CN110428377A - Data extending method, apparatus, equipment and medium - Google Patents
Data extending method, apparatus, equipment and medium Download PDFInfo
- Publication number
- CN110428377A CN110428377A CN201910683255.5A CN201910683255A CN110428377A CN 110428377 A CN110428377 A CN 110428377A CN 201910683255 A CN201910683255 A CN 201910683255A CN 110428377 A CN110428377 A CN 110428377A
- Authority
- CN
- China
- Prior art keywords
- target image
- source
- chromaticity
- component
- coloration
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000012937 correction Methods 0.000 claims description 61
- 238000004458 analytical method Methods 0.000 claims description 34
- 238000012216 screening Methods 0.000 claims description 27
- 238000012549 training Methods 0.000 claims description 21
- 238000003860 storage Methods 0.000 claims description 18
- 239000000284 extract Substances 0.000 claims description 8
- 238000004422 calculation algorithm Methods 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000004590 computer program Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 abstract description 8
- 230000002207 retinal effect Effects 0.000 description 20
- 238000004088 simulation Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 230000003287 optical effect Effects 0.000 description 6
- 201000010099 disease Diseases 0.000 description 5
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 5
- 210000004220 fundus oculi Anatomy 0.000 description 5
- 230000011218 segmentation Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 230000005291 magnetic effect Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 238000009826 distribution Methods 0.000 description 3
- 208000035719 Maculopathy Diseases 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000003902 lesion Effects 0.000 description 2
- 208000002780 macular degeneration Diseases 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 210000001525 retina Anatomy 0.000 description 2
- 206010012689 Diabetic retinopathy Diseases 0.000 description 1
- 208000010412 Glaucoma Diseases 0.000 description 1
- 206010018473 Glycosuria Diseases 0.000 description 1
- 206010025421 Macule Diseases 0.000 description 1
- 208000009857 Microaneurysm Diseases 0.000 description 1
- 208000017442 Retinal disease Diseases 0.000 description 1
- 206010038923 Retinopathy Diseases 0.000 description 1
- 208000000260 Warts Diseases 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000004204 blood vessel Anatomy 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000008450 motivation Effects 0.000 description 1
- 210000003733 optic disk Anatomy 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 210000000056 organ Anatomy 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 201000010153 skin papilloma Diseases 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
- G06T5/90—Dynamic range modification of images or parts thereof
-
- 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
- G06T7/90—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
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30041—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)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the invention discloses a kind of data extending method, apparatus, equipment and media, are related to field of image processing.This method comprises: obtaining source target image from target type, the chrominance information of the source target image is extracted；Based on target image chromaticity range, the chrominance information of the source target image is adjusted, generates fresh target image.The embodiment of the invention provides a kind of data extending method, apparatus, equipment and medium, the expansion using a kind of target image of type to other type target images is realized.
Description
Technical field
The present embodiments relate to field of image processing more particularly to a kind of data extending method, apparatus, equipment and Jie
Matter.
Background technique
Retinal fundus images played an important role in fundus oculi disease screening and diagnosis.High-performance eye fundus image point
The acquisition for analysing model needs large batch of high quality mark training data.In order to guarantee model output in practical application scene
Generalization and robustness, the distribution of training data should be almost the same with the data distribution in practical application scene.
And the fundus camera as retinal fundus images acquisition equipment, model currently used in the market are numerous.To not
With the fundus camera of model, due to using different hardware (such as lighting source, photosensitive element etc.) and the software (numbers such as used
Image Post-processing Techniques etc.) configuration, there is also apparent differences (such as Fig. 1 institute in coloration for obtained retinal fundus images
Show).This phenomenon for resulting in the retinal fundus images of different type of machines shooting inconsistent there are data distribution.
However, usual model only includes a kind of type in the data set that the training stage uses due to the limitation of equipment cost
Retinal fundus images or only a small amount of type retinal fundus images, asked so as to cause training dataset is single
Topic.And then the model obtained based on single data set training, the analysis to trained type retinal fundus images are had neither part nor lot in
Accuracy rate is low, and this restrict the practical applications of retinal fundus images analysis model.
Summary of the invention
The embodiment of the present invention provides a kind of data extending method, apparatus, equipment and medium, utilizes a kind of type to realize
Expansion of the target image to other type target images, wherein the target image of other types expanded can be used for model training,
To improve model to the analysis accuracy rate of different type of machines target image.
In a first aspect, the embodiment of the invention provides a kind of data extending methods, this method comprises:
Source target image is obtained from target type, extracts the chrominance information of the source target image；
Based on target image chromaticity range, the chrominance information of the source target image is adjusted, generates fresh target image.
Second aspect, the embodiment of the invention also provides a kind of data extending device, which includes:
Chrominance information extraction module extracts the color of the source target image for obtaining source target image from target type
Spend information；
Chrominance information adjusts module, for being based on target image chromaticity range, adjusts the coloration letter of the source target image
Breath generates fresh target image.
The third aspect, the embodiment of the invention also provides a kind of electronic equipment, the equipment includes:
One or more processors；
Storage device, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processing
Device realizes the data extending method as described in any one of embodiment of the present invention.
Fourth aspect, the embodiment of the invention also provides a kind of computer readable storage mediums, are stored thereon with computer
Program, which is characterized in that the data extending as described in any one of embodiment of the present invention is realized when the program is executed by processor
Method.
