CN110390680A - Image partition method, computer equipment and storage medium - Google Patents
Image partition method, computer equipment and storage medium Download PDFInfo
- Publication number
- CN110390680A CN110390680A CN201910597372.XA CN201910597372A CN110390680A CN 110390680 A CN110390680 A CN 110390680A CN 201910597372 A CN201910597372 A CN 201910597372A CN 110390680 A CN110390680 A CN 110390680A
- Authority
- CN
- China
- Prior art keywords
- image
- structure image
- segmentation
- split
- primary structure
- 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
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20036—Morphological image processing
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- 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
- G06T2207/30016—Brain
Abstract
This application involves a kind of image partition method, computer equipment and storage mediums.The described method includes: obtaining image to be split;It is at least one primary structure image by the image segmentation to be split;The primary structure image is the minor structure image of the image to be split;It is at least one secondary structure image by least one described primary structure image segmentation;The secondary structure image is the minor structure image of the primary structure image.By the above method, image to be split can be obtained to the minor structure image of different levels in a manner of being classified segmentation.
Description
Technical field
This application involves image technique fields, are situated between more particularly to a kind of image partition method, computer equipment and storage
Matter.
Background technique
Universal with deep learning and convolutional neural networks, more and more R&D institutions and enterprise march deep learning
Field.The semantic segmentation of medical image is also more and more applied in the workflow of doctor, such as the segmentation of brain image can be with
The segmentation result of convenient classification brain map is provided.
However brain image segmentation method common in the art, all it is difficult disposably to obtain the brain map point of different levels
Cut result.
Summary of the invention
Based on this, it is necessary in view of the above technical problems, provide a kind of image partition method, computer equipment and storage and be situated between
Matter.
A kind of image partition method, which comprises
Obtain image to be split;
It is at least one primary structure image by the image segmentation to be split;The primary structure image is described wait divide
Cut the minor structure image of image;
It is at least one secondary structure image by least one described primary structure image segmentation;The secondary structure image
For the minor structure image of the primary structure image.
In one of the embodiments, based on by training determining first nerves network model by the image to be split
It is divided at least one primary structure image;Based on by the determining nervus opticus network model of training by least one described one
Level structure image segmentation is at least one secondary structure image.
In one of the embodiments, the method also includes:
It is tertiary structure image by least one described secondary structure image segmentation;The tertiary structure image is described two
The minor structure image of level structure image.
In one of the embodiments, it is described by the image segmentation to be split be at least one primary structure image it
Before, further includes:
The image to be split is carried out down-sampled;
Coarse segmentation is carried out to down-sampled obtained image, obtains down-sampled more structural images to be split;
Down-sampled more structural images to be split are restored into original resolution.
In one of the embodiments, the method also includes:
It, will be described wait divide according to primary structure image segmentation request when receiving the request of primary structure image segmentation
Cutting image segmentation is at least one primary structure image;
Or, when receiving the request of secondary structure image segmentation, it will be described according to secondary structure image segmentation request
Primary structure image segmentation is at least one secondary structure image.
In one of the embodiments, the method also includes:
By the image segmentation to be split be at least one primary structure image after, further includes: receiving level-one
When structural images display request, the primary structure image is shown;
Or, by the primary structure image segmentation be at least one secondary structure image after, further includes: receiving
When the display request of secondary structure image, the secondary structure image is shown.
In one of the embodiments, when receiving the display request of primary structure image, further includes: show the level-one
The segmentation morphological feature parameter of structural images;
Or, when receiving the display request of secondary structure image, further includes: show the segmentation shape of the secondary structure image
State characteristic parameter.
The image to be split is medical image in one of the embodiments,.
A kind of computer equipment, including memory and processor, the memory are stored with computer program, the processing
The step of device realizes the above method when executing the computer program.
A kind of computer readable storage medium, is stored thereon with computer program, and the computer program is held by processor
The step of above method is realized when row.
The one or more embodiments of the detail of the application propose in following attached drawing and description.Other spies of the application
Advantage of seeking peace will become obvious from specification, attached drawing and claims.
Detailed description of the invention
Fig. 1 is the applied environment figure of image partition method in one embodiment;
Fig. 2 is the flow diagram of image partition method in one embodiment;
Fig. 3 is the flow diagram of image partition method in another embodiment;
Fig. 4 is the different levels structural images schematic diagram of brain image in a specific embodiment;
Fig. 5 is the internal structure chart of computer equipment in one embodiment.
