CN109447964A - Method for processing fundus images and equipment - Google Patents
Method for processing fundus images and equipment Download PDFInfo
- Publication number
- CN109447964A CN109447964A CN201811236773.4A CN201811236773A CN109447964A CN 109447964 A CN109447964 A CN 109447964A CN 201811236773 A CN201811236773 A CN 201811236773A CN 109447964 A CN109447964 A CN 109447964A
- Authority
- CN
- China
- Prior art keywords
- pixel value
- mapping
- image
- region
- pixel
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- 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)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Processing (AREA)
Abstract
The present invention provides a kind of method for processing fundus images and equipment, and wherein described image processing method includes that original image is divided into multiple regions；The first mapping pixel value of each pixel is obtained to the pixel value progress Nonlinear Processing of penetrating of each pixel in described each region respectively, the Nonlinear Processing makes first of each pixel in the region to map near the median of maximum value and minimum value of the pixel value integrated distribution in the pixel value in the region；Linear process is carried out to the first mapping pixel value of each pixel in each region respectively and obtains the second mapping pixel value of each pixel；Treated image is generated using the second mapping pixel value of each pixel in described each region.
Description
Technical field
The present invention relates to field of image processings, and in particular to a kind of method for processing fundus images and equipment.
Background technique
During shooting image, the problems such as due to personnel, environment or equipment, depth of exposure difference will lead to, and
The content that different exposures can show image, which is brought, to be directly affected.
At present specific objective, especially in the medical field, eyeground figure can be identified according to picture material in many fields
It seem the important evidence for judging eye disease, the quality of eye fundus image can bring direct influence to doctor and machine diagosis.
Such as over-exposed eye fundus image will affect judgement of the doctor to optic disk and cup area, and this area
Judgment bias will lead to the mistaken diagnosis of disease such as glaucoma.Because the area of optic cup optic disk is the color according to eye fundus image optic disk
Always judge with walking for blood vessel, the calculating that excessive exposure frequently can lead to cup area is excessive.And under-exposure can be led
It causes image smudgy, can not identify specific content from image by naked eyes.
Existing technical solution typically directly enhances brightness of image or reduces brightness of image, and this processing mode can make
It is too high or too low and can not show actual content that the suitable position of script brightness becomes brightness, it can be seen that prior art
It is poor to the recovery effects of image.
Summary of the invention
In view of this, the present invention provides a kind of image processing method, include the following steps:
Original image is divided into multiple regions；
Each pixel is obtained to the pixel value progress Nonlinear Processing of penetrating of each pixel in described each region respectively
First mapping pixel value of point, the Nonlinear Processing make first of each pixel in the region to map pixel value
Near the median of maximum value and minimum value of the integrated distribution in the pixel value in the region；
Linear process is carried out to the first mapping pixel value of each pixel in each region respectively and obtains each pixel
Second mapping pixel value of point；
Treated image is generated using the second mapping pixel value of each pixel in described each region.
Optionally, the Nonlinear Processing is carried out using following formula:
F ' (z)=V3fmap(f (z)),
Wherein, f ' (z) indicates that the first mapping pixel value, f (z) indicate the pixel value, fmapFor default mapping letter
Number, V3For preset constant.
Optionally, the linear process is carried out using following formula:
Wherein, f " (z) indicates that the second mapping pixel value, f ' (z) indicate the first mapping pixel value, V1And V2It is default normal
Number.
Optionally, described that linear process is carried out to the first mapping pixel value of each pixel in each region respectively
Obtain each pixel second mapping pixel value include:
The maximum first in each region is obtained respectively maps pixel value and/or minimum the first mapping pixel value；
Pixel value is mapped according to the maximum first of each region respectively and/or minimum first maps pixel value to respective area
First mapping pixel value of each pixel in domain carries out linear process and obtains the second mapping pixel value of each pixel.
Optionally, the linear process is carried out using following formula:
Wherein f " (z) indicates that the second mapping pixel value, f ' (z) indicate the first mapping pixel value, f ' (z)min
Indicate the minimum first mapping pixel value, f ' (z)maxIndicate the maximum first mapping pixel value.
Optionally, the linear process is carried out using following formula:
Wherein f " (z) indicates that the second mapping pixel value, f ' (z) indicate the first mapping pixel value, f ' (z)max
Indicate the maximum first mapping pixel value.