The embodiment of the present invention by be based on target image chromaticity range, adjust the chrominance information of the source target image, with
It realizes the simulation to other target images of other types, and then utilizes other target images of other types of simulation and described
Source target image is trained initial analysis model, with improve training complete target analysis model to other types its
The analysis accuracy rate of his target image.
Detailed description of the invention
Fig. 1 is that the prior art uses the fundus camera type of different vendor's production to shoot the retina eye that same eyeground obtains
Base map picture；
Fig. 2 is a kind of flow chart for data extending method that the embodiment of the present invention one provides；
Fig. 3 is a kind of flow chart of data extending method provided by Embodiment 2 of the present invention；
Fig. 4 is a kind of flow chart for data extending method that the embodiment of the present invention three provides；
Fig. 5 is that the Color Channel to same eye fundus image that the embodiment of the present invention three provides carries out gamma using different γ values
The eye fundus image schematic diagram obtained after correction；
Fig. 6 is a kind of flow chart for data extending method that the embodiment of the present invention four provides；
Fig. 7 is a kind of structural schematic diagram for data extending device that the embodiment of the present invention five provides；
Fig. 8 is a kind of structural schematic diagram for equipment that the embodiment of the present invention six provides.
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention rather than limiting the invention.It also should be noted that in order to just
Only the parts related to the present invention are shown in description, attached drawing rather than entire infrastructure.
Embodiment one
Fig. 2 is a kind of flow chart for data extending method that the embodiment of the present invention one provides.The present embodiment is applicable to benefit
Other target images of other types are simulated with the source target image of target type, and utilize source target image and other target figures
As being trained to the analysis model of target image, to improve analysis model to the feelings of the analysis accuracy rate of other target images
Condition.
Optionally, target image can be the image of arbitrary content.Typically, target image is retinal fundus images.
This method can be executed by a kind of data extending device, which can be realized by the mode of software and/or hardware, referring to figure
2, data extending method provided in this embodiment includes:
S110, source target image is obtained from target type, extracts the chrominance information of the source target image.
Wherein, target type is the type of acquisition source target image.
Source target image includes object content, and the image for expanding fresh target image.
Object content can be arbitrary content to be analyzed, such as diseased organ, environment to be analyzed etc..
Typically, object content is eye ground.
Chrominance information is to describe the information of image chroma, specifically can be the numerical value of tri- Color Channels of RGB in image,
It can be the U component and V component of image.
S120, it is based on target image chromaticity range, adjusts the chrominance information of the source target image, generate fresh target figure
Picture.
Wherein, target image chromaticity range refers to the coloration variation range that target image is presented in different type of machines.
Specifically, target image chromaticity range can not do this by those skilled in the art's flexible choice as needed
It limits.
Optionally, described based on target image chromaticity range, it is described before the chrominance information for adjusting the source target image
Method further include:
According to other target images obtained from other at least one types, the target image chromaticity range is determined.
Wherein, other types refer to the type in addition to target type.
Other target images are the target images collected by the equipment of other types.
For improve target image chromaticity range determination accuracy rate, the basis obtained from other at least one types its
His target image, determines the target image chromaticity range, comprising:
Based on chrominance information, other described target images are clustered；
Calculate the coloration mean value of other target images in each classification；
According to coloration mean value of all categories, the target image chromaticity range is determined.
Specifically, it is based on chrominance information, other described target images are clustered, comprising:
Summation is weighted to the numerical value of Color Channel in other target images, determines the coloration of other target images
Value；
Other same or similar target images of chromatic value are gathered for same category.
Optionally, it is based on chrominance information, other described target images are clustered, comprising:
The numerical value in same color channel in other different target images is compared；
Compared according to the numerical value of each Color Channel as a result, being clustered to other described target images.
According to coloration mean value of all categories, the target image chromaticity range is determined, comprising:
The maximum value and minimum value for determining coloration mean value of all categories determine the mesh according to determining maximum value and minimum value
Logo image chromaticity range.
Specifically, it is determined that the maximum value and minimum value of coloration mean value of all categories, true according to determining maximum value and minimum value
The fixed target image chromaticity range, comprising:
Coloration mean value more of all categories determines maximum value and minimum in coloration mean value of all categories according to comparison result
Value；
The chromaticity range that will be made of determining maximum value and minimum value, as the target image chromaticity range.
Optionally, on the basis of above disclosure, those skilled in the art will also be appreciated that other feasible programs, this
Embodiment does not limit this.
The technical solution of the embodiment of the present invention adjusts the source target image by being based on target image chromaticity range
Chrominance information, to realize the simulation to other target images of other types, and then other mesh of other types using simulation
Logo image and the source target image can assist realizing single model while be adapted to a variety of different model fundus cameras, and be not necessarily to
Training data is additionally acquired and marked for different type of machines, and save the cost is trained initial analysis model and improves model
Robustness and universality, be obviously improved the practical value of Related product, improve target analysis model to other types its
The analysis accuracy rate of his target image.
Further, after generating fresh target image, the method also includes:
Using the fresh target image of the source target image and generation, initial analysis model is trained, to obtain mesh
Mark analysis model.
Wherein, target analysis model is the model analyzed target image.The analysis can be point on any mesh ground
Analysis, the present embodiment do not limit this.
Using the technical solution of the embodiment of the present invention, initial analysis model is trained by the data set after expanding and is mentioned
The robustness and universality for having risen model have been obviously improved the practical value of Related product, improve target analysis model to other
The analysis accuracy rate of other target images of type.