Specific embodiment
It is with reference to the accompanying drawings and embodiments, right in order to which the objects, technical solutions and advantages of the application are more clearly understood
The application is further elaborated.It should be appreciated that specific embodiment described herein is only used to explain the application, not
For limiting the application.
Image partition method provided by the present application can be applied in application environment as shown in Figure 1.Wherein, terminal 110
It is communicated by network with server 120.Terminal 120 is after getting image to be split, by image segmentation to be split
At least one primary structure image, then again by primary structure image segmentation therein be a secondary structure image, wherein one
Level structure image is the minor structure image of image to be split, and secondary structure image is the minor structure image of primary structure image;Its
In, in one embodiment, terminal 110 obtains image to be split from server 120.Wherein, terminal 110 can be, but not limited to
It is various personal computers, laptop, smart phone, tablet computer and portable wearable device, server 120 can be with
It is realized with the server cluster of the either multiple server compositions of independent server.
In one embodiment, as shown in Fig. 2, providing a kind of image partition method, it is applied in Fig. 1 in this way
It is illustrated for terminal, including step S210 to step S230.
Step S210 obtains image to be split.
Wherein, image to be split is the image for needing to be split operation.In one embodiment, terminal is obtained wait divide
Image is cut, can be by manually by image input terminal to be split;Either it is also possible to terminal from other terminals or server
In get image to be split.In one embodiment, image to be split is medical image.In a specific embodiment, to
Segmented image is brain image.
Image segmentation to be split is at least one primary structure image by step S220.
Wherein, primary structure image is the minor structure image of image to be split;That is, primary structure image be to
The structural images of the next stage split in segmented image.
In one embodiment, it treats segmented image and is split to can be using predetermined manner and be split, such as
In one embodiment, it can use through the determining neural network model of training and realize image segmentation to be split as level-one
Structural images;In another embodiment, it is also possible to be achieved by other means and image to be split is divided into primary structure
Image.It in one embodiment, is extremely by image segmentation to be split based on the preset neural network model determining by training
A few primary structure image.
Further, in one embodiment, an image to be split can be divided into multiple primary structure images, at this
In embodiment, segmented image can be treated when receiving the instruction of primary structure image segmentation and be split, according to primary structure
Image segmentation instruction come determine need by image segmentation to be split be several primary structure images, and determine need be partitioned into
Which of segmented image or which primary structure image.
At least one primary structure image segmentation is at least one secondary structure image by step S230.
Wherein, secondary structure image is the minor structure image of primary structure image, that is, secondary structure image is from level-one knot
The structural images of the next stage split in composition picture.
In one embodiment, primary structure image is split to can be using predetermined manner and is split, example
As in one embodiment, can use through the neural network model of training determination realize by primary structure image segmentation for
Secondary structure image;In another embodiment, it is also possible to be achieved by other means and primary structure image is divided into two
Level structure image.In one embodiment, based on preset by training determining neural network model by primary structure image
It is divided at least one secondary structure image.
Further, in one embodiment, a primary structure image can be divided into multiple secondary structure images, In
In the present embodiment, primary structure image can be split when receiving the instruction of secondary structure image segmentation, according to instruction
It needs primary structure image segmentation to be several secondary structure images to determine;And determination is partitioned into primary structure image
Which or which secondary structure image.
By taking image to be split is brain image as an example, primary structure cutting operation is that brain image can be divided into multiple one
Level structure image, in one embodiment, the primary structure of brain image include white matter, grey matter and cerebrospinal fluid.In one embodiment
In, terminal in a primary structure image segmentation operations, can be after getting brain image and obtain brain image segmentation
Three white matter, grey matter and cerebrospinal fluid primary structure images, are also possible to mind map in a primary structure image segmentation operations
A primary structure image of white matter is obtained as only dividing, or can also be in a primary structure image segmentation operations,
Brain image segmentation is obtained into two primary structure images of white matter and cerebrospinal fluid.It is to be appreciated that in other embodiments, terminal
Brain image segmentation can also be obtained to the primary structure figure of other number and type in a primary structure image segmentation operations
Picture, specifically can be according to before executing primary structure image segmentation operations, and the primary structure image segmentation operations instruction of acquisition comes
Determine the type and number for needing to divide the primary structure image obtained.
