverage
Edge feature is to form edge feature image;Gray threshold based on the edge feature image extracts in the effective coverage
Noise region;The area of the noise region is calculated, and the effective district is judged according to the size of the area of the noise region
It whether there is noise in domain;And when judging in the effective coverage to utilize Uniform measurement method to assess institute there are when noise
State the noise size in noise region.
In one embodiment, the essential information includes clarity, is commented for being used as in subsequent image treatment process
Estimate the parameter of the eye fundus image quality, the essential information that the eye fundus image is obtained according to the effective coverage, comprising:
Extract the line feature of the effective coverage;Determine the resolution of the line feature;And according to the resolution of the line feature come
Assess the clarity of the effective coverage.
In one embodiment, the essential information includes maximum lesion size, in subsequent image treatment process
As the threshold value for assessing the eye fundus image lesion, the basic letter that the eye fundus image is obtained according to the effective coverage
Breath, comprising: extract maximum light tone region or the maximum dark areas of the effective coverage;And calculate the effective coverage most
Pixel number shared by big light tone region or maximum dark areas, using the pixel number as the maximum lesion size.
In one embodiment, the essential information includes dimension information, grayscale information, the tone letter of the effective coverage
At least one of breath, noise information, sharpness information and lesion size information.
In one embodiment, the dimension information of the effective coverage includes effective area and/or effective radius, the ash
Degree information includes the average gray of the effective coverage, and the hue information includes the average tone of the effective coverage, described
Noise information includes the noise of the effective coverage, and the sharpness information includes the clarity of the effective coverage, the disease
Become the maximum lesion size that dimension information includes the effective coverage.
According to a second aspect of the embodiments of the present invention, a kind of device for handling eye fundus image is provided, comprising: effective coverage is true
Cover half block is configured to determine the effective coverage of the eye fundus image;Essential information obtains module, is configured to according to the effective district
Domain obtains the essential information of the eye fundus image;And preprocessing module, it is configured to mention to eye fundus image progress feature
Before taking, the eye fundus image is pre-processed based on the essential information.
According to a third aspect of the embodiments of the present invention, a kind of computer readable storage medium is provided, calculating is stored thereon with
Machine executable instruction realizes the method for processing eye fundus image as described above when the executable instruction is executed by processor.
According to a fourth aspect of the embodiments of the present invention, a kind of device for handling eye fundus image is provided, including computer-readable
Storage medium and processor are stored with computer executable instructions on the computer readable storage medium, and the computer can
Execute instruction the method that processing eye fundus image as described above is realized when being executed by the processor.
A kind of method for handling eye fundus image, passes through having for the determination eye fundus image provided by the embodiment of the present invention
Imitate region;The essential information of the eye fundus image is obtained according to the effective coverage;And spy is being carried out to the eye fundus image
Before sign is extracted, the eye fundus image is pre-processed based on the essential information, realizes and obtains eye in effective coverage
The essential information of base map picture provides basic reference information for the pretreatment of eye fundus image, to improve subsequent to eyeground
Image carries out effect when feature extraction.
Detailed description of the invention
Fig. 1 show the flow diagram of the method for processing eye fundus image provided by one embodiment of the present invention.
Fig. 2 show the block diagram of the device of processing eye fundus image provided by one embodiment of the present invention.
Fig. 3 show the block diagram of the device of the processing eye fundus image of another embodiment of the present invention offer.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that the described embodiment is only a part of the embodiment of the present invention, instead of all the embodiments.Based on this
Embodiment in invention, every other reality obtained by those of ordinary skill in the art without making creative efforts
Example is applied, shall fall within the protection scope of the present invention.
Herein, term " first ", " second " and other similar words are not meant to imply any sequence, quantity and again
The property wanted, but be used only for distinguishing different elements.Herein, term " one ", "one" is with other similar words
It is not intended to mean that and only exists the things, but indicate in relation to describing only for one in the things, it is described
Things may have one or more.Herein, term "comprising", " comprising " are intended to mean that in logic with other similar words
Correlation, and the relationship that cannot be regarded as in representation space structure.For example, " A includes B " is intended to mean that logically B belongs to
A, without indicating that spatially B is located at the inside of A.In addition, term "comprising", " comprising " should be regarded with the meaning of other similar words
To be open, and misclosure.For example, " A includes B " is intended to mean that B belongs to A, but B not necessarily constitutes the whole of A, A
It is also possible that other elements such as C, D, E.
Herein, term " embodiment ", " the present embodiment ", " embodiment ", " one embodiment " are not offered as related
Description is suitable only for a specific embodiment, but indicates that these descriptions are also likely to be applicable to another or multiple implementations
In example.It will be understood by those skilled in the art that herein, any description done for some embodiment can in addition
Related description in one or more embodiments is substituted, combined or is combined in other ways, it is described substitution, combination or
It is that those skilled in the art can be readily apparent that person combines generated new embodiment in other ways, belongs to guarantor of the invention
Protect range.
The current calculating for eye fundus image essential information obtains the standard for having no a set of maturation, and there are no mature calculating sides
Method, thus in the past before being pre-processed to image it is rare consider image itself essential information or to image essential information
Consider less, while which essential information pre-processed subsequent image in the past useful, and needed to obtain, and how to obtain etc. and ask
Topic is not resolved, so that eye fundus image be made to lack basic reference information during pretreated, leads to the later period
The extraction process of eyeground feature is more difficult and extraction effect is unsatisfactory.