Embodiment two
Fig. 3 is a kind of flow chart of data extending method provided by Embodiment 2 of the present invention.The present embodiment is in above-mentioned reality
On the basis of applying example, to " being based on target image chromaticity range, adjust the chrominance information of the source target image, generate fresh target
The step of image ", further spreads out, a kind of optinal plan of proposition.Referring to Fig. 3, data extending method packet provided in this embodiment
It includes:
S210, source target image is obtained from target type, extracts the chrominance information of the source target image.
S220, random generation chromaticity correction coefficient.
Wherein, chromaticity correction coefficient is the parameter that chrominance information adjustment is carried out to the source target image.
Specifically, chromaticity correction coefficient can be generated by random generator.
S230, according to the target image chromaticity range, the chromaticity correction coefficient is screened.
Specifically, according to the target image chromaticity range, the chromaticity correction coefficient is screened, comprising:
According to the chrominance information of the target image chromaticity range and the source target image, determine the source target figure
The minimum correction coefficient of minimal color value of the chromaticity correction of picture into the target image chromaticity range, and by the source target
The maximum correction coefficient of maximum chrominance value of the chromaticity correction of image into the target image chromaticity range；
If the chromaticity correction coefficient is greater than or equal to the minimum correction coefficient, and is less than or equal to the maximum correction
Coefficient, it is determined that the chromaticity correction coefficient passes through screening.
S240, new mesh is generated to source target image progress coloration adjustment using the chromaticity correction coefficient by screening
Logo image.
Specifically, the chromaticity correction coefficient using by screening carries out coloration adjustment to the source target image, raw
At fresh target image, comprising:
If the source target image is rgb format, gamma correction algorithms are based on, the coloration school by screening is utilized
Positive coefficient is corrected three Color Channel numerical value in the source target image respectively；
Merge corrected Color Channel numerical value, generates fresh target image.
Specifically, if the Color Channel number of source target image is three, the chromaticity correction coefficient generated at random be can be
Three, it is also possible to one.If one, then can be based on the same chromaticity correction coefficient respectively to the number of three Color Channels
Value is corrected.
Optionally, the color correction coefficient using by screening carries out coloration adjustment to the source target image, raw
At fresh target image, comprising:
If the source target image is yuv format, using the color correction coefficient by screening respectively to the source mesh
Logo image Central Plains U component and former V component are adjusted, and generate new U component and new V component；
Merge source target image Central Plains Y-component, the new U component and the new V component, generates fresh target image.
Source target image Central Plains U component and former V component are carried out respectively using the color correction coefficient by screening
Adjustment, generates new U component and new V component, comprising:
The product of the color correction coefficient and source target image Central Plains U component by screening is calculated, and by the product
As new U component；
The product of the color correction coefficient and source target image Central Plains V component by screening is calculated, and by the product
As new V component.
The technical solution of the embodiment of the present invention, by generating chromaticity correction coefficient at random；According to the target image coloration
Range screens the chromaticity correction coefficient；Using by screening chromaticity correction coefficient, to the source target image into
The adjustment of row coloration.To realize the random adjustment to source target image.Because being to adjust at random, from probability simulation
The probability of other target images of other types is identical, and then realizes the equalizing training to different type of machines.
Embodiment three
Fig. 4 is a kind of flow chart for data extending method that the embodiment of the present invention three provides.The present embodiment is in above-mentioned reality
On the basis of applying example, by taking target image is retinal fundus images as an example, a kind of optinal plan of proposition.Referring to fig. 4, this implementation
Example provide data extending method include:
S310, source colour retinal fundus images are read.
Source colour retinal fundus images are expressed as h × w × c size matrix in a computer, and wherein h indicates that image is high
Degree, w indicate that picture traverse, c indicate image channel number.Colored retinal fundus images include that (R is indicated tri- Color Channels of RGB
Red, G indicate green, and B indicates blue).
S320, three-dimensional numerical value vector γ={ γ is generated using random number generatorR,γG,γB, it is used as correction coefficient, it takes
Value range is
Wherein n is the positive number greater than 1 manually set, and the specific value of n is true according to retinal fundus images chromaticity range
It is fixed, it can generally set 2 or 3.
S330, gamma correction is carried out respectively to the numerical value of three Color Channels of source colour retinal fundus images.
Wherein R channel correcting coefficient is set as γR, G channel correcting coefficient is set as γG, channel B correction coefficient is set as
γB。
Gamma correction is to carry out nonlinear operation to input picture gray value, makes to export gray value of image and input picture ash
Angle value has exponent relation, and mathematic(al) representation is
Wherein VoIndicate the gray value of output, ViIndicate that the gray value of input, γ indicate correction coefficient.When to colored eyeground
When different γ value progress gamma corrections are respectively adopted in the different color channels numerical value of image, the different colour of coloration can be obtained
Eye fundus image.Fig. 5 is to obtain after carrying out gamma correction using different γ values to same eye fundus image different color channels gray value
Eye fundus image example.
S340, tri- Color Channels of RGB after gamma correction are reconsolidated to obtain new colored eye fundus image.
The technical solution of the embodiment of the present invention is applied different by the numerical value to retinal fundus images different color channels
The gamma correction of degree changes input picture chrominance information, to simulate the quality scheme of a variety of different type of machines.