In the embodiment that brain image is only divided the primary structure image for obtaining white matter, by primary structure image segmentation
When obtaining secondary structure image, when primary structure is white matter, corresponding secondary structure includes corpus callosum, brain stem etc..In a reality
It applies in example, terminal can be after obtaining the primary structure image that primary structure is white matter in a secondary structure image point
Two secondary structure images for being split in operation to white matter and obtaining corpus callosum, brain stem are cut, are also possible in a second level knot
In structure image segmentation operations, a secondary structure image for only obtaining corpus callosum is split to white matter, or can also be
In secondary structure image segmentation operations, two secondary structure images for obtaining corpus callosum, brain stem are split to white matter.It can
To understand ground, in other embodiments, white matter can also be divided into it in a secondary structure image segmentation operations by terminal
The secondary structure image of its number and type, specifically can according to execute secondary structure image segmentation operations before, the two of acquisition
The instruction of level structure image segmentation operations needs to divide the number and type for obtaining secondary structure image to determine.If desired two are carried out
The primary structure image of level structure image segmentation is grey matter or cerebrospinal fluid, and cutting procedure and primary structure image are white matter
Cutting procedure is similar, and details are not described herein.
It in one embodiment, is extremely by image segmentation to be split based on the first nerves network model by training determination
A few primary structure image;At least one primary structure image is divided based on by the determining nervus opticus network model of training
It is segmented at least one secondary structure image.
Wherein, first nerves network model and nervus opticus network can be the same neural network model, be also possible to
Different neural network model;In the embodiment that the two is same neural network model, the neural network is obtained in training
During model, the image to be split, primary structure image, secondary structure image of voice annotation will be carried as nerve net
The training sample of network model is trained preset neural network model, obtains in the present embodiment through the mind of training determination
Through network model.In the embodiment that the two is different neural network models, the process of each neural network model is obtained in training
In, using the image to be split for carrying voice annotation, primary structure image as training sample, input preset neural network mould
It is trained in type, obtains first nerves network model;Primary structure image, the secondary structure image of voice annotation will be carried
It inputs in preset neural network model and is trained, obtain nervus opticus network model.
Further, in one embodiment, above-mentioned image partition method further include: by least one secondary structure image
It is divided at least one tertiary structure image.Wherein, tertiary structure image is the minor structure image of secondary structure image.
In the present embodiment, by least one primary structure image segmentation be at least one secondary structure image after, two
Level structure image can also continue to segmentation and obtain next stage minor structure image, therefore further divide secondary structure image
It cuts, obtains tertiary structure image.
It is to be appreciated that when being split to obtain tertiary structure image to secondary structure image, and by image to be split
Segmentation obtains primary structure image, process that primary structure image segmentation is obtained secondary structure image is similar, implements at one
In example, it can be using preset mode and realize that segmentation obtains tertiary structure image;Such as it can be based on true by training
Fixed neural network model come realize by least one secondary structure image segmentation be at least one tertiary structure image.At one
In embodiment, a secondary structure image can be divided into multiple tertiary structure images, in the present embodiment, can receive
Tertiary structure image segmentation obtains tertiary structure image to secondary structure image segmentation when instructing, according to tertiary structure image segmentation
Instruction determination needs to be split which, which secondary structure image, needs to obtain which or which three-level knot
Composition picture.
In one embodiment, at least one secondary structure image is divided based on by the determining third nerve network of training
It is segmented at least one tertiary structure image.
It is to be appreciated that in one embodiment, tertiary structure image can also be divided into quaternary structure image, then above-mentioned
Image partition method further include: by least one tertiary structure image segmentation be at least one quaternary structure image.
Above-mentioned image partition method, first will be wait divide using the strategy of classification segmentation after getting image to be split
Cutting image segmentation is at least one primary structure image, is then at least one by wherein at least one primary structure image segmentation
Secondary structure image;Wherein, primary structure image is the minor structure image of image to be split, and secondary structure image is primary structure
The minor structure image of image.By the above method, image grading to be split is divided to the minor structure image for obtaining different levels, it is right
In each minor structure correlation for higher and more minor structure data image to be split, it can not only reduce video memory and may be used also
To improve the precision of segmentation.
Further, in one embodiment, as shown in figure 3, being at least one level-one knot by image segmentation to be split
It further include step S310 to step S330 before composition picture.
It is down-sampled to treat segmented image progress by step S310.