Optionally, the linear process is carried out using following formula:
Wherein f " (z) indicates that the second mapping pixel value, f ' (z) indicate the first mapping pixel value, f ' (z)min
Indicate the minimum first mapping pixel value, f ' (z)maxIndicate the maximum first mapping pixel value.
The present invention also provides a kind of method for processing fundus images, include the following steps:
Specific region image is extracted in eye fundus image according to eyeground feature；
Handled to obtain treated eye fundus image to the specific region image using above-mentioned image processing method.
Optionally, described the step of original image is divided into multiple regions, comprising:
Determine the diameter of the blood vessel image in the eye fundus image；
The specific region image is divided into multiple regions, the length and width in divided region according to the diameter
It is all larger than the diameter.
Correspondingly, the present invention also provides a kind of image processing equipments, comprising: at least one processor and with it is described
The memory of at least one processor communication connection；Wherein, the memory, which is stored with, to be held by least one described processor
Capable instruction, described instruction are executed by least one described processor, so that at least one described processor executes above-mentioned image
Processing method.
Correspondingly, the present invention also provides a kind of eye fundus image processing equipments, comprising: at least one processor and with
The memory of at least one processor communication connection；Wherein, the memory be stored with can by it is described at least one processing
The instruction that device executes, described instruction are executed by least one described processor, so that the execution of at least one described processor is above-mentioned
Method for processing fundus images.
The method for processing fundus images and equipment provided according to the present invention, is divided into multiple regions for original image first,
Then it is directed to each region respectively, the distribution that Nonlinear Processing changes pixel values in regions is carried out to original pixel value, then to non-
Pixel value after linear process, which carries out linear process, is mapped to pixel value in one fixed interval, utilizes final linear process
Obtained pixel value substitution original pixel value generates treated image, it is possible thereby to the visibility of picture material is improved, for
Crossing dark or overexposure light image has stronger recovery effects, especially particularly evident to the recovery effects of local detail.
Detailed description of the invention
It, below will be to tool in order to illustrate more clearly of the specific embodiment of the invention or technical solution in the prior art
Body embodiment or attached drawing needed to be used in the description of the prior art be briefly described, it should be apparent that, it is described below
Attached drawing be some embodiments of the present invention, for those of ordinary skill in the art, what is do not made the creative labor
Under the premise of, it is also possible to obtain other drawings based on these drawings.
Fig. 1 is the original eye fundus image that the width in the embodiment of the present invention is acquired by image capture device；
Fig. 2 is the flow chart of one of embodiment of the present invention image processing method；
Fig. 3 is the eye fundus image that the image processing method provided according to embodiments of the present invention generates；
Fig. 4 is the flow chart of another image processing method in the embodiment of the present invention；
Fig. 5 is the flow chart of one of embodiment of the present invention method for processing fundus images；
Fig. 6 is the schematic diagram that specific region is determined in original eye fundus image；
Fig. 7 is the eye fundus image obtained after handling specific region.
Specific embodiment
Technical solution of the present invention is clearly and completely described below in conjunction with attached drawing, it is clear that described reality
Applying example is a part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, this field is common
Technical staff's every other embodiment obtained without making creative work belongs to what the present invention protected
Range.
In addition, as long as technical characteristic involved in invention described below different embodiments is each other not
Constituting conflict can be combined with each other.
The embodiment of the present invention provides a kind of image processing method, and this method can have image by computer, server etc.
The electronic equipment of processor executes.In the present embodiment, handled image is the eye fundus image as shown in Fig. 1, but this method
Be not limited to handle eye fundus image, but can for it is any by image capture device acquired image at
Reason.This method comprises the following steps as shown in Figure 2:
Original image is divided into multiple regions, in the present embodiment by the way of equal part, divided an image by S1A
Multiple as broad as long zonules.Such as the original image for a width 1000x1000,100 can be divided into
The region of 100x100.Other division modes can also be taken in addition to equal part mode, such as can be according to the feature of picture material
Size not equal multiple regions are divided into, so that specific position (interested position) is divided sufficiently small, and other positions
Area size can be relatively large.
S2A, respectively to each pixel in described each region penetrate pixel value carry out Nonlinear Processing obtain it is each
First mapping pixel value of pixel.Image handled by this method can be triple channel image (color image), be also possible to
Single channel image (gray level image).By taking single channel image as an example, the information of each pixel, including pixel can be extracted first
Position (x, y) (coordinate) and original pixel value f (z) of the point in original image.