Eye fundus image, which refers to, shoots the obtained image in eyeball bottom by general or special purpose capture apparatus, display
Content generally includes the physiological structures such as retina, blood vessel, optic disk, macula lutea.Eye fundus image can be analog image, be also possible to count
Word image.In various embodiments of the present invention, eye fundus image is preferably digital picture.
Fig. 1 show the flow diagram of the method for processing eye fundus image provided by one embodiment of the present invention.Such as Fig. 1,
This method comprises:
S101: the effective coverage of the eye fundus image is determined.
It should be appreciated that the effective coverage of eye fundus image refers to the area-of-interest other than background area, region of interest
Domain includes the physiological structures such as blood vessel, optic disk, macula lutea.Because when the eyeground to everyone is shot, eye fundus image
Effective coverage is not necessarily identical, therefore the screening of effective coverage will be carried out respectively for everyone eye fundus image.Meanwhile it is first
First determine that the effective coverage of eye fundus image can also make the acquisition of the subsequent essential information to eye fundus image become more square
Just quick.
S102: the essential information of the eye fundus image is obtained according to the effective coverage.
Specifically, it is necessary first to the effective coverage of eye fundus image is determined by step S101, it is right in effective coverage
Eye fundus image carries out relevant calculating and assessment, to obtain the essential information of eye fundus image for locating in advance in subsequent image
As basic reference information during reason.It should be noted that the essential information of eye fundus image can have very much, obtaining
When each essential information, the embodiment of the present invention is not intended to limit the sequencing that each essential information obtains.
S103: before carrying out feature extraction to the eye fundus image, based on the essential information to the eye fundus image
It is pre-processed.
In conclusion eye fundus image is pre-processed first before carrying out feature extraction to eye fundus image, and to eye
It is exactly to obtain the essential information of eye fundus image that base map picture, which carries out pretreated most basic step,.Eye is got in effective coverage
The essential information of base map picture provides basic reference information for the pretreatment of eye fundus image, to improve subsequent to eyeground figure
As carrying out effect when feature extraction.
In another embodiment of the present invention, the effective coverage of the determination eye fundus image, comprising: described in determining
First gray threshold of eye fundus image;And according to first gray threshold, the pixel of the eye fundus image is screened
To obtain the effective coverage.
It should be appreciated that being color image by the eye fundus image that shooting obtains, which includes tri- waves of R, G, B
Section can obtain tri- wave bands of H, I and S when color image is after HIS space converts, when color image is empty by LAB color
Between convert after, tri- wave bands of L, A and B can be obtained, other spatial alternations are also may also pass through certainly and obtain other color spaces
Wave band.A gray values of any one above-described wave band in eye fundus image are chosen as the first gray threshold,
Gray values change in 0 to 255 range, and first gray threshold can be the empirical value obtained by statistical analysis,
It can be and be calculated by algorithm, the embodiment of the present invention is not construed as limiting the value of the first gray threshold specifically taken.It needs
It is noted that the embodiment of the present invention is not intended to limit which kind of wave band any one wave band is specially, any one wave band can be
A kind of wave band (appointing i.e. in the wave bands such as R wave band, G-band, B wave band, H band, I wave band, S-band, L-band, A band, B wave band
Meaning one), or combination (i.e. R wave band, G-band, B wave band, H band, I wave band, S-band, the L-band, A of several wave bands
Any combination of wave band, B wave band).
Because the area-of-interest of the eye fundus image and the gray value of background area be it is different, from the point of view of visually, sense
The brightness in interest region is brighter (being in light tone), and the brightness ratio of background area is darker (in dead color), so first gray threshold
Setting mainly aim at divide eye fundus image area-of-interest and background area.It should be noted that when be determined the
When one gray threshold, the embodiment of the present invention is not intended to limit the specific embodiment of the eye fundus image screening process, to eye fundus image
Gray value be less than the pixel of first gray threshold and be removed, which is greater than to the gray value of eye fundus image
Pixel retained, then the pixel remained is exactly the effective coverage to be obtained, or, to eye fundus image
The pixel that gray value is greater than first gray threshold is removed, and is less than first gray threshold to the gray value of eye fundus image
Pixel is retained, then the pixel remained is exactly the effective coverage to be obtained.
In another embodiment of the present invention, described according to first gray threshold, to the picture of the eye fundus image
Element screened with obtain the effective coverage include: according to first gray threshold, to the pixel of the eye fundus image into
Row screening, obtains initial virtual zone;Calculate the average gray and standard deviation of the eye fundus image;It is average according to the gray scale
Value and standard deviation, determine the second gray value threshold value of the eye fundus image;And according to the second gray value threshold value, to described
Initial virtual zone is further screened, and the effective coverage is obtained.
It should be appreciated that screen according to pixel of first gray threshold to the eye fundus image obtained effective
The effective coverage of the not necessarily final eye fundus image to be obtained in region, this is possible to only be merely a preliminary screening
Process has substantially determined the range (referred to herein as initial virtual zone) of the effective coverage of the eye fundus image.It is right also to need again
The initial virtual zone is further screened, to obtain the final effective coverage to be obtained.