It is down-sampled in digital signal process field, also referred to as subtract acquisition, is a kind of skill of multi-rate digital signal processing
Art or the process for reducing signal sampling rate, commonly used in reducing message transmission rate or size of data.In the present embodiment,
Before treating segmented image and being split, first treat segmented image carry out it is down-sampled, obtain it is down-sampled after image;Such as
This, can reduce the size of data to be split, when carrying out next cutting operation to image, can reduce video memory, thus
Improve the precision of segmentation.
Step S320 carries out coarse segmentation to down-sampled obtained image, obtains down-sampled more structural images to be split.
Coarse segmentation is carried out to down-sampled obtained image wherein it is possible to realize using any one mode, is adopted with obtaining drop
More structural images to be split of sample.
Down-sampled more structural images to be split are restored original resolution by step S330.
After carrying out coarse segmentation, down-sampled more structural images to be split are obtained, at this time again by down-sampled wait divide
The more structural images cut restore original resolution, keep the resolution ratio of finally obtained image higher.
In one embodiment, down-sampled more structural images to be split are reverted to original resolution can be use
It is realized with down-sampled corresponding interpolation method, difference is used to increase sampling frequency, so that image be made to be restored to original resolution.
In one embodiment, above-mentioned image partition method further include:
When receiving the request of primary structure image segmentation, image to be split is divided according to the request of primary structure image segmentation
It is segmented at least one primary structure image;In another embodiment, when receiving the request of secondary structure image segmentation, according to
Primary structure image segmentation is at least one secondary structure image by the request of secondary structure image segmentation.
In the present embodiment, terminal requests to determine needs pair according to the image segmentation when receiving image segmentation request
Which rank of structural images is split, and requests to divide corresponding structural images according to the image segmentation.Further, image
Segmentation request includes the request of primary structure image segmentation, the request of secondary structure image segmentation;Wherein, segmented image segmentation will be treated
The cutting procedure for obtaining primary structure image is denoted as primary structure image segmentation operations process;To be to primary structure image segmentation
The cutting procedure of secondary structure image is denoted as secondary structure image segmentation operations process.When terminal receives primary structure image point
When cutting request, the specific partitioning scheme of image to be split is determined according to the request of primary structure image segmentation, and need to obtain
The number and type of primary structure image;When terminal receives the request of secondary structure image segmentation, according to secondary structure image
Segmentation requests to determine the specific partitioning scheme of primary structure image, and the number and kind of the secondary structure image for needing to obtain
Class.
In one embodiment, image segmentation to be split is being at least one primary structure figure by above-mentioned image partition method
As after, further includes: when receiving the display request of primary structure image, show primary structure image;By primary structure figure
As being divided into after at least one secondary structure image, further includes: when receiving the display request of secondary structure image, display two
Level structure image.
According to image segmentation request by image segmentation to be split be different levels structural images after, further may be used also
To show the different levels structural images of acquisition, in the present embodiment, according to the structural images display request received
Determining needs structural images to be shown, wherein structural images display request includes primary structure image display request, secondary structure
Image display request;In the present embodiment, when receiving the display request of primary structure image, primary structure image is shown;In
In another embodiment, when receiving the display request of secondary structure image, secondary structure image is shown.It is to be appreciated that In
In other embodiments, in the structural images display request for receiving other levels, display shows that request is corresponding with structural images
Structural images.
Further, in one embodiment, terminal can also not hold the segmentation of the structural images of corresponding level also currently
Structural images display request is received when operation, such as is received primary structure image when not treating segmented image segmentation also and shown
Request, terminal is treated segmented image according to the display request of primary structure image and is split at this time, obtains at least one level-one knot
Composition picture, and show the primary structure image of acquisition.That is in the present embodiment, doctor directly sends level-one knot to terminal
The display request of composition picture, terminal are the process for completing to obtain primary structure image to segmentation, show primary structure image.Another
In a embodiment, doctor directly sends the display request of secondary structure image to terminal, shows that request corresponds to secondary structure image
It is for display secondary structure corpus callosum mouth, terminal is in the secondary structure image display request for receiving corpus callosum mouth
When, it treats segmented image and is split the primary structure image for obtaining white matter, then obtain the primary structure image segmentation of white matter
The secondary structure image for obtaining corpus callosum mouth, finally shows the secondary structure image of corpus callosum mouth.It is to be appreciated that other
In embodiment, the display request of other way can also be, that is to say, that doctor can choose any one layer of different brackets point
The segmentation result of the structural images of grade is checked.
In one embodiment, above-mentioned image partition method is when receiving the display request of primary structure image, further includes:
Show the segmentation morphological feature parameter of primary structure image;When receiving the display request of secondary structure image, further includes: aobvious
Show the segmentation morphological feature parameter of secondary structure image.