For a certain region, extreme value of the possible integrated distribution of the pixel value f (z) of each pixel in its section is attached
Closely, i.e., the maximum value or minimum value of f (z) of the f (z) of most of pixels in the region.At Nonlinear Mapping
Reason can change the distribution of numerical value, this step make first of each pixel in region map pixel value integrated distribution in
Near the median of the maximum value and minimum value in pixel value in region；
It illustrating as one, it is assumed that the pixel value f (z) of each pixel is distributed in [10,150] in a certain region, and
And the f (z) of wherein most pixel is between 10-20 or 140-150, f of this step to the pixel in the region
(z) Nonlinear Mapping processing is carried out, in the case where not changing section maximum value and minimum value, first made maps picture
Element value f ' (z) is still distributed in [10,150], and the f ' (z) of wherein most pixel is distributed near 70, this will make low
The data value of bright area is shifted toward high brightness, or shifts the data value of highlight regions toward low-light level.
S3A respectively obtains the first mapping pixel value progress linear process of each pixel in each region each
Second mapping pixel value of pixel.Go out second according to the first mapping calculated for pixel values by a kind of specific linear transformation to reflect
Penetrate pixel value f " (z).After linear process calculates, for each region, first of the pixel in region maps pixel
Value will be mapped in a fixed interval.
It illustrates as one, for the image of width under-exposure, it is assumed that there is n pixel in some region, they
The first mapping pixel value be distributed in [0,50] this section, the first of this n pixel the mapping pixel value is carried out linearly
Mapping obtains the second mapping pixel value, in the section after them can be made to be distributed in [30,180] this mapping；
As another citing, the image over-exposed for one, it is assumed that there is n pixel in some region, it
The first mapping pixel value be distributed in [190,250] this section, to the first of this n pixel map pixel value into
Row Linear Mapping obtains the second mapping pixel value, in the section after them can be made to be distributed in [90,120] this mapping.
For under-exposed and both over-exposed images, different mapping function and parameter can be used.At this
The maximum value and Returning to one for minimum valueization processing of reason also referred to as the first mapping pixel value.
S4A generates treated image using the second mapping pixel value of each pixel in each region.Rear
The location information of obtained pixel is extracted in use before in platform treatment process, uses the second mapping picture changed after being distributed
Corresponding part in element value substitution original image, or multiple regions storage is arranged in big figure, processing shown in available Fig. 3
Image afterwards, the details such as sightless optic disk, blood vessel are shown well in Fig. 3 in Fig. 1 originally as shown in the figure.
For triple channel image, then above-mentioned Linear Mapping processing and non-thread is carried out for the pixel value in three channels respectively
Property mapping processing, obtain the second mapping pixel value on three channels, finally producible triple channel image.
Original image is divided into multiple regions first, then by the image processing method provided according to embodiments of the present invention
It is directed to each region respectively, the distribution that Nonlinear Processing changes pixel values in regions is carried out to original pixel value, then to non-linear
Treated, and pixel value progress linear process is mapped to pixel value in one fixed interval, is obtained using final linear process
Pixel value substitution original pixel value generate treated image, it is possible thereby to the visibility of picture material be improved, for excessively dark
Or overexposure light image has stronger recovery effects, it is especially particularly evident to the recovery effects of local detail.
About Nonlinear Processing, as an alternative embodiment, can use following formula progress in above-mentioned steps S2A
Nonlinear Processing:
F ' (z)=V3fmap(f (z)),
Wherein, f ' (z) indicates that the first mapping pixel value, f (z) indicate original pixel value, fmapTo preset mapping function, V3
For preset constant.
Wherein fmapSigmoidal formula can be used, it may be assumed that
σ is the standard deviation of f (z),It is the mean value of the f (z) of all pixels point in region.Letter used in the present invention
Number is not limited to sigmoidal function, as long as function makes being more evenly distributed for image pixel numerical value, does not reaccees maximum value
(such as 255) or minimum value (such as 0).
Because the value range of the output of sigmoidal function is 0 to 1, under the situation using sigmoidal formula,
No matter to excessively dark or over-exposed picture, the value of V3 is both configured to 255.Different for other output areas is non-thread
Property function, the value of V3 are also different.