Specifically, the pixel of the eye fundus image is screened according to first gray threshold initial effective to obtain
Region, then, calculating the average gray of any one wave band of the initial virtual zone of eye fundus image, (any one wave band exists
Had been described in upper one embodiment, details are not described herein), standard deviation, the average gray are calculated further according to average gray
It is grey when solving second according to average gray and standard deviation with the important parameter that standard deviation is as the second gray threshold of solution
After spending threshold value, the initial virtual zone is further screened, obtains the effective district of the final eye fundus image to be obtained
Domain.It should be noted that after solving the second gray threshold, the embodiment of the present invention be not intended to limit this to initial virtual zone into
The specific embodiment of the further screening process of row can be to be greater than second ash to the gray value within the scope of initial virtual zone
The pixel of degree threshold value is removed, and the pixel for being less than first gray threshold to the gray value within the scope of initial virtual zone carries out
Retain, then the pixel remained is exactly the final effective coverage to be obtained, or, to initial virtual zone range
The pixel that interior gray value is less than second gray threshold is removed, and being greater than to the gray value within the scope of initial virtual zone should
The pixel of first gray threshold is retained, then the pixel remained is exactly the final effective coverage to be obtained.
In another embodiment of the present invention, the area threshold for determining the effective coverage of the eye fundus image, when
By according to first gray threshold, the initial effective district obtained after being screened to the pixel of the eye fundus image
When the area in domain is greater than the area threshold, it is also necessary to carry out recited above further screened to the initial virtual zone
Journey;Or determine the area threshold of the effective coverage of the eye fundus image, when by the picture to the eye fundus image
When the area of element initial virtual zone obtained after being screened is less than the area threshold, it is also necessary to described initial
Effective coverage carries out further screening process recited above.
It should be appreciated that the area of the initial virtual zone after above-mentioned preliminary screening process may obtain than final institute
Effective coverage area it is big, it is also possible to the area of the effective coverage obtained than final is small, so in order to obtain final institute
The effective coverage to be obtained will further screen the initial virtual zone Jing Guo preliminary screening.It needs to illustrate
It is that the determination of the area threshold can be the empirical value obtained by statistical analysis, is also possible to be calculated by algorithm
Area value.
It is also understood that can also whether continuous to determine whether carrying out institute above by observing the initial virtual zone
The further screening process stated can carry out further screening process when the initial virtual zone is discontinuous to obtain
Final effective coverage.There may be the time of shooting and the personal information of patient in the initial virtual zone of eye fundus image,
It is not connect when the personal information of the time and patient that there is shooting in initial virtual zone just represent entire initial virtual zone
Continuous, need to carry out further screening process recited above.
In another embodiment of the present invention, the essential information include the effective coverage effective radius and effectively
Area, the essential information that the eye fundus image is obtained according to the effective coverage, comprising: determine the effective coverage most
Small circumscribed circle;Pixel number shared by the radius of the minimum circumscribed circle is calculated, using the pixel number as the effective radius
Size;And the effective area is calculated according to the effective radius.
It should be appreciated that the essential information includes the effective radius and effective area of the effective coverage, the effective radius
It is used to characterize the information of the effective coverage with effective area.It should be noted that the side of determining minimum circumscribed circle can be passed through
Method determines the effective radius, by calculate pixel number shared by the radius of the minimum circumscribed circle can obtain it is described effectively
The size of radius, but the embodiment of the present invention is not intended to limit the determination method of the effective radius, can also first pass through algorithm calculating
The effective area (for example, calculating effective area using pixel number) of effective coverage out, is calculated effectively further according to effective area
Radius, other methods that can determine effective radius are also suitable for the application.
It should be noted that the embodiment of the present invention is not intended to limit the minimum circumscribed circle it is confirmed that the effective coverage,
The minimum circumscribed rectangle of available effective coverage, minimum external oval, maximum circumscribed circle, maximum boundary rectangle, maximum are external
Ellipse etc., the present invention is to this and with no restrictions.
In another embodiment of the present invention, the essential information includes average gray, in subsequent image processing
In the process as the underlying parameter value for calculating contrast, the basic letter that the eye fundus image is obtained according to the effective coverage
Breath, comprising: the effective coverage of the eye fundus image is converted into gray level image, using the average gray of the gray level image as
The average gray of the effective coverage.
It should be appreciated that being color image by the eye fundus image that shooting obtains, chromatic image is being converted into gray level image
Afterwards, tri- spectrum conversions of R, G, B of the color image at gray level image a wave band, so determining the eye fundus image
The color image of the effective coverage of eye fundus image first can be converted into gray level image before the average gray of effective coverage, with this
Average gray of the average gray of the only one wave band of gray level image as the effective coverage.
It is also understood that average gray is for the underlying parameter in subsequent image treatment process as calculating contrast
Value, meanwhile, average gray can also provide underlying parameter for subsequent image compensation and subsequent image enhancement, to assess figure
As the size of compensation or the degree of image enhancement.
In another embodiment of the present invention, the essential information includes average tone, in subsequent image processing
In the process as the underlying parameter value for calculating contrast, the basic letter that the eye fundus image is obtained according to the effective coverage
Breath, comprising: the eye fundus image is converted into HIS color space by RGB color;And in the HIS color space
The average gray for calculating the H band of the effective coverage, using the average gray of the H band as the effective coverage
Average tone.