In the present embodiment, when receiving structural images display request, in addition to display and structural images display request pair
The structural images answered also show the segmentation morphological feature parameter of the structural images;In this way, can be convenient doctor check it is relevant
Information.In the present embodiment, doctor can check the segmentation morphological feature parameter of any primary structure image.
In a specific embodiment, by taking image to be split is brain image as an example, according to Anatomical Structure Knowledge to brain image
It is split;As shown in figure 4, the schematic diagram of the different levels structural images for the brain image in the present embodiment;Wherein, brain image
Primary structure include white matter, grey matter and cerebrospinal fluid, when primary structure is white matter, secondary structure includes: corpus callosum, brain stem
Deng;When primary structure is grey matter, secondary structure includes: top, temporal lobe, frontal lobe and Basal ganglia;When primary structure is cerebrospinal fluid
When, secondary structure includes the ventricles of the brain;When secondary structure be corpus callosum when, tertiary structure include: corpus callosum cadre, corpus callosum mouth and
Splenium of corpus callosum;When secondary structure is temporal lobe, tertiary structure includes: superior temporal gyrus, gyrus temporalis meduus, inferior temporal gyrus and temporo pole;Work as second level
When structure is Basal ganglia, three section structures include amygdaloid nucleus, caudate nucleus, shell core and globus pallidus.Wherein, level-one knot in the present embodiment
Structure can be divided into 3 classes, and secondary structure can be divided into altogether 27 classes, and tertiary structure can be divided into altogether 112 classes.In one embodiment, together
Different structure image under one level can be distinguished by different colours or different filling contents.
The brain map result of the different levels of above-mentioned segmentation is shown that doctor can choose different brackets on software interface
Point a kind of (corresponding diagram mesencephalic tissue), three classes (white matter, grey matter, cerebrospinal fluid in corresponding diagram), 27 classes (callosity in corresponding diagram
Body, brain stem, top ... the ventricles of the brain) different segmentation results is checked, and can check each grade of segmentation morphological feature
Parameter.
By above-mentioned image partition method, the workflow that the segmentation result of the different levels of brain image is applied to doctor is worked as
In, facilitate doctor to check relevant information;Brain tissue is split according to different brain maps, using the plan of classification segmentation
Slightly, situation higher for each minor structure correlation and that minor structure data are more.Video memory, which can not only be reduced, to be mentioned
The precision of height segmentation.
It should be understood that although each step in the flow chart of Fig. 2-3 is successively shown according to the instruction of arrow,
These steps are not that the inevitable sequence according to arrow instruction successively executes.Unless expressly stating otherwise herein, these steps
Execution there is no stringent sequences to limit, these steps can execute in other order.Moreover, at least one in Fig. 2-3
Part steps may include that perhaps these sub-steps of multiple stages or stage are not necessarily in synchronization to multiple sub-steps
Completion is executed, but can be executed at different times, the execution sequence in these sub-steps or stage is also not necessarily successively
It carries out, but can be at least part of the sub-step or stage of other steps or other steps in turn or alternately
It executes.
In one embodiment, a kind of image segmentation device is provided, comprising: image collection module, primary structure segmentation
Module and secondary structure divide module, in which:
Image collection module, for obtaining image to be split.
Primary structure divides module, for being at least one primary structure image by image segmentation to be split;Primary structure
Image is the minor structure image of image to be split.
Secondary structure divides module, for being secondary structure image by least one primary structure image segmentation;Second level knot
Composition picture is the minor structure image of primary structure image.
In one embodiment, above-mentioned image segmentation device further includes tertiary structure figure segmentation module, for will at least one
A secondary structure image segmentation is at least one tertiary structure image.
Specific about image segmentation device limits the restriction that may refer to above for image partition method, herein not
It repeats again.Modules in above-mentioned image segmentation device can be realized fully or partially through software, hardware and combinations thereof.On
Stating each module can be embedded in the form of hardware or independently of in the processor in computer equipment, can also store in a software form
In memory in computer equipment, the corresponding operation of the above modules is executed in order to which processor calls.