About linear process, it can use following formula in above-mentioned steps S3A and carry out linear process:
Wherein, f " (z) indicates that the second mapping pixel value, f ' (z) indicate the first mapping pixel value, V1And V2It is default normal
Number.For both over-exposed and excessively dark images, V1And V2Value mode can be different be also possible to it is identical.
Such as it can be according to the maximum pixel of the first mapping pixel value and the first mapping pixel value in the region divided
The smallest pixel carries out value.As an alternative embodiment, may include steps of in above-mentioned steps S3A:
S3A1 obtains the maximum first in each region respectively and maps pixel value and/or minimum the first mapping pixel value,
By taking a certain region as an example, it is assumed that the first mapping pixel value of all pixels point in the region is distributed in [0,50] this section
In, then 0 maps pixel value for the minimum first in the region, and 50 map pixel value for the maximum first in the region；
S3A2 maps pixel value according to the maximum first of each region respectively and/or minimum first maps pixel value to phase
It answers first of each pixel in region to map pixel value progress linear process and obtains the second mapping pixel of each pixel
Value.The maximum first of each region maps pixel value and minimum first maps pixel value and may all be different, this can make line
Property processing embody the difference of each region.
It can map pixel value according to above-mentioned maximum first mapping pixel value and minimum first and determine V1And V2The two ginsengs
Number, for example, by using such as under type:
Wherein f " (z) indicates the second mapping pixel value, f ' (z)minThat is V1 is all f ' (z) in the regionminIn
Minimum value (minimum first map pixel)；V2=f ' (z)max-f′(z)min, it is the maximum value in the region in all f ' (z)
The difference of (maximum first maps pixel) and minimum value.This calculation can be suitable for dark and excessively bright image simultaneously.
Such as following formula can also be used linearly to be calculated:
Wherein f " (z) indicates the second mapping pixel value, V1=0, f ' (z)maxThat is V2 is all f ' in the region
(z)minIn maximum value (maximum first map pixel), this processing mode can obtain preferable effect to excessively dark image.
Such as following formula can also be used linearly to be calculated:
Wherein f " (z) indicates the second mapping pixel value, f ' (z)minThat is V1, f ' (z)maxThat is V2, this mode is to exposure
The excessive image of light can obtain preferable effect.
Another embodiment of the invention additionally provides a kind of image processing method, the difference with previous embodiment
It is the sequence of reverse linear process and non-linear processing steps, specifically, this method comprises the following steps as shown in Figure 4:
Original image is divided into multiple regions, specifically can refer to above-mentioned steps S1A by S1B.
S2B carries out linear process to the pixel value of each pixel in each region respectively and obtains each pixel
First mapping pixel value, specifically can refer to above-mentioned steps S3A.
S3B carries out Nonlinear Processing to the first mapping pixel value of each pixel in described each region respectively and obtains
Second to each pixel maps pixel value, and the Nonlinear Processing makes second of each pixel in the region
The median for mapping maximum value and minimum value of the pixel value integrated distribution in the first mapping pixel value in the region is attached
Closely, it specifically can refer to above-mentioned steps S2A.
S4B generates treated image using the second mapping pixel value of each pixel in each region, specifically
It can refer to above-mentioned steps S4A.
Original image is divided into multiple regions first, then by the image processing method provided according to embodiments of the present invention
It is directed to each region respectively, carrying out linear process to the pixel value of wherein pixel makes pixel value be mapped to a fixed interval
In, then the distribution that Nonlinear Processing changes pixel values in regions is carried out to the pixel value after Linear Mapping, it is final to utilize change point
The pixel value substitution original pixel value of cloth generates treated image, it is possible thereby to the visibility of picture material be improved, for mistake
Dark or overexposure light image has stronger recovery effects, especially particularly evident to the recovery effects of local detail.
As an optional embodiment, it can use following formula in above-mentioned steps S2B and carry out the linear process:
Wherein, f ' (z) indicates that the first mapping pixel value, f (z) indicate the pixel value, V1And V2For preset constant.
As an optional embodiment, it can use following formula in above-mentioned steps S3B and carry out the Nonlinear Processing:
F " (z)=V3fmap(f′(z))
Wherein, f " (z) indicates that the second mapping pixel value, f ' (z) indicate the first mapping pixel value, fmapIt is pre-
If mapping function, V3For preset constant.