It should be appreciated that HIS (Hue-Intensity-Saturation) is in color space with tone, brightness and saturation
The color mode for spending to indicate.The relational model between RGB color and HIS color space and the mutual transformation carried out
Treatment process be known as HIS transformation.HIS transformation is also referred to as color transformation or Meng Saier transformation.In eye fundus image processing usually
Application there are two types of hue coordinate system or color space: first is that be made of RGB three primary colors color space (RGB coordinate system or
Person's RGB color);Another is by tone (hue), saturation degree (saturation) and brightness (intensity) three
The color space (HIS coordinate system or HIS color space) that a variable is constituted.That is a kind of color can use RGB color
R, G and B in space are described, and can also be described with I, H and S of HIS color space, the former is from physics angle
Color is described, the latter is then to describe color from the subjective sensation of human eye.HIS transformation is exactly RGB color and HIS color
Transformation between space.Therefore, tone just refers to the gray value on the H band in the HIS color space of eye fundus image, then having
The average tone in effect region can be averaged by doing average calculating operation to all gray values in the effective coverage on H band
The obtained average gray of operation is the average tone of effective coverage.
It is also understood that average color is called in the underlying parameter in subsequent image treatment process as calculating contrast
Value, meanwhile, average gray can also provide underlying parameter for subsequent image compensation and subsequent image enhancement, to assess figure
As the size of compensation or the degree of image enhancement.
It should be noted that average gray and average color tune combination can be used in subsequent image treatment process
The middle underlying parameter value as calculating contrast, meanwhile, average gray and average color tune combination can also be subsequent
Image compensation and subsequent image enhancement provide underlying parameter, to assess the size of image compensation or the degree of image enhancement.
In another embodiment of the present invention, the essential information includes noise, for denoising process in subsequent image
In decide whether carry out denoising, select which kind of denoising method and in subsequent image treatment process as calculating
The underlying parameter value of contrast, the essential information that the eye fundus image is obtained according to the effective coverage, comprising: extract institute
The edge feature of effective coverage is stated to form edge feature image;Described in gray threshold based on the edge feature image extracts
Noise region in effective coverage;Calculating proposes the area for stating noise region, and the size of the area according to the noise region
Judge in the effective coverage with the presence or absence of noise;And when judging to utilize uniformity there are when noise in the effective coverage
Measure assesses the noise size in the noise region.
It should be appreciated that edge is the discontinuous place of characteristic in eye fundus image (such as pixel grey scale, texture) distribution, it is image
Surrounding characteristic has those of Spline smoothing or ridge variation pixel set.Another definition of image border refers to picture around it
The set of plain those discontinuous pixels of grey scale change.Noise is assessed by edge feature can be good at making an uproar at random to some
Sound and additive noise are assessed, therefore may determine that in the effective coverage according to the size of the area of the noise region
With the presence or absence of random noise and additive noise.Since the contrast of edge feature and the contrast of noise spot are than in effective coverage
Muting pixel contrast it is high, so edge feature is assessed to noise easily to be evaluated noise spot
Come.
Specifically, multiple edge features in effective coverage are extracted first, constitute edge feature figure by multiple edge features
Picture obtains the gray threshold of the edge feature image by related algorithm or statistical analysis, according to the gray scale of edge feature image
Threshold value assesses the noise spot in effective coverage, can be determined as to the pixel for the gray threshold for being greater than the edge feature image
The noise spot of effective coverage can also be determined as effective coverage to the pixel for the gray threshold for being less than the edge feature image
Noise spot, the embodiment of the present invention is to this and without limitation.After determining the noise spot in effective coverage, by all noise spot structures
At the noise region of effective coverage, and the calculating of area is carried out to the noise region, i.e., calculated using the pixel number of noise region
The area of noise region.Judge whether the size of the area of the noise region constitutes the noise of the effective coverage, one can be set
A noise area threshold value can determine that the noise region is constituted and be somebody's turn to do when the area of noise region is greater than the noise area threshold value
The noise of effective coverage can determine that the noise region is not enough to when the area of noise region is less than the noise area threshold value
The noise for constituting effective coverage, can be ignored the noise region.When the size to noise area is enough to constitute effective coverage
Noise region after, the noise size in the noise region is assessed using Uniform measurement method.
It should be appreciated that a gray values of any one wave band in edge feature image can be chosen as edge spy
The gray threshold of image is levied, wherein any one wave band can be a kind of wave band (i.e. R wave band, G-band, B wave band, H band, I wave
Any one in the wave bands such as section, S-band, L-band, A band, B wave band), or the combinations of several wave bands (i.e. R wave band,
G-band, B wave band, H band, I wave band, S-band, L-band, A band, any combination of B wave band).Since noise region is by very
What multiple noise spots were constituted, thus using uniformity metric quantity algorithm be the size of each noise spot in noise region is carried out it is flat
It calculates, finally obtains the average noise size of the effective coverage, commonly referred to herein as noise size, but need to illustrate
It is that the embodiment of the present invention is to the method for how calculating noise size and is not specifically limited, other can calculate noise size
Method be also suitable for the application.
It is also understood that the assessment of noise carries out at denoising for deciding whether during subsequent image denoises
It manages, select which kind of denoising method, as the underlying parameter value for calculating contrast in subsequent image treatment process and subsequent
As the threshold value for assessing the eye fundus image noise during image denoising.Specifically, when utilization noise area threshold value is to effective
It when the area of the noise region in region compares, can determine whether the noise region constitutes the noise of effective coverage, change
Sentence is talked about, in that case it can be decided that whether needs to carry out denoising, it, can when the noise region constitutes the noise of effective coverage
Carry out denoising;After the noise size of noise region determines, process can be denoised for subsequent image by the noise size
It provides noise threshold (the noise size for calculating acquisition can be used as noise threshold), when some pixel of eye fundus image is made an uproar
It when sound value is greater than the noise threshold, determines that the pixel is noise spot, subsequent denoising can be carried out to the pixel.