Above-mentioned image segmentation device, first will be wait divide using the strategy of classification segmentation after getting image to be split
Cutting image segmentation is at least one primary structure image, is then at least one by wherein at least one primary structure image segmentation
Secondary structure image;Wherein, primary structure image is the minor structure image of image to be split, and secondary structure image is primary structure
The minor structure image of image.By the above method, image grading to be split is divided to the minor structure image for obtaining different levels, it is right
In each minor structure correlation for higher and more minor structure data image to be split, it can not only reduce video memory and may be used also
To improve the precision of segmentation.
In one embodiment, a kind of computer equipment is provided, which can be terminal, internal structure
Figure can be as shown in Figure 5.The computer equipment includes processor, the memory, network interface, display connected by system bus
Screen and input unit.Wherein, the processor of the computer equipment is for providing calculating and control ability.The computer equipment is deposited
Reservoir includes non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with operating system and computer journey
Sequence.The built-in storage provides environment for the operation of operating system and computer program in non-volatile memory medium.The calculating
The network interface of machine equipment is used to communicate with external terminal by network connection.When the computer program is executed by processor with
Realize a kind of image partition method.The display screen of the computer equipment can be liquid crystal display or electric ink display screen,
The input unit of the computer equipment can be the touch layer covered on display screen, be also possible to be arranged on computer equipment shell
Key, trace ball or Trackpad, can also be external keyboard, Trackpad or mouse etc..
It will be understood by those skilled in the art that structure shown in Fig. 5, only part relevant to application scheme is tied
The block diagram of structure does not constitute the restriction for the computer equipment being applied thereon to application scheme, specific computer equipment
It may include perhaps combining certain components or with different component layouts than more or fewer components as shown in the figure.
In one embodiment, a kind of computer equipment, including memory and processor are provided, is stored in memory
Computer program, the processor perform the steps of when executing computer program
Obtain image to be split;
It is at least one primary structure image by image segmentation to be split;Primary structure image is the son knot of image to be split
Composition picture;
It is at least one secondary structure image by least one primary structure image segmentation;Secondary structure image is level-one knot
The minor structure image of composition picture.
In one embodiment, it also performs the steps of when processor executes computer program and is determined based on by training
First nerves network model by image segmentation to be split be at least one primary structure image;Based on preset true by training
At least one primary structure image segmentation is at least one secondary structure image by fixed nervus opticus network model.
In one embodiment, it also performs the steps of when processor executes computer program by least one second level knot
Composition picture is divided into tertiary structure image;Tertiary structure image is the minor structure image of secondary structure image.
In one embodiment, it is also performed the steps of when processor executes computer program and treats segmented image progress
It is down-sampled;
Coarse segmentation is carried out to down-sampled obtained image, obtains down-sampled more structural images to be split;
Down-sampled more structural images to be split are restored into original resolution.
In one embodiment, it is also performed the steps of when processor executes computer program and is receiving primary structure
When image segmentation is requested, request image segmentation to be split to be at least one primary structure figure according to primary structure image segmentation
Picture;
Or, being requested according to secondary structure image segmentation by primary structure when receiving the request of secondary structure image segmentation
Image segmentation is at least one secondary structure image.
In one embodiment, it also performs the steps of when processor executes computer program and divides by image to be split
It is segmented into after at least one primary structure image, further includes: when receiving the display request of primary structure image, show level-one knot
Composition picture;Or, by primary structure image segmentation be at least one secondary structure image after, further includes: receiving second level
When structural images display request, secondary structure image is shown.
In one embodiment, it is also performed the steps of when processor executes computer program and is receiving primary structure
When image display request, further includes: the segmentation morphological feature parameter of display primary structure image;Or, receiving second level knot
When the display request of composition picture, further includes: the segmentation morphological feature parameter of display secondary structure image.
In one embodiment, it is medicine that image to be split is also performed the steps of when processor executes computer program
Image.
In one embodiment, a kind of computer readable storage medium is provided, computer program is stored thereon with, is calculated
Machine program performs the steps of when being executed by processor
Obtain image to be split;
It is at least one primary structure image by image segmentation to be split;Primary structure image is the son knot of image to be split
Composition picture;
It is at least one secondary structure image by least one primary structure image segmentation;Secondary structure image is level-one knot
The minor structure image of composition picture.
In one embodiment, it is also performed the steps of when computer program is executed by processor
It is at least one primary structure based on the determining first nerves network model of training is passed through by image segmentation to be split
Image;Based on by the determining nervus opticus network model of training by least one primary structure image segmentation be at least one two
Level structure image.
In one embodiment, it is also performed the steps of when computer program is executed by processor
It is tertiary structure image by least one secondary structure image segmentation;Tertiary structure image is secondary structure image
Minor structure image.
In one embodiment, it is also performed the steps of when computer program is executed by processor
It is down-sampled to treat segmented image progress;
Coarse segmentation is carried out to down-sampled obtained image, obtains down-sampled more structural images to be split;
Down-sampled more structural images to be split are restored into original resolution.
In one embodiment, it is also performed the steps of when computer program is executed by processor
When receiving the request of primary structure image segmentation, image to be split is divided according to the request of primary structure image segmentation
It is segmented at least one primary structure image;
Or, being requested according to secondary structure image segmentation by primary structure when receiving the request of secondary structure image segmentation
Image segmentation is at least one secondary structure image.
In one embodiment, it is also performed the steps of when computer program is executed by processor
By image segmentation to be split be at least one primary structure image after, further includes: receiving primary structure
When image display request, primary structure image is shown;By primary structure image segmentation be at least one secondary structure image it
Afterwards, further includes: when receiving the display request of secondary structure image, show secondary structure image.
In one embodiment, it is also performed the steps of when computer program is executed by processor
When receiving the display request of primary structure image, further includes: the segmentation morphology of display primary structure image is special
Levy parameter;When receiving the display request of secondary structure image, further includes: the segmentation morphological feature of display secondary structure image
Parameter.
In one embodiment, image to be split is performed the steps of when computer program is executed by processor also as doctor
Learn image.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, the computer program can be stored in a non-volatile computer
In read/write memory medium, the computer program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein,
To any reference of memory, storage, database or other media used in each embodiment provided herein,
Including non-volatile and/or volatile memory.Nonvolatile memory may include read-only memory (ROM), programming ROM
(PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include
Random access memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms,
Such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhancing
Type SDRAM (ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM
(RDRAM), direct memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
Each technical characteristic of above embodiments can be combined arbitrarily, for simplicity of description, not to above-described embodiment
In each technical characteristic it is all possible combination be all described, as long as however, the combination of these technical characteristics be not present lance
Shield all should be considered as described in this specification.
The several embodiments of the application above described embodiment only expresses, the description thereof is more specific and detailed, but simultaneously
It cannot therefore be construed as limiting the scope of the patent.It should be pointed out that coming for those of ordinary skill in the art
It says, without departing from the concept of this application, various modifications and improvements can be made, these belong to the protection of the application
Range.Therefore, the scope of protection shall be subject to the appended claims for the application patent.
Claims (10)
1. a kind of image partition method, which comprises
Obtain image to be split;
It is at least one primary structure image by the image segmentation to be split;The primary structure image is the figure to be split
The minor structure image of picture;
It is at least one secondary structure image by least one described primary structure image segmentation;The secondary structure image is institute
State the minor structure image of primary structure image.
2. the method according to claim 1, wherein based on will by the determining first nerves network model of training
The image segmentation to be split is at least one primary structure image;It will based on the nervus opticus network model by training determination
At least one described primary structure image segmentation is at least one secondary structure image.
3. the method according to claim 1, wherein further include:
It is tertiary structure image by least one described secondary structure image segmentation;The tertiary structure image is the second level knot
The minor structure image of composition picture.
4. the method according to claim 1, wherein it is described by the image segmentation to be split be at least one
Before primary structure image, further includes:
The image to be split is carried out down-sampled;
Coarse segmentation is carried out to down-sampled obtained image, obtains down-sampled more structural images to be split;
Down-sampled more structural images to be split are restored into original resolution.
5. the method according to claim 1, which is characterized in that further include:
When receiving the request of primary structure image segmentation, requested according to the primary structure image segmentation by the figure to be split
As being divided at least one primary structure image;
Or, being requested according to the secondary structure image segmentation by the level-one when receiving the request of secondary structure image segmentation
Structural images are divided at least one secondary structure image.
6. the method according to claim 1, which is characterized in that further include:
By the image segmentation to be split be at least one primary structure image after, further includes: receiving primary structure
When image display request, the primary structure image is shown;
Or, by the primary structure image segmentation be at least one secondary structure image after, further includes: receiving second level
When structural images display request, the secondary structure image is shown.
7. according to the method described in claim 6, it is characterized in that, also being wrapped when receiving the display request of primary structure image
It includes: showing the segmentation morphological feature parameter of the primary structure image;
Or, when receiving the display request of secondary structure image, further includes: show the segmentation morphology of the secondary structure image
Characteristic parameter.
8. the method according to claim 1, which is characterized in that the image to be split is medical image.
9. a kind of computer equipment, including memory and processor, the memory are stored with computer program, feature exists
In the step of processor realizes any one of claims 1 to 8 the method when executing the computer program.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program
The step of method described in any item of the claim 1 to 8 is realized when being executed by processor.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910597372.XA CN110390680A (en) | 2019-07-04 | 2019-07-04 | Image partition method, computer equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910597372.XA CN110390680A (en) | 2019-07-04 | 2019-07-04 | Image partition method, computer equipment and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110390680A true CN110390680A (en) | 2019-10-29 |
Family
ID=68286210
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910597372.XA Pending CN110390680A (en) | 2019-07-04 | 2019-07-04 | Image partition method, computer equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110390680A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111340209A (en) * | 2020-02-18 | 2020-06-26 | 北京推想科技有限公司 | Network model training method, image segmentation method and focus positioning method |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180308237A1 (en) * | 2017-04-21 | 2018-10-25 | Samsung Electronics Co., Ltd. | Image segmentation method and electronic device therefor |
CN109598728A (en) * | 2018-11-30 | 2019-04-09 | 腾讯科技(深圳)有限公司 | Image partition method, device, diagnostic system and storage medium |
CN109785306A (en) * | 2019-01-09 | 2019-05-21 | 上海联影医疗科技有限公司 | Organ delineation method, device, computer equipment and storage medium |
-
2019
- 2019-07-04 CN CN201910597372.XA patent/CN110390680A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180308237A1 (en) * | 2017-04-21 | 2018-10-25 | Samsung Electronics Co., Ltd. | Image segmentation method and electronic device therefor |
CN109598728A (en) * | 2018-11-30 | 2019-04-09 | 腾讯科技(深圳)有限公司 | Image partition method, device, diagnostic system and storage medium |
CN109785306A (en) * | 2019-01-09 | 2019-05-21 | 上海联影医疗科技有限公司 | Organ delineation method, device, computer equipment and storage medium |
Non-Patent Citations (1)
Title |
---|
唐树良等: "《神经系统三维坐标定位诊断学》", 31 October 2019, 辽宁科学技术出版社 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111340209A (en) * | 2020-02-18 | 2020-06-26 | 北京推想科技有限公司 | Network model training method, image segmentation method and focus positioning method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
DE102016211642B4 (en) | PATCH MEMORY SYSTEM | |
CN111161269B (en) | Image segmentation method, computer device, and readable storage medium | |
CN110211130A (en) | Image partition method, computer equipment and storage medium | |
DE102020007951A1 (en) | Use of a neural network used for coloring to generate colored images based on interactive colored edges | |
CN110210543B (en) | Image classification system, method, apparatus and storage medium | |
CN110390679A (en) | Image processing method, computer equipment and readable storage medium storing program for executing | |
CN110189306A (en) | The corresponding compressing vessels of abnormal area determine method and apparatus in brain image | |
CN110751187A (en) | Training method of abnormal area image generation network and related product | |
CN110414570A (en) | Image classification model generating method, device, equipment and storage medium | |
CN110222741A (en) | Prediction technique, model, device, equipment and the storage medium of medical image | |
CN114067389A (en) | Facial expression classification method and electronic equipment | |
CN110390680A (en) | Image partition method, computer equipment and storage medium | |
CN110309855A (en) | Training method, computer equipment and the storage medium of image segmentation | |
EP2469434A1 (en) | Method and device for displaying medical image data | |
CN111383207A (en) | Musculoskeletal ultrasonic image processing method, system and device and readable storage medium | |
CN110209970A (en) | Sequence image display method, system, device, equipment and storage medium | |
CN116415000A (en) | Visual knowledge graph configuration method, device, equipment and storage medium | |
CN110766653B (en) | Image segmentation method and device, computer equipment and storage medium | |
CN113313705B (en) | Pathological image processing system, method and medium | |
CN111899263B (en) | Image segmentation method, device, computer equipment and storage medium | |
CN111368889B (en) | Image processing method and device | |
CN106407294A (en) | Method and device for displaying test data of user | |
CN113628221A (en) | Image processing method, image segmentation model training method and related device | |
CN110400317A (en) | More structural images dividing methods, computer equipment and storage medium | |
CN110533158A (en) | Model Construction method, system and non-volatile readable in computer recording medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191029 |
|
RJ01 | Rejection of invention patent application after publication |