The present invention also provides a kind of method for processing fundus images, as shown in figure 5, this method comprises the following steps:
S1C extracts specific region image according to eyeground feature in eye fundus image, this step can be from as shown in FIG. 6
A specific region 61 is extracted in eye fundus image first, specific region image can be the area image in addition to black border,
It can be more specific organic image.Extracted in the present embodiment is optic disk region, specifically can use machine vision or
The methods of person's artificial intelligence identifies optic disk from image, is then based on optic disk and determines a region.
Specific region image is divided into multiple regions by S2C, and S2C-S5C is divided and located just for specific region 61
Reason, detailed process is in combination with Fig. 2 method referring to described in above-described embodiment, and details are not described herein again.
S3C carries out Nonlinear Processing to the pixel value of each pixel in each region respectively and obtains each pixel
First mapping pixel value, Nonlinear Processing make each pixel in region first map pixel value integrated distribution in
Near the median of the maximum value and minimum value in pixel value in region；
S4C respectively obtains the first mapping pixel value progress linear process of each pixel in each region each
Second mapping pixel value of pixel,
S5C generates treated eye fundus image using the second mapping pixel value of each pixel in each region,
Image as shown in Figure 7 finally can be obtained, it can be seen that the details such as sightless blood vessel obtain in Fig. 7 in original image Fig. 6
To show.
It should be noted that above-mentioned steps S3C and S4C can overturn execution, i.e., first original pixel value is linearly located
Reason, then to the pixel value after linear process carry out Nonlinear Processing be also it is feasible, it is specific in combination with Fig. 2 and Fig. 4 referring to above-mentioned
Method described in embodiment, details are not described herein again.
In step S2C, it can specifically include following steps:
S2C1 determines that the diameter d of the blood vessel image in eye fundus image, diameter d can be preset constant, also can use
The modes such as machine vision or artificial intelligence are obtained by the blood vessel image in measurement image.
Specific region image is divided into multiple regions according to diameter d by S2C2, and the length and width in divided region is equal
Greater than the diameter d.
The suitable region of this available size of preferred division mode avoids causing since division region is excessive
Subsequent detail recovery effect is not obvious enough, or the problem for causing subsequent calculation amount excessive since division region is too small.
It should be understood by those skilled in the art that, the embodiment of the present invention can provide as method, system or computer journey
Sequence product.Therefore, complete hardware embodiment, complete software embodiment or combining software and hardware aspects can be used in the present invention
The form of embodiment.Moreover, it wherein includes the calculating of computer usable program code that the present invention, which can be used in one or more,
The computer program implemented in machine usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
The form of product.
The present invention be referring to according to the method for the embodiment of the present invention, the process of equipment (system) and computer program product
Figure and/or block diagram describe.It should be understood that can be realized by computer program instructions each in flowchart and/or the block diagram
The combination of process and/or box in process and/or box and flowchart and/or the block diagram.It can provide these computers
Processor of the program instruction to general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices
To generate a machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute
For realizing the function of being specified in one or more flows of the flowchart and/or one or more blocks of the block diagram
Device.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that instruction stored in the computer readable memory generation includes
The manufacture of command device, the command device are realized in one box of one or more flows of the flowchart and/or block diagram
Or the function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that
Series of operation steps are executed on computer or other programmable devices to generate computer implemented processing, thus calculating
The instruction executed on machine or other programmable devices is provided for realizing in one or more flows of the flowchart and/or side
The step of function of being specified in block diagram one box or multiple boxes.
Obviously, the above embodiments are merely examples for clarifying the description, and does not limit the embodiments.
For those of ordinary skill in the art, other various forms of changes can also be made on the basis of the above description
Change or changes.There is no necessity and possibility to exhaust all the enbodiments.And obvious change extended from this
Change or changes still within the protection scope of the invention.
Claims (10)
1. a kind of image processing method, which comprises the steps of:
Original image is divided into multiple regions；
Each pixel is obtained to the pixel value progress Nonlinear Processing of penetrating of each pixel in described each region respectively
First mapping pixel value, the Nonlinear Processing make first of each pixel in the region to map pixel value concentration point
It is distributed near the median of maximum value and minimum value in the pixel value in the region；
Linear process is carried out to the first mapping pixel value of each pixel in each region respectively and obtains each pixel
Second mapping pixel value；
Treated image is generated using the second mapping pixel value of each pixel in described each region.