Example IV
Fig. 6 is a kind of flow chart for data extending method that the embodiment of the present invention four provides.The present embodiment is in above-mentioned reality
On the basis of applying example, by taking the new colored eye fundus image that will expand is applied to the training of retinal fundus images model as an example, propose
A kind of optinal plan.Referring to Fig. 6, data extending method provided in this embodiment includes:
S410, source colour retinal fundus images are read.
S420, three-dimensional numerical value vector γ={ γ is generated using random number generatorR,γG,γB, it is used as correction coefficient, it takes
Value range is
S430, gamma correction is carried out respectively to the numerical value of three Color Channels of source colour retinal fundus images.
S440, tri- Color Channels of RGB after gamma correction are reconsolidated to obtain new colored eye fundus image.
S450, the source eye fundus image of the new colored eye fundus image of expansion and reading is used for subsequent model training.
Wherein, the new colored eye fundus image of expansion is the colored eye fundus image of other types of simulation.
Source eye fundus image is the colored eye fundus image of target type.
Specifically, the source eye fundus image of the new colored eye fundus image of expansion and reading is used for subsequent model training, wrapped
It includes:
Using the source eye fundus image of the new colored eye fundus image and reading of expansion, initial analysis model is trained, is obtained
To the target analysis model of retinal fundus images.The technical solution of the embodiment of the present invention passes through other types that will simulate
The colored eye fundus image of colored eye fundus image and target type is used for subsequent model training as sample data.To improve
The robustness and universality of model promote the practical value of Related product, improve model to other colored eyeground of other types
The analysis accuracy rate of image.
In other words, following effect may be implemented in the embodiment of the present invention:
(1) embodiment of the present invention can assist realizing single model while be adapted to a variety of different model fundus cameras, and be not necessarily to
Training data is additionally acquired and marked for different type of machines, and the robustness of model and pervasive is improved while save the cost
Property, it has been obviously improved the practical value of Related product.
(2) embodiment of the present invention is suitable for a variety of different retinas eyeground figure as a kind of general data extending method
As the training process of analysis model, including but not limited to: fundus oculi disease classification and classification, as diabetic retinopathy classification,
Maculopathy classification etc.；The segmentation of eyeground key structure and positioning, such as optic disk segmentation, central fovea of macula positioning, optical fundus blood vessel segmentation
Deng；The detection of fundus oculi disease key lesion and segmentation, such as microaneurysm detection, the segmentation of glass-film wart.
The application of the embodiment of the present invention can be described as:
It is all kinds of eye fundus image analysis systems using the product or project of the invention, including but not limited to fundus oculi disease is divided
Grade is with categorizing system, the positioning of eyeground key structure with segmenting system, the detection of eyeground lesion with segmenting system etc..With AI
For (Artificial Intelligence, artificial intelligence) eyeground screening all-in-one machine, operator use fundus camera for
Screening person shoots eyeground figure, and the AI algorithm on backstage automatically to the eyeground, analyze by figure, output glaucoma, maculopathy and glycosuria
The risk indicator of the fundus oculi diseases such as sick retinopathy, if a kind of image of type is used only in the disaggregated model that the system uses
As trained source data, and be used only general image extending method (such as Image Reversal, image rotation, setting contrast,
Exposure adjustment, superposition Gaussian noise etc.) expand training dataset, then when being applied to other type fundus cameras, model is defeated
Accuracy will receive influence out, and performance is difficult to be protected.If the embodiment of the present invention is added during model training to propose
Method carry out data extending, then apply in other type fundus cameras, model still can be exported reliably.
It should be noted that by the technical teaching of the present embodiment, those skilled in the art have motivation by above-described embodiment
Described in any embodiment carry out the combination of scheme, to realize that improve model quasi- to the analysis of different type of machines target image
True rate.
Embodiment five
Fig. 7 is a kind of structural schematic diagram for data extending device that the embodiment of the present invention five provides.Referring to Fig. 7, this implementation
The data extending device that example provides includes: chrominance information extraction module 10 and chrominance information adjustment module 20.
Wherein, chrominance information extraction module 10 extracts the source target figure for obtaining source target image from target type
The chrominance information of picture；
Chrominance information adjusts module 20, for being based on target image chromaticity range, adjusts the coloration of the source target image
Information generates fresh target image.
The embodiment of the present invention by be based on target image chromaticity range, adjust the chrominance information of the source target image, with
It realizes the simulation to other target images of other types, and then utilizes other target images of other types of simulation and described
Source target image is trained initial analysis model, with improve training complete target analysis model to other types its
The analysis accuracy rate of his target image.
Further, the chrominance information adjusts module, comprising: correction coefficient generation unit, coefficient screening unit and color
Spend adjustment unit.
Wherein, correction coefficient generation unit, for generating chromaticity correction coefficient at random；
Coefficient screening unit, for being screened to the chromaticity correction coefficient according to the target image chromaticity range；
Coloration adjustment unit, for carrying out coloration to the source target image using the chromaticity correction coefficient by screening
Adjustment generates fresh target image.
Further, the coloration adjustment unit, is specifically used for:
If the source target image is rgb format, gamma correction algorithms are based on, the coloration school by screening is utilized
Positive coefficient is corrected three Color Channel numerical value in the source target image respectively；
Merge corrected Color Channel numerical value, generates fresh target image.