When noise difference, the method for calculating contrast is not identical, the size of the method and compensation numerical value that compensate to contrast
It is not identical, so after noise has been determined, so that it may determine to calculate the method for contrast, the method for compensating contrast
And the size etc. of compensation numerical value.
It should be noted why the embodiment of the present invention provides using edge feature the side for assessing eye fundus image noise
Method, be because present inventor verified through a large number of experiments its assessment effect it is preferable, but the present invention implement
Example be not intended to limit assessment eye fundus image noise method, can also using based on Shearlet frame multiple dimensioned Denoising Algorithm,
Method based on the multiple dimensioned Denoising Algorithm etc. of Ridgelet transformation, other assessment eye fundus image noises is also suitable for the application.
In another embodiment of the present invention, the essential information includes clarity, for processed in subsequent image
It is described that the basic of the eye fundus image is obtained according to the effective coverage as the parameter for assessing the eye fundus image quality in journey
Information, comprising: extract the line feature of the effective coverage;Determine the resolution of the line feature;And according to the line feature
Resolution assess the clarity of the effective coverage.
It should be appreciated that the essential information further includes clarity, for being used as assessment institute in subsequent image treatment process
State the threshold value of eye fundus image clarity.Since the line feature of eye fundus image is mainly blood vessel, the resolution of blood vessel is higher, then represents
Eye fundus image is more clear.Specifically, clarity can be assessed according to the total length for precomputing blood vessel, is obtained then calculating
Blood vessel length it is longer, then the clarity for representing eye fundus image is higher.Specifically, when the eye fundus image that we are observed visually
Clarity it is very high, it is relatively good for the extraction effect of blood vessel to represent selected algorithm, can be extracted by the algorithm thinner
Small blood vessel, therefore the length of vessel calculated is also longer, is based on a large amount of data statistic analysis, can obtain a blood vessel
Length threshold, i.e., when the length of vessel in extracted eye fundus image is lower than the length of vessel threshold value, then it is assumed that fogging image.
Therefore, when the length of vessel in the effective coverage for the eye fundus image being calculated is greater than the length of vessel threshold value, described in determination
The clarity of eye fundus image is relatively high (picture quality is high), the length of vessel in the effective coverage for the eye fundus image being calculated
When less than the length of vessel threshold value, it is relatively low (picture quality is low) to determine the clarity of the eye fundus image, then can weigh
Newly shot to obtain the high eye fundus image of clarity.
It should be appreciated that the clarity of image can refer to naked eyes or computer to the size of the identification capacity of image detail.
For Computer Automatic Recognition and the clarity of the eye fundus image of diagnosis can refer to computer to the details of eye fundus image (such as
Blood vessel) ability or accuracy that are read out and identify.It should be noted that choosing line feature to identify the main original of clarity
Because being that assessment result is objective, the structure of blood vessel is conducive to reference standard unification when being unified in assessment, and blood vessel refers to and divides on retina
The blood vessel of cloth is display content generally existing in eye fundus image.Meanwhile why the embodiment of the present invention provides and utilizes line feature
Resolution come the method for assessing image definition, be because present inventor has verified it through a large number of experiments and commented
The effect estimated is preferable, but the method that the embodiment of the present invention is not intended to limit assessment image definition, can also pass through other methods
Assess the clarity of image.For example, can be according to the identification of other schematic structures (such as optic disk, macula lutea etc.) in addition to blood vessel
Degree can also calculate the clear of image according to resolution ratio, contrast, other image parameters or combinations thereof to calculate clarity
Degree.
It is also understood that there are certain to be associated with for the parameters such as the concept of image definition and the resolution ratio of image, contrast.
In some cases, the resolution ratio or contrast of image are higher, and the clarity of image may be higher.However, above-mentioned rule is simultaneously
It is non-to be generally applicable in.Clarity is as a kind of standard for measuring picture quality, independently of other picture qualities such as resolution ratio, contrast
Parameter.It determines that clarity can refer to the clarity for calculating image by exact algorithm, can refer to big by fuzzy algorithmic approach
Cause calculates the clarity of image, can also refer to the clarity that image is substantially estimated according to statistics or empirical data.Specifically
Ground can also determine clarity according to the contrast of blood vessel, can refer to the numerical value for directlying adopt calculated contrast
As the numerical value of clarity, it can refer to the numerical value using the numerical value of the contrast after mathematic(al) manipulation as clarity, it can be with
Refer to using contrast as a parameter, the number of its clarity is calculated together with preparatory or temporarily determining other parameters
Value.
In another embodiment of the present invention, the essential information includes maximum lesion size, in subsequent image
It is described that the eye fundus image is obtained according to the effective coverage as the threshold value for assessing the eye fundus image lesion in treatment process
Essential information, comprising: extract the effective coverage maximum light tone region or maximum dark areas;And it calculates described effective
Pixel number shared by the maximum light tone region in region or maximum dark areas, using the pixel number as the maximum lesion ruler
It is very little.
It should be appreciated that the essential information further includes maximum lesion size, maximum lesion size is used at subsequent image
As the threshold value for assessing the eye fundus image lesion during reason, at the same time it can also be provided to be subsequent when carrying out lesion extraction
One screening criteria, or the identification of subsequent image feature and classification provide parameter support.