2. the method according to claim 1, wherein carrying out the Nonlinear Processing using following formula:
F ' (z)=V3fmap(f (z)),
Wherein, f ' (z) indicates that the first mapping pixel value, f (z) indicate the pixel value, fmapTo preset mapping function, V3For
Preset constant.
3. method according to claim 1 or 2, which is characterized in that carry out the linear process using following formula:
Wherein, f " (z) indicates that the second mapping pixel value, f ' (z) indicate the first mapping pixel value, V1And V2For preset constant.
4. method according to any one of claim 1-3, which is characterized in that respectively to each pixel in each region
Point first mapping pixel value carry out linear process obtain each pixel second mapping pixel value include:
The maximum first in each region is obtained respectively maps pixel value and/or minimum the first mapping pixel value；
Pixel value is mapped according to the maximum first of each region respectively and/or minimum first maps pixel value in corresponding region
First mapping pixel value of each pixel carries out linear process and obtains the second mapping pixel value of each pixel.
5. according to the method described in claim 4, it is characterized in that, carrying out the linear process using following formula:
Wherein f " (z) indicates that the second mapping pixel value, f ' (z) indicate the first mapping pixel value, f ' (z)minIndicate institute
It states minimum first and maps pixel value, f ' (z)maxIndicate the maximum first mapping pixel value.
6. according to the method described in claim 4, it is characterized in that, carrying out the linear process using following formula:
Wherein f " (z) indicates that the second mapping pixel value, f ' (z) indicate the first mapping pixel value, f ' (z)maxIndicate institute
It states maximum first and maps pixel value.
7. according to the method described in claim 4, it is characterized in that, carrying out the linear process using following formula:
Wherein f " (z) indicates that the second mapping pixel value, f ' (z) indicate the first mapping pixel value, f ' (z)minIndicate institute
It states minimum first and maps pixel value, f ' (z)maxIndicate the maximum first mapping pixel value.
8. a kind of method for processing fundus images, which comprises the steps of:
Specific region image is extracted in eye fundus image according to eyeground feature；
The specific region image is carried out handling everywhere using the image processing method any in claim 1-7
Eye fundus image after reason.
9. according to the method described in claim 8, it is characterized in that, described the step of original image is divided into multiple regions,
Include:
Determine the diameter of the blood vessel image in the eye fundus image；
The specific region image is divided into multiple regions according to the diameter, the length and width in divided region is all larger than
The diameter.
10. a kind of image processing equipment characterized by comprising at least one processor and with it is described at least one processing
The memory of device communication connection；Wherein, the memory is stored with the instruction that can be executed by least one described processor, described
Instruction is executed by least one described processor, so that at least one described processor perform claim requires any one of 1-9 institute
The image processing method stated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811236773.4A CN109447964A (en) | 2018-10-23 | 2018-10-23 | Method for processing fundus images and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811236773.4A CN109447964A (en) | 2018-10-23 | 2018-10-23 | Method for processing fundus images and equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109447964A true CN109447964A (en) | 2019-03-08 |
Family
ID=65548235
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811236773.4A Pending CN109447964A (en) | 2018-10-23 | 2018-10-23 | Method for processing fundus images and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109447964A (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102842136A (en) * | 2012-07-19 | 2012-12-26 | 湘潭大学 | Optic disc projection location method synthesizing vascular distribution with video disc appearance characteristics |
| CN105139364A (en) * | 2015-10-10 | 2015-12-09 | 湖北知本信息科技有限公司 | Image enhancement method and application thereof |
| CN106407917A (en) * | 2016-09-05 | 2017-02-15 | 山东大学 | Dynamic scale distribution-based retinal vessel extraction method and system |
| CN107133932A (en) * | 2017-05-04 | 2017-09-05 | 季鑫 | Retina image preprocessing method and device and computing equipment |
| JP2017189530A (en) * | 2016-04-15 | 2017-10-19 | キヤノン株式会社 | Image processing device and imaging processing method and program |
| 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 |