Further, the coloration adjustment unit, is specifically used for:
If the source target image is yuv format, using the color correction coefficient by screening respectively to the source mesh
Logo image Central Plains U component and former V component are adjusted, and generate new U component and new V component；
Merge source target image Central Plains Y-component, the new U component and the new V component, generates fresh target image.
Further, described device further include: chromaticity range determining module.
Wherein, chromaticity range determining module is based on target image chromaticity range for described, adjusts the source target image
Chrominance information before, according to other target images obtained from other at least one types, determine the target image coloration
Range.
Further, the chromaticity range determining module, comprising: cluster cell, coloration average calculation unit and coloration model
Enclose determination unit.
Wherein, cluster cell clusters other described target images for being based on chrominance information；
Coloration average calculation unit, for calculating the coloration mean value of other target images in each classification；
Chromaticity range determination unit, for determining the target image chromaticity range according to coloration mean value of all categories.
Further, described device further include: model training module.
Wherein, the model training module, for utilizing the source target image and life after generating fresh target image
At fresh target image, initial analysis model is trained, to obtain target analysis model
Data extending device provided by the embodiment of the present invention can be performed data provided by any embodiment of the invention and expand
Method is filled, has the corresponding functional module of execution method and beneficial effect.
Embodiment six
Fig. 8 is a kind of structural schematic diagram for equipment that the embodiment of the present invention six provides.Fig. 8, which is shown, to be suitable for being used to realizing this
The block diagram of the example devices 12 of invention embodiment.The equipment 12 that Fig. 8 is shown is only an example, should not be to of the invention real
The function and use scope for applying example bring any restrictions.
As shown in figure 8, equipment 12 is showed in the form of universal computing device.The component of equipment 12 may include but unlimited
In one or more processor or processing unit 16, system storage 28, connecting different system components, (including system is deposited
Reservoir 28 and processing unit 16) bus 18.
Bus 18 indicates one of a few class bus structures or a variety of, including memory bus or Memory Controller,
Peripheral bus, graphics acceleration port, processor or the local bus using any bus structures in a variety of bus structures.It lifts
For example, these architectures include but is not limited to industry standard architecture (ISA) bus, microchannel architecture (MAC)
Bus, enhanced isa bus, Video Electronics Standards Association (VESA) local bus and peripheral component interconnection (PCI) bus.
Equipment 12 typically comprises a variety of computer system readable media.These media can be it is any can be by equipment 12
The usable medium of access, including volatile and non-volatile media, moveable and immovable medium.
System storage 28 may include the computer system readable media of form of volatile memory, such as arbitrary access
Memory (RAM) 30 and/or cache memory 32.Equipment 12 may further include it is other it is removable/nonremovable,
Volatile/non-volatile computer system storage medium.Only as an example, storage system 34 can be used for reading and writing irremovable
, non-volatile magnetic media (Fig. 8 do not show, commonly referred to as " hard disk drive ").Although being not shown in Fig. 8, use can be provided
In the disc driver read and write to removable non-volatile magnetic disk (such as " floppy disk "), and to removable anonvolatile optical disk
The CD drive of (such as CD-ROM, DVD-ROM or other optical mediums) read-write.In these cases, each driver can
To be connected by one or more data media interfaces with bus 18.Memory 28 may include at least one program product,
The program product has one group of (for example, at least one) program module, these program modules are configured to perform each implementation of the invention
The function of example.
Program/utility 40 with one group of (at least one) program module 42 can store in such as memory 28
In, such program module 42 include but is not limited to operating system, one or more application program, other program modules and
It may include the realization of network environment in program data, each of these examples or certain combination.Program module 42 is usual
Execute the function and/or method in embodiment described in the invention.
Equipment 12 can also be communicated with one or more external equipments 14 (such as keyboard, sensing equipment, display 24 etc.),
Can also be enabled a user to one or more equipment interacted with the equipment 12 communication, and/or with enable the equipment 12 with
One or more of the other any equipment (such as network interface card, modem etc.) communication for calculating equipment and being communicated.It is this logical
Letter can be carried out by input/output (I/O) interface 22.Also, equipment 12 can also by network adapter 20 and one or
The multiple networks of person (such as local area network (LAN), wide area network (WAN) and/or public network, such as internet) communication.As shown,
Network adapter 20 is communicated by bus 18 with other modules of equipment 12.It should be understood that although not shown in the drawings, can combine
Equipment 12 use other hardware and/or software module, including but not limited to: microcode, device driver, redundant processing unit,
External disk drive array, RAID system, tape drive and data backup storage system etc..
Processing unit 16 by the program that is stored in system storage 28 of operation, thereby executing various function application and
Data processing, such as realize data extending method provided by the embodiment of the present invention.
Embodiment seven
The embodiment of the present invention seven additionally provides a kind of computer readable storage medium, is stored thereon with computer program, should
The data extending method as described in any one of embodiment of the present invention is realized when program is executed by processor, this method comprises:
Source target image is obtained from target type, extracts the chrominance information of the source target image；
Based on target image chromaticity range, the chrominance information of the source target image is adjusted, generates fresh target image.
The computer storage medium of the embodiment of the present invention, can be using any of one or more computer-readable media
Combination.Computer-readable medium can be computer-readable signal media or computer readable storage medium.It is computer-readable
Storage medium for example may be-but not limited to-the system of electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor, device or
Device, or any above combination.The more specific example (non exhaustive list) of computer readable storage medium includes: tool
There are electrical connection, the portable computer diskette, hard disk, random access memory (RAM), read-only memory of one or more conducting wires
(ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-
ROM), light storage device, magnetic memory device or above-mentioned any appropriate combination.In this document, computer-readable storage
Medium can be any tangible medium for including or store program, which can be commanded execution system, device or device
Using or it is in connection.
Computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal,
Wherein carry computer-readable program code.The data-signal of this propagation can take various forms, including but unlimited
In electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be that computer can
Any computer-readable medium other than storage medium is read, which can send, propagates or transmit and be used for
By the use of instruction execution system, device or device or program in connection.
The program code for including on computer-readable medium can transmit with any suitable medium, including --- but it is unlimited
In wireless, electric wire, optical cable, RF etc. or above-mentioned any appropriate combination.
The computer for executing operation of the present invention can be write with one or more programming languages or combinations thereof
Program code, described program design language include object oriented program language-such as Java, Smalltalk, C++,
It further include conventional procedural programming language-such as " C " language or similar programming language.Program code can be with
It fully executes, partly execute on the user computer on the user computer, being executed as an independent software package, portion
Divide and partially executes or executed on a remote computer or server completely on the remote computer on the user computer.In
Be related in the situation of remote computer, remote computer can pass through the network of any kind --- including local area network (LAN) or
Wide area network (WAN)-be connected to subscriber computer, or, it may be connected to outer computer (such as mentioned using Internet service
It is connected for quotient by internet).
Note that the above is only a better embodiment of the present invention and the applied technical principle.It will be appreciated by those skilled in the art that
The invention is not limited to the specific embodiments described herein, be able to carry out for a person skilled in the art it is various it is apparent variation,
It readjusts and substitutes without departing from protection scope of the present invention.Therefore, although being carried out by above embodiments to the present invention
It is described in further detail, but the present invention is not limited to the above embodiments only, without departing from the inventive concept, also
It may include more other equivalent embodiments, and the scope of the invention is determined by the scope of the appended claims.
Claims (16)
1. a kind of data extending method characterized by comprising
Source target image is obtained from target type, extracts the chrominance information of the source target image；
Based on target image chromaticity range, the chrominance information of the source target image is adjusted, generates fresh target image.
2. adjusting the source the method according to claim 1, wherein described be based on target image chromaticity range
The chrominance information of target image generates fresh target image, comprising:
It is random to generate chromaticity correction coefficient；
According to the target image chromaticity range, the chromaticity correction coefficient is screened；
Using the chromaticity correction coefficient by screening, coloration adjustment is carried out to the source target image, generates fresh target image.
3. according to the method described in claim 2, it is characterized in that, described utilize the chromaticity correction coefficient for passing through screening, to institute
It states source target image and carries out coloration adjustment, generate fresh target image, comprising:
If the source target image is rgb format, gamma correction algorithms are based on, the chromaticity correction system by screening is utilized
Number, is corrected three Color Channel numerical value in the source target image respectively；
Merge corrected Color Channel numerical value, generates fresh target image.
4. according to the method described in claim 2, it is characterized in that, described utilize the color correction coefficient for passing through screening, to institute
It states source target image and carries out coloration adjustment, generate fresh target image, comprising:
If the source target image is yuv format, using the color correction coefficient by screening respectively to the source target figure
As Central Plains U component and former V component are adjusted, new U component and new V component are generated；
Merge source target image Central Plains Y-component, the new U component and the new V component, generates fresh target image.
5. adjusting the source the method according to claim 1, wherein described be based on target image chromaticity range
Before the chrominance information of target image, the method also includes:
According to other target images obtained from other at least one types, the target image chromaticity range is determined.
6. according to the method described in claim 5, it is characterized in that, the basis obtained from other at least one types other
Target image determines the target image chromaticity range, comprising:
Based on chrominance information, other described target images are clustered；
Calculate the coloration mean value of other target images in each classification；
According to coloration mean value of all categories, the target image chromaticity range is determined.
7. adjusting the source the method according to claim 1, wherein described be based on target image chromaticity range
The chrominance information of target image, after generating fresh target image, the method also includes:
Using the fresh target image of the source target image and generation, initial analysis model is trained, to obtain target point
Analyse model.
8. a kind of data extending device characterized by comprising
Chrominance information extraction module extracts the coloration letter of the source target image for obtaining source target image from target type
Breath；
Chrominance information adjusts module, for being based on target image chromaticity range, adjusts the chrominance information of the source target image, life
At fresh target image.
9. device according to claim 8, which is characterized in that the chrominance information adjusts module, comprising:
Correction coefficient generation unit, for generating chromaticity correction coefficient at random；
Coefficient screening unit, for being screened to the chromaticity correction coefficient according to the target image chromaticity range；
Coloration adjustment unit, for carrying out coloration adjustment to the source target image using the chromaticity correction coefficient by screening,
Generate fresh target image.
10. device according to claim 9, which is characterized in that the coloration adjustment unit is specifically used for:
If the source target image is rgb format, gamma correction algorithms are based on, the chromaticity correction system by screening is utilized
Number, is corrected three Color Channel numerical value in the source target image respectively；
Merge corrected Color Channel numerical value, generates fresh target image.
11. device according to claim 9, which is characterized in that the coloration adjustment unit is specifically used for:
If the source target image is yuv format, using the color correction coefficient by screening respectively to the source target figure
As Central Plains U component and former V component are adjusted, new U component and new V component are generated；
Merge source target image Central Plains Y-component, the new U component and the new V component, generates fresh target image.