It is also understood that eyeground pathological changes are divided into light tone lesion and dark-coloured lesion, the effect that light tone lesion is showed be
The significantly partially bright region of brightness in effective coverage, the effect that dark-coloured lesion is showed is that the brightness in effective coverage is significant
Partially dark region.So by extracting maximum light tone region or maximum dark areas, and calculate the most light of the effective coverage
Pixel number shared by color region or maximum dark areas is finally maximum light tone region or maximum dark-coloured area with the pixel number
The maximum lesion size in domain.It should be noted that can be according to the pixel number for calculating maximum light tone region or maximum dark areas
Obtain maximum lesion size, but the embodiment of the present invention is not intended to limit the method for obtaining maximum lesion size, other methods are also fitted
For the application.
It should be noted that can be the disease of eye fundus image described in further evaluation by calculating the maximum lesion size obtained
Become size and lesion threshold value (the maximum lesion size for calculating acquisition can be used as lesion threshold value) is provided, when some of eye fundus image
When the size in the region of brightness exception is less than the lesion threshold value, determine that the region is lesion region, when some of eye fundus image is bright
When spending the size in abnormal region greater than the lesion threshold value, determine that the region is not lesion region, then at this time can be again
Adjustment algorithm finds the position of lesion region or makees preliminary screening to the position of lesion region.
In another embodiment of the present invention, the essential information includes the dimension information of the effective coverage, gray scale
At least one of information, hue information, noise information, sharpness information and lesion size information.
In another embodiment of the present invention, the dimension information of the effective coverage include effective area and/or effectively
Radius, the grayscale information include the average gray of the effective coverage, and the hue information includes the flat of the effective coverage
Equal tone, the noise information include the noise of the effective coverage, and the sharpness information includes the clear of the effective coverage
Clear degree, the lesion size information include the maximum lesion size of the effective coverage.
It should be noted that the embodiment of the present invention is not intended to limit the type and number of the essential information of effective coverage, remove
Include other than above-mentioned essential information, can also include other essential informations, can be subsequent eye fundus image
The essential information that pretreatment provides basic parameter information is all contained in the protection scope of the application.
Fig. 2 show the block diagram of the device of processing eye fundus image provided by one embodiment of the present invention.As shown in Fig. 2, should
Device includes: effective coverage determining module 210, is configured to determine the effective coverage of the eye fundus image;Essential information obtains mould
Block 220 is configured to obtain the essential information of the eye fundus image according to the effective coverage;Preprocessing module 230, is configured to
Before carrying out feature extraction to the eye fundus image, the eye fundus image is pre-processed based on the essential information.
It should be appreciated that effective coverage determining module 210 is used to determine the effective coverage of the eye fundus image, eye fundus image
Effective coverage refers to that the area-of-interest other than background area, area-of-interest include the physiology knot such as blood vessel, optic disk, macula lutea
Structure.Because when the eyeground to everyone is shot, the effective coverage of eye fundus image is not necessarily identical, therefore will needle
Carry out the screening of effective coverage respectively to everyone eye fundus image.Simultaneously, it is first determined go out the effective coverage of eye fundus image also
The acquisition of the subsequent essential information to eye fundus image can be made to become enhanced convenience quick.
It is used to obtain the eye fundus image according to the effective coverage it is also understood that essential information obtains module 220
Essential information is obtained after determining the effective coverage of eye fundus image by effective coverage determining module 210 by essential information
Module 220 carries out relevant calculating and assessment to eye fundus image in effective coverage, to obtain the essential information of eye fundus image
Using for during subsequent image preprocessing as basic reference information.
It is also understood that preprocessing module 230 is used for before carrying out feature extraction to the eye fundus image, based on described
Essential information pre-processes the eye fundus image, before carrying out feature extraction to eye fundus image, passes through pretreatment first
Module 230 pre-processes eye fundus image.And carrying out pretreated most basic step to eye fundus image is exactly to obtain eyeground
The essential information of image provides then getting the essential information of eye fundus image in effective coverage for the pretreatment of eye fundus image
Basic reference information, to improve subsequent effect when carrying out feature extraction to eye fundus image.
It should be noted that the essential information of eye fundus image can have very much, and when obtaining each essential information, the present invention
Embodiment is not intended to limit the sequencing that each essential information obtains.
Fig. 3 show the block diagram of the device of the processing eye fundus image of another embodiment of the present invention offer.Referring to Fig. 3, dress
Setting 300 includes processing component 310, further comprises one or more processors, and the storage as representated by memory 320
Device resource, can be by the instruction of the execution of processing component 310, such as application program for storing.The application stored in memory 320
Program may include it is one or more each correspond to one group of instruction module.In addition, processing component 310 is configured
Method to execute instruction, to execute above-mentioned processing eye fundus image.
Device 300 can also include that a power supply module be configured as the power management of executive device 300, one it is wired or
Radio network interface is configured as device 300 being connected to network and input and output (I/O) interface.Device 300 can be grasped
Make based on the operating system for being stored in memory 320, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM,
FreeBSDTMOr it is similar.
A kind of non-transitorycomputer readable storage medium, when the instruction in storage medium is by the processing of above-mentioned apparatus 300
When device executes, so that above-mentioned apparatus 300 is able to carry out a kind of method for handling eye fundus image, comprising: determine the eye fundus image
Effective coverage;The essential information of the eye fundus image is obtained according to the effective coverage;And to the eye fundus image into
Before row feature extraction, the eye fundus image is pre-processed based on the essential information.
Those of ordinary skill in the art may be aware that list described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and algorithm steps can be realized with the combination of electronic hardware or computer software and electronic hardware.These functions are actually
It is implemented in hardware or software, the specific application and design constraint depending on technical solution.Professional technician
Each specific application can be used different methods to achieve the described function, but this realization is it is not considered that exceed
The scope of the present invention.
It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description,
The specific work process of device and unit, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
In several embodiments provided herein, it should be understood that disclosed systems, devices and methods, it can be with
It realizes by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the unit
It divides, only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units or components
It can be combined or can be integrated into another system, or some features can be ignored or not executed.Another point, it is shown or
The mutual coupling, direct-coupling or communication connection discussed can be through some interfaces, the indirect coupling of device or unit
It closes or communicates to connect, can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.
It, can be with if the function is realized in the form of SFU software functional unit and when sold or used as an independent product
It is stored in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, server or network equipment etc.) it performs all or part of the steps of the method described in the various embodiments of the present invention.
And storage medium above-mentioned includes: that USB flash disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), arbitrary access are deposited
The various media that can store program ver-ify code such as reservoir (Random Access Memory, RAM), magnetic or disk.
In addition, it should also be noted that, institute in the combination of each technical characteristic and unlimited this case claim in this case
Combination documented by the combination or specific embodiment of record, all technical characteristics documented by this case can be to appoint
Where formula is freely combined or is combined, unless generating contradiction between each other.
It should be noted that the above list is only specific embodiments of the present invention, it is clear that the present invention is not limited to above real
Example is applied, there are many similar variations therewith.If those skilled in the art directly exported from present disclosure or
All deformations associated, are within the scope of protection of the invention.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Within mind and principle, made any modification, equivalent replacement etc. be should all be included in the protection scope of the present invention.
Claims (10)
1. a kind of method for handling eye fundus image characterized by comprising
Determine the effective coverage of the eye fundus image;
The essential information of the eye fundus image is obtained according to the effective coverage;And
Before carrying out feature extraction to the eye fundus image, the eye fundus image is located in advance based on the essential information
Reason.
2. the method according to claim 1, wherein the effective coverage of the determination eye fundus image, comprising:
Determine the first gray threshold of the eye fundus image;And
According to first gray threshold, the pixel of the eye fundus image is screened to obtain the effective coverage.
3. according to the method described in claim 2, it is characterized in that, described according to first gray threshold, to the eyeground
The pixel of image is screened to obtain the effective coverage and include:
According to first gray threshold, the pixel of the eye fundus image is screened, initial virtual zone is obtained;
Calculate the average gray and standard deviation of the eye fundus image;
According to the average gray and standard deviation, the second gray value threshold value of the eye fundus image is determined;And
According to the second gray value threshold value, the initial virtual zone is further screened, obtains the effective coverage,
Wherein, the essential information include the dimension information of the effective coverage, it is grayscale information, hue information, noise information, clear
At least one of clear degree information and lesion size information, the dimension information of the effective coverage include effective area and/or
Effective radius, the grayscale information include the average gray of the effective coverage, and the hue information includes the effective coverage
Average tone, the noise information includes the noise of the effective coverage, and the sharpness information includes the effective coverage
Clarity, the lesion size information includes the maximum lesion size of the effective coverage.
4. method according to any one of claims 1 to 3, which is characterized in that the essential information has including described
Imitate the effective radius and effective area in region, the essential information that the eye fundus image is obtained according to the effective coverage, packet
It includes:
Determine the minimum circumscribed circle of the effective coverage;
Pixel number shared by the radius of the minimum circumscribed circle is calculated, using the pixel number as the size of the effective radius;
And
The effective area is calculated according to the effective radius.
5. method according to any one of claims 1 to 3, which is characterized in that the essential information includes average ash
Degree, it is described to be obtained according to the effective coverage for the underlying parameter value in subsequent image treatment process as calculating contrast
Take the essential information of the eye fundus image, comprising:
The effective coverage of the eye fundus image is converted into gray level image, using the average gray of the gray level image described in
The average gray of effective coverage.
6. method according to any one of claims 1 to 3, which is characterized in that the essential information includes average color
It adjusts, it is described to be obtained according to the effective coverage for the underlying parameter value in subsequent image treatment process as calculating contrast
Take the essential information of the eye fundus image, comprising:
The eye fundus image is converted into HIS color space by RGB color;And
The average gray of the H band of the effective coverage is calculated, in the HIS color space with the gray scale of the H band
Average tone of the average value as the effective coverage.
7. method according to any one of claims 1 to 3, which is characterized in that the essential information includes noise, is used
In decided whether during subsequent image denoises carry out denoising, select which kind of denoising method and in subsequent image
As the underlying parameter value for calculating contrast, the base that the eye fundus image is obtained according to the effective coverage in treatment process
This information, comprising:
The edge feature of the effective coverage is extracted to form edge feature image;
Gray threshold based on the edge feature image extracts the noise region in the effective coverage;
Calculate the area of the noise region, and judged according to the size of the area of the noise region be in the effective coverage
It is no that there are noises;And
When judging to assess the noise in the noise region using Uniform measurement method there are when noise in the effective coverage
Size.
8. method according to any one of claims 1 to 3, which is characterized in that the essential information includes clarity,
It is described to be obtained according to the effective coverage for the parameter in subsequent image treatment process as the assessment eye fundus image quality
Take the essential information of the eye fundus image, comprising:
Extract the line feature of the effective coverage;
Determine the resolution of the line feature;And
The clarity of the effective coverage is assessed according to the resolution of the line feature.
9. method according to any one of claims 1 to 3, which is characterized in that the essential information includes most serious disease
Become size, it is described to have according to for the threshold value in subsequent image treatment process as the assessment eye fundus image lesion
Effect region obtains the essential information of the eye fundus image, comprising:
Extract maximum light tone region or the maximum dark areas of the effective coverage;And
Calculate the effective coverage maximum light tone region or maximum dark areas shared by pixel number, using the pixel number as
The maximum lesion size.
10. a kind of device for handling eye fundus image, including computer readable storage medium and processor are described computer-readable to deposit
Computer executable instructions are stored on storage media, which is characterized in that the computer executable instructions are held by the processor
The method of the processing eye fundus image as described in any one in claim 1 to 9 is realized when row.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910754948.9A CN110503639A (en) | 2019-08-15 | 2019-08-15 | The method and apparatus for handling eye fundus image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910754948.9A CN110503639A (en) | 2019-08-15 | 2019-08-15 | The method and apparatus for handling eye fundus image |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110503639A true CN110503639A (en) | 2019-11-26 |
Family
ID=68586613
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910754948.9A Pending CN110503639A (en) | 2019-08-15 | 2019-08-15 | The method and apparatus for handling eye fundus image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110503639A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112734701A (en) * | 2020-12-28 | 2021-04-30 | 深圳市优必选科技股份有限公司 | Fundus focus detection method, fundus focus detection device and terminal equipment |
CN112907548A (en) * | 2021-02-26 | 2021-06-04 | 依未科技(北京)有限公司 | Image evaluation method and device, computer-readable storage medium and electronic device |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102096818A (en) * | 2011-01-31 | 2011-06-15 | 中国林业科学研究院资源信息研究所 | Object-oriented automatic extracting method and system for outline and parameter of remote sensing image crown |
CN108272434A (en) * | 2017-12-07 | 2018-07-13 | 江威 | The method and device that eye fundus image is handled |
CN108629769A (en) * | 2018-05-02 | 2018-10-09 | 山东师范大学 | Eye fundus image optic disk localization method and system based on best fraternal similarity |
WO2019077494A1 (en) * | 2017-10-16 | 2019-04-25 | King Abdullah University Of Science And Technology | System, apparatus, and method for sequence-based enzyme ec number prediction by deep learning |
-
2019
- 2019-08-15 CN CN201910754948.9A patent/CN110503639A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102096818A (en) * | 2011-01-31 | 2011-06-15 | 中国林业科学研究院资源信息研究所 | Object-oriented automatic extracting method and system for outline and parameter of remote sensing image crown |
WO2019077494A1 (en) * | 2017-10-16 | 2019-04-25 | King Abdullah University Of Science And Technology | System, apparatus, and method for sequence-based enzyme ec number prediction by deep learning |
CN108272434A (en) * | 2017-12-07 | 2018-07-13 | 江威 | The method and device that eye fundus image is handled |
CN108629769A (en) * | 2018-05-02 | 2018-10-09 | 山东师范大学 | Eye fundus image optic disk localization method and system based on best fraternal similarity |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112734701A (en) * | 2020-12-28 | 2021-04-30 | 深圳市优必选科技股份有限公司 | Fundus focus detection method, fundus focus detection device and terminal equipment |
CN112907548A (en) * | 2021-02-26 | 2021-06-04 | 依未科技(北京)有限公司 | Image evaluation method and device, computer-readable storage medium and electronic device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7248736B2 (en) | Enhancing images superimposed on uneven or partially obscured background | |
CN108272434B (en) | Method and device for processing fundus images | |
JP5426475B2 (en) | Skin color unevenness analysis apparatus, skin color unevenness analysis method, and skin color unevenness analysis program | |
US20080304736A1 (en) | Method of estimating a visual evaluation value of skin beauty | |
CN109993731A (en) | A kind of eyeground pathological changes analysis method and device | |
Jan et al. | Retinal image analysis aimed at blood vessel tree segmentation and early detection of neural-layer deterioration | |
CN110503639A (en) | The method and apparatus for handling eye fundus image | |
CN113610844A (en) | Intelligent skin care method, device, equipment and storage medium | |
Hussein et al. | Retinex theory for color image enhancement: A systematic review | |
Kundu et al. | Visual attention guided quality assessment of tone-mapped images using scene statistics | |
JP2009082338A (en) | Skin discrimination method using entropy | |
JP6740329B2 (en) | Method and system for facial feature analysis and delivery of personalized advice | |
CN106446805B (en) | A kind of eyeground shine in optic cup dividing method and system | |
JP6530703B2 (en) | Skin condition evaluation method | |
CN110675325A (en) | Method and device for enhancing fundus image | |
JP7234086B2 (en) | Biometric information acquisition device and program | |
Kauppi et al. | Simple and robust optic disc localisation using colour decorrelated templates | |
EP3048580A1 (en) | Method of extracting the optic disc of a retinal image | |
CN115984206A (en) | Image classification method and device and electronic equipment | |
JP2011118655A (en) | Apparatus and method for processing image | |
US20210319551A1 (en) | 3d analysis with optical coherence tomography images | |
Troglio et al. | Unsupervised change-detection in retinal images by a multiple-classifier approach | |
Yang et al. | Multi-Scale histogram tone mapping algorithm enables better object detection in wide dynamic range images | |
Relan et al. | Robustness of Fourier fractal analysis in differentiating subgroups of retinal images | |
CN108830847A (en) | Visible light is objectively evaluated with infrared grayscale fusion image perceptual contrast |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191126 |