| CN107516083A (en) * | 2017-08-29 | 2017-12-26 | 电子科技大学 | A kind of remote facial image Enhancement Method towards identification |
| CN107578413A (en) * | 2017-10-11 | 2018-01-12 | 吉林大学 | Method, apparatus, equipment and the readable storage medium storing program for executing of retinal images layering |
| CN108198184A (en) * | 2018-01-09 | 2018-06-22 | 北京理工大学 | The method and system of contrastographic picture medium vessels segmentation |
| CN108272434A (en) * | 2017-12-07 | 2018-07-13 | 江威 | The method and device that eye fundus image is handled |
| CN108648193A (en) * | 2018-06-06 | 2018-10-12 | 南方医科大学 | Biological tissue's image knows method for distinguishing and its system, computer storage media |
-
2018
- 2018-10-23 CN CN201811236773.4A patent/CN109447964A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102842136A (en) * | 2012-07-19 | 2012-12-26 | 湘潭大学 | Optic disc projection location method synthesizing vascular distribution with video disc appearance characteristics |
| CN105139364A (en) * | 2015-10-10 | 2015-12-09 | 湖北知本信息科技有限公司 | Image enhancement method and application thereof |
| JP2017189530A (en) * | 2016-04-15 | 2017-10-19 | キヤノン株式会社 | Image processing device and imaging processing method and program |
| CN106407917A (en) * | 2016-09-05 | 2017-02-15 | 山东大学 | Dynamic scale distribution-based retinal vessel extraction method and system |
| CN107133932A (en) * | 2017-05-04 | 2017-09-05 | 季鑫 | Retina image preprocessing method and device and computing equipment |
| 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 |
| CN107516083A (en) * | 2017-08-29 | 2017-12-26 | 电子科技大学 | A kind of remote facial image Enhancement Method towards identification |
| CN107578413A (en) * | 2017-10-11 | 2018-01-12 | 吉林大学 | Method, apparatus, equipment and the readable storage medium storing program for executing of retinal images layering |
| CN108272434A (en) * | 2017-12-07 | 2018-07-13 | 江威 | The method and device that eye fundus image is handled |
| CN108198184A (en) * | 2018-01-09 | 2018-06-22 | 北京理工大学 | The method and system of contrastographic picture medium vessels segmentation |
| CN108648193A (en) * | 2018-06-06 | 2018-10-12 | 南方医科大学 | Biological tissue's image knows method for distinguishing and its system, computer storage media |
Non-Patent Citations (4)
| Title |
|---|
| 何东健: "《数字图像处理》", 28 February 2015 * |
| 尹方平: "《新编生物特征识别与应用》", 30 April 2016 * |
| 张歆雅: "视网膜血管增强与分割算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 张正修 等编著: "《印前图像处理》", 31 August 2002 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Galdran et al. | Automatic red-channel underwater image restoration | |
| CN108491809B (en) | Method and apparatus for generating near infrared image generation model | |
| JP2020530920A (en) | Image lighting methods, devices, electronics and storage media | |
| CN109377462A (en) | Method for processing fundus images and equipment | |
| CN108876742A (en) | image color enhancement method and device | |
| CN105069453A (en) | Image correction method and apparatus | |
| CN109816666B (en) | Symmetrical full convolution neural network model construction method, fundus image blood vessel segmentation device, computer equipment and storage medium | |
| CN107194869B (en) | Image processing method and terminal, computer storage medium and computer equipment | |
| CN107564048A (en) | Based on bifurcation feature registration method | |
| CN113592739B (en) | Lens shading correction method, device and storage medium | |
| CN105701496B (en) | A kind of go disk recognition methods based on artificial intelligence technology | |
| CN110503705A (en) | Image labeling method and equipment | |
| EP4071708B1 (en) | Methods and systems for localised smoke removal and color restoration of real-time video | |
| CN110473176B (en) | Image processing method and device, fundus image processing method and electronic equipment | |
| CN113327263B (en) | Image vision-based fish school activity monitoring method | |
| CN110136140A (en) | Eye fundus image blood vessel image dividing method and equipment | |
| CN114219758A (en) | Defect detection method, system, electronic device and computer readable storage medium | |
| CN109447964A (en) | Method for processing fundus images and equipment | |
| CN116033273A (en) | Image processing method, system and device for eliminating laser layering | |
| AU2021101135A4 (en) | Fundus Image Processing Method and Device | |
| CN112686851B (en) | Image detection method, device and storage medium | |
| CN110189296B (en) | Method and equipment for marking reflecting state of blood vessel wall of fundus image | |
| CN113674157A (en) | Fundus image stitching method, computer device and storage medium | |
| CN116503304A (en) | Image processing and detecting method, computer device and storage medium | |
| KR20180064064A (en) | Method for Detecting Edges on Color Image Based on Fuzzy Theory |
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 |