12. device according to claim 8, which is characterized in that described device further include:
Chromaticity range determining module is based on target image chromaticity range for described, adjusts the coloration letter of the source target image
Before breath, according to other target images obtained from other at least one types, the target image chromaticity range is determined.
13. device according to claim 12, which is characterized in that the chromaticity range determining module, comprising:
Cluster cell clusters other described target images for being based on chrominance information；
Coloration average calculation unit, for calculating the coloration mean value of other target images in each classification；
Chromaticity range determination unit, for determining the target image chromaticity range according to coloration mean value of all categories.
14. device according to claim 8, which is characterized in that described device further include:
Model training module is based on target image chromaticity range for described, adjusts the chrominance information of the source target image, raw
After fresh target image, using the fresh target image of the source target image and generation, initial analysis model is trained,
To obtain target analysis model.
15. a kind of electronic equipment, which is characterized in that the equipment includes:
One or more processors；
Storage device, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processors are real
Now such as data extending method of any of claims 1-7.
16. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is by processor
Such as data extending method of any of claims 1-7 is realized when execution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910683255.5A CN110428377B (en) | 2019-07-26 | 2019-07-26 | Data expansion method, device, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910683255.5A CN110428377B (en) | 2019-07-26 | 2019-07-26 | Data expansion method, device, equipment and medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110428377A true CN110428377A (en) | 2019-11-08 |
| CN110428377B CN110428377B (en) | 2023-06-30 |
Family
ID=68412754
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910683255.5A Active CN110428377B (en) | 2019-07-26 | 2019-07-26 | Data expansion method, device, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110428377B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112710996A (en) * | 2020-12-08 | 2021-04-27 | 中国人民解放军海军航空大学 | Data set expansion method and system for radar micro-motion target identification |
Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002027266A (en) * | 2000-07-04 | 2002-01-25 | Ricoh Co Ltd | Image processing system and image processing method and recording medium |
| CN1642220A (en) * | 2004-01-16 | 2005-07-20 | 精工爱普生株式会社 | Image processing device, image display device, image processing method, and image processing program |
| US20060158704A1 (en) * | 2005-01-18 | 2006-07-20 | Fuji Photo Film Co., Ltd. | Image correction apparatus, method and program |
| US20070177029A1 (en) * | 2006-01-31 | 2007-08-02 | Olympus Corporation | Color correction apparatus |
| JP2012088551A (en) * | 2010-10-20 | 2012-05-10 | Mitsubishi Electric Corp | Color correction processing unit, color correction processing method and multi-display device |
| CN104586397A (en) * | 2015-01-26 | 2015-05-06 | 北京工业大学 | Traditional Chinese medicine tongue color classification perception quantification method combined by equal sense distance method and cluster analysis |
| US20150281664A1 (en) * | 2014-03-31 | 2015-10-01 | Canon Kabushiki Kaisha | Image processing apparatus that performs tone correction, image processing method, and storage medium |
| CN105488509A (en) * | 2015-11-19 | 2016-04-13 | Tcl集团股份有限公司 | Image clustering method and system based on local chromatic features |
| CN106920218A (en) * | 2015-12-25 | 2017-07-04 | 展讯通信（上海）有限公司 | A kind of method and device of image procossing |
| CN107798661A (en) * | 2017-10-17 | 2018-03-13 | 华南理工大学 | A kind of adaptive image enchancing method |
| CN107958470A (en) * | 2017-12-18 | 2018-04-24 | 维沃移动通信有限公司 | A kind of color correcting method, mobile terminal |
| CN108230233A (en) * | 2017-05-16 | 2018-06-29 | 北京市商汤科技开发有限公司 | Data enhancing, treating method and apparatus, electronic equipment and computer storage media |
| CN109345469A (en) * | 2018-09-07 | 2019-02-15 | 苏州大学 | A Speckle Denoising Method in OCT Imaging Based on Conditional Generative Adversarial Networks |
| CN109523485A (en) * | 2018-11-19 | 2019-03-26 | Oppo广东移动通信有限公司 | Image color correction method, device, storage medium and mobile terminal |
| CN109671036A (en) * | 2018-12-26 | 2019-04-23 | 上海联影医疗科技有限公司 | A kind of method for correcting image, device, computer equipment and storage medium |
| CN109902717A (en) * | 2019-01-23 | 2019-06-18 | 平安科技（深圳）有限公司 | Method, device and computer-readable storage medium for automatic lesion identification |
| CN109903256A (en) * | 2019-03-07 | 2019-06-18 | 京东方科技集团股份有限公司 | Model training method, chromatic aberration calibrating method, device, medium and electronic equipment |
| CN110009574A (en) * | 2019-02-13 | 2019-07-12 | 中山大学 | A kind of method that brightness, color adaptively inversely generate high dynamic range images with details low dynamic range echograms abundant |
-
2019
- 2019-07-26 CN CN201910683255.5A patent/CN110428377B/en active Active
Patent Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002027266A (en) * | 2000-07-04 | 2002-01-25 | Ricoh Co Ltd | Image processing system and image processing method and recording medium |
| CN1642220A (en) * | 2004-01-16 | 2005-07-20 | 精工爱普生株式会社 | Image processing device, image display device, image processing method, and image processing program |
| US20060158704A1 (en) * | 2005-01-18 | 2006-07-20 | Fuji Photo Film Co., Ltd. | Image correction apparatus, method and program |
| US20070177029A1 (en) * | 2006-01-31 | 2007-08-02 | Olympus Corporation | Color correction apparatus |
| JP2012088551A (en) * | 2010-10-20 | 2012-05-10 | Mitsubishi Electric Corp | Color correction processing unit, color correction processing method and multi-display device |
| US20150281664A1 (en) * | 2014-03-31 | 2015-10-01 | Canon Kabushiki Kaisha | Image processing apparatus that performs tone correction, image processing method, and storage medium |
| CN104586397A (en) * | 2015-01-26 | 2015-05-06 | 北京工业大学 | Traditional Chinese medicine tongue color classification perception quantification method combined by equal sense distance method and cluster analysis |
| CN105488509A (en) * | 2015-11-19 | 2016-04-13 | Tcl集团股份有限公司 | Image clustering method and system based on local chromatic features |
| CN106920218A (en) * | 2015-12-25 | 2017-07-04 | 展讯通信（上海）有限公司 | A kind of method and device of image procossing |
| CN108230233A (en) * | 2017-05-16 | 2018-06-29 | 北京市商汤科技开发有限公司 | Data enhancing, treating method and apparatus, electronic equipment and computer storage media |
| CN107798661A (en) * | 2017-10-17 | 2018-03-13 | 华南理工大学 | A kind of adaptive image enchancing method |
| CN107958470A (en) * | 2017-12-18 | 2018-04-24 | 维沃移动通信有限公司 | A kind of color correcting method, mobile terminal |
| CN109345469A (en) * | 2018-09-07 | 2019-02-15 | 苏州大学 | A Speckle Denoising Method in OCT Imaging Based on Conditional Generative Adversarial Networks |
| CN109523485A (en) * | 2018-11-19 | 2019-03-26 | Oppo广东移动通信有限公司 | Image color correction method, device, storage medium and mobile terminal |
| CN109671036A (en) * | 2018-12-26 | 2019-04-23 | 上海联影医疗科技有限公司 | A kind of method for correcting image, device, computer equipment and storage medium |
| CN109902717A (en) * | 2019-01-23 | 2019-06-18 | 平安科技（深圳）有限公司 | Method, device and computer-readable storage medium for automatic lesion identification |
| CN110009574A (en) * | 2019-02-13 | 2019-07-12 | 中山大学 | A kind of method that brightness, color adaptively inversely generate high dynamic range images with details low dynamic range echograms abundant |
| CN109903256A (en) * | 2019-03-07 | 2019-06-18 | 京东方科技集团股份有限公司 | Model training method, chromatic aberration calibrating method, device, medium and electronic equipment |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112710996A (en) * | 2020-12-08 | 2021-04-27 | 中国人民解放军海军航空大学 | Data set expansion method and system for radar micro-motion target identification |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110428377B (en) | 2023-06-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021031458A1 (en) | Method and device for image color correction applicable in endoscope, and storage medium | |
| Niu et al. | 2D and 3D image quality assessment: A survey of metrics and challenges | |
| JP3810776B2 (en) | A method for detecting and correcting red eyes in digital images. | |
| US5548697A (en) | Non-linear color corrector having a neural network and using fuzzy membership values to correct color and a method thereof | |
| CN111476849B (en) | Object color recognition method, device, electronic equipment and storage medium | |
| CN102867295B (en) | A kind of color correction method for color image | |
| CN105229697A (en) | Multi-modal prospect background segmentation | |
| CN113723385B (en) | Video processing method and device, neural network training method and device | |
| CN108463823A (en) | A reconstruction method, device and terminal of a user's hair model | |
| CN106910184A (en) | Endoscopic images intestinal bleeding detection method based on depth convolutional neural networks | |
| CN110533612A (en) | Imaging method, device, equipment and the medium of endoscopic images | |
| CN102567969B (en) | Color image edge detection method | |
| CN109427041A (en) | A kind of image white balance method and system, storage medium and terminal device | |
| CN110428377A (en) | Data extending method, apparatus, equipment and medium | |
| CN110310341A (en) | Method, device, equipment and storage medium for generating default parameters in color algorithm | |
| Duan et al. | BoostGAN for occlusive profile face frontalization and recognition | |
| CN118196218B (en) | Fundus image processing method, device and equipment | |
| US20190378249A1 (en) | Recolored collage generation based on color hue distances | |
| CN114494084B (en) | Image color homogenizing method and device, electronic equipment and storage medium | |
| WO2024138793A1 (en) | Image correction method and apparatus based on fpga, and device and medium | |
| US12062175B2 (en) | Method for processing images, electronic device, and storage medium | |
| CN116453171A (en) | Method and device for detecting blood vessel color in white eye area, electronic equipment and medium | |
| CN115657859A (en) | Intelligent interaction system based on virtual reality | |
| AU2021240232A1 (en) | Data collection method and apparatus, device and storage medium | |
| CN107038705A (en) | Retinal image bleeding area segmentation method and device and computing equipment |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20210419 Address after: 100000 Room 220, 2nd Floor, Building 4, No. 1, Shangdi East Road, Haidian District, Beijing Applicant after: Beijing Confucius Health Technology Co.,Ltd. Address before: 100085 Beijing, Haidian District, No. ten on the ground floor, No. 10 Baidu building, layer 2 Applicant before: BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |