CN110415262A - Realize 3-D image multiphase segmentation computer installation and three-dimensional dividing method, equipment - Google Patents
Realize 3-D image multiphase segmentation computer installation and three-dimensional dividing method, equipment Download PDFInfo
- Publication number
- CN110415262A CN110415262A CN201910726082.0A CN201910726082A CN110415262A CN 110415262 A CN110415262 A CN 110415262A CN 201910726082 A CN201910726082 A CN 201910726082A CN 110415262 A CN110415262 A CN 110415262A
- Authority
- CN
- China
- Prior art keywords
- image
- segmentation
- multiphase
- function
- region
- 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
- 230000011218 segmentation Effects 0.000 title claims abstract description 121
- 238000000034 method Methods 0.000 title claims abstract description 69
- 238000009434 installation Methods 0.000 title claims abstract description 18
- 239000002365 multiple layer Substances 0.000 claims abstract description 38
- 238000003709 image segmentation Methods 0.000 claims abstract description 36
- 238000002924 energy minimization method Methods 0.000 claims abstract description 8
- 230000006870 function Effects 0.000 claims description 118
- 238000004422 calculation algorithm Methods 0.000 claims description 7
- 238000005457 optimization Methods 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 description 15
- 238000010586 diagram Methods 0.000 description 6
- 230000014509 gene expression Effects 0.000 description 6
- 239000010410 layer Substances 0.000 description 4
- 238000002591 computed tomography Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000003384 imaging method Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000002860 competitive effect Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000010429 evolutionary process Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000005192 partition Methods 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/194—Segmentation; Edge detection involving foreground-background 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
- G06T2207/10—Image acquisition modality
- G06T2207/10028—Range image; Depth image; 3D point clouds
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Image Processing (AREA)
Abstract
The embodiment of the invention discloses a kind of computer installation for realizing 3-D image multiphase segmentation and three-dimensional dividing method, equipment, computer readable storage mediums.Wherein, device includes for the characteristic function based on Heaviside function to be arranged for each region to be split with each zone boundary of determination, to ensure that each pixel automatically falls into the region division module of unique corresponding region, the image segmentation module for being multiple disjoint ranges by three-dimensional image segmentation to be processed using 3-D image multiphase segmentation model.Image segmentation module includes model prebuild submodule and parameter computation module；Multiple-layer horizontal collection of the model prebuild submodule based on 1 continuous horizontal set function by three-dimensional image segmentation be it is multiple do not overlap and region independent construct 3-D image multiphase segmentation model；Parameter computation module calculates the various parameter values of 3-D image multiphase segmentation model using energy minimization method by successive ignition.The application efficiently and accurately realizes the multiphase segmentation to 3-D image.
Description
Technical field
The present embodiments relate to technical field of image processing, more particularly to a kind of 3-D image multiphase segmentation of realizing
Computer installation and three-dimensional dividing method, equipment, computer readable storage medium.
Background technique
Multiphase image segmentation is widely used in the fields such as image procossing, image analysis, computer vision, especially three-dimensional figure
The multiphase segmentation of picture, pattern-recognition, medical diagnosis, geophysical exploration, in terms of have it is important
Researching value.
Multiphase image is divided into the process that image is divided into multiple phases according to certain area division scheme, and one effectively
Multiphase image parted pattern problem to be solved include the expression of energy functional, the partition strategy in multiphase region and evolution
The numerical value of equation calculates.Wherein, the core for being designed as image multiphase segmentation of the selection of area division scheme and provincial characteristics function
The heart, to prevent the overlapping and leakage point between region, multiphase image segmentation must resolve the competitive strategy between region.Current
Multiphase image dividing method has the effect of the multiphase segmentation of two dimensional image, and due to two dimensional image and 3-D image
Difference, suitable for two dimensional image multiphase segmentation method and the multiphase segmentation of 3-D image can not be directly applied for.
In consideration of it, how to realize efficiently and accurately to 3-D image carry out multiphase segmentation be those skilled in the art urgently
It solves the problems, such as.
Summary of the invention
The embodiment of the present disclosure provide a kind of computer installation for realizing 3-D image multiphase segmentation and three-dimensional dividing method,
Equipment, computer readable storage medium efficiently and accurately realize the multiphase segmentation to 3-D image.
In order to solve the above technical problems, the embodiment of the present invention the following technical schemes are provided:
On the one hand the embodiment of the present invention provides a kind of realization 3-D image multiphase segmentation device, comprising:
Region division module, for being based on Heaviside function for 3-D image to be processed each region setting to be split
Characteristic function with each zone boundary of determination, to ensure that each pixel of the 3-D image to be processed is automatically fallen into uniquely
In one region；
Image segmentation module, for utilizing the 3-D image multiphase segmentation model constructed in advance by the three-dimensional figure to be processed
As being divided into multiple disjoint regions；
Wherein, described image segmentation module includes model prebuild submodule and parameter computation module；The model is pre-
Building submodule for the multiple-layer horizontal collection based on 1 continuous horizontal set function by three-dimensional image segmentation be it is multiple do not overlap and
Region independent constructs the 3-D image multiphase segmentation model；The parameter computation module is used for minimum using energy
Change method calculates the various parameter values of the 3-D image multiphase segmentation model by successive ignition.
Optionally, it is n-th of region that the region division module, which is using Regularization Heaviside functional relation calculating formula,
ΩnCharacteristic function χ is setn(φ), the Regularization Heaviside functional relation calculating formula are as follows:
χn(φ)=H (φ-ln-1)(1-H(φ-ln)),ln-1< φ (x)≤ln, n=1,2,3 ...；
In formula, φ (x) ∈ [l0,ln], (0=l0< l1< ... < ln), andN is the region that segmentation obtains
Number, H () are Heaviside function, lnFor n-th of level set of multiple-layer horizontal set function.
Optionally, the model prebuild submodule is to divide the building of variation level set relationship calculating formula based on multiphase image
The 3-D image multiphase segmentation model, the multiphase image divide variation level set relationship calculating formula are as follows:
And φ (x) ∈ [l0,ln], (0=l0< l1< ... < ln), | ▽ φ |=1, li∈Z；
In formula, f is the image intensity of the 3-D image to be processed, and φ is multiple-layer horizontal set function, χi(φ) is i-th
The characteristic function in region, H () are Heaviside function, ui=(u1,u2,…,un) be the 3-D image to be processed image
Piecewise constant of the intensity in each region, lnFor described n-th of level set of multiple-layer horizontal set function, γiFor edge item ∫Ω|▽H(φ-
li) | the weight parameter of dx, n are the areal that segmentation obtains.
It optionally, further include image denoising module；Described image denoises module and is used for the 3-D image to be processed
Segmentation result figure carries out geometry denoising, to obtain multiple smooth subgraphs.
Optionally, the parameter computation module includes:
Energy functional computing unit, for calculating the multiple-layer horizontal set function based on energy functional equivalence relation calculating formula
Energy functional, the energy functional equivalence relation calculating formula are as follows:
In formula, f is the image intensity of the 3-D image to be processed, and φ is multiple-layer horizontal set function, χi(φ) is i-th
The characteristic function in region, ui=(u1,u2,…,un) be the 3-D image to be processed image intensity each region Piecewise Constant
Value, lnFor described n-th of level set of multiple-layer horizontal set function, n is the areal that segmentation obtains, γiFor weight parameter, δ ()
For Dirac function；
Level set function computing unit, for calculating institute using Split-Bregman projection algorithm and alternate optimization method
State the extreme value of energy functional.
On the other hand the embodiment of the present invention provides a kind of 3-D image multiphase segmentation method, comprising:
It is each to determine that the characteristic function based on Heaviside function is set for 3-D image to be processed each region to be split
Zone boundary, to ensure that each pixel of the 3-D image to be processed is automatically fallen into a unique region；
Based on the characteristic function in each region, using the 3-D image multiphase segmentation model constructed in advance by described to be processed three
Dimension image segmentation is multiple disjoint regions；
Wherein, the 3-D image multiphase segmentation model is the multiple-layer horizontal collection based on 1 continuous horizontal set function by three
Dimension image segmentation be it is multiple do not overlap and region independent, and using energy minimization method by successive ignition calculating
Obtained by various parameter values.
Optionally, the 3-D image multiphase segmentation model is that the multiple-layer horizontal collection based on 1 continuous horizontal set function will
Three-dimensional image segmentation be it is multiple do not overlap and region independent, and pass through successive ignition meter using energy minimization method
It calculates obtained by various parameter values and includes:
The 3-D image multiphase segmentation model is constructed based on multiphase image segmentation variation level set relationship calculating formula, it is described
Multiphase image divides variation level set relationship calculating formula are as follows:
And φ (x) ∈ [l0,ln], (0=l0< l1< ... < ln), | ▽ φ |=1, li∈Z；
In formula, f is the image intensity of the 3-D image to be processed, and φ is multiple-layer horizontal set function, χi(φ) is i-th
The characteristic function in region, H () are Heaviside function, ui=(u1,u2,…,un) be the 3-D image to be processed image
Piecewise constant of the intensity in each region, lnFor n-th of level set of multiple-layer horizontal set function, γiFor edge item ∫Ω|▽H(φ-li)|
The weight parameter of dx, n are the areal that segmentation obtains.
Optionally, it is described using the 3-D image multiphase segmentation model that constructs in advance by the three-dimensional image segmentation to be processed
After multiple disjoint regions, further includes:
Geometry denoising is carried out to the segmentation result figure of the 3-D image to be processed, obtains multiple smooth subgraphs.
The embodiment of the invention also provides a kind of 3-D image multiphase segmentation equipment, including processor, the processor is used
The step of realizing the 3-D image multiphase segmentation method as described in preceding any one when executing the computer program stored in memory.
The embodiment of the present invention finally additionally provides a kind of computer readable storage medium, the computer readable storage medium
On be stored with realize 3-D image multiphase segmentation program, it is described realize 3-D image multiphase segmentation program be executed by processor
Shi Shixian is as described in preceding any one the step of 3-D image multiphase segmentation method.
The advantages of technical solution provided by the present application, is, constructs 3-D image multiphase using a multiple-layer horizontal set function
Parted pattern, so that image segmentation module realizes the surface evolution of the n-layer level set of multiple-layer horizontal set function in the picture to figure
As being split, it is only necessary to the parameter that 3-D image multiphase segmentation model can be calculated in the level set function extreme value is calculated,
Be conducive to improve 3-D image multiphase segmentation efficiency；In addition, region division module is special using each region of Heaviside construction of function
Sign function guarantees three-dimensional figure to can ensure that each pixel of 3-D image to be processed is automatically fallen into a unique region
As multiphase segmentation accuracy and precision.
In addition, the embodiment of the present invention provides corresponding reality also directed to the computer installation for realizing 3-D image multiphase segmentation
Existing method, equipment and computer readable storage medium, further such that described device has more practicability and feasibility, the side
Method, equipment and computer readable storage medium have the advantages that corresponding.
It should be understood that the above general description and the following detailed description are merely exemplary, this can not be limited
It is open.
Detailed description of the invention
It, below will be to embodiment or correlation for the clearer technical solution for illustrating the embodiment of the present invention or the relevant technologies
Attached drawing needed in technical description is briefly described, it should be apparent that, the accompanying drawings in the following description is only this hair
Bright some embodiments for those of ordinary skill in the art without creative efforts, can be with root
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is a kind of specific implementation of the computer installation provided in an embodiment of the present invention for realizing 3-D image multiphase segmentation
Mode structure chart；
Fig. 2 is that the another kind of the computer installation provided in an embodiment of the present invention for realizing 3-D image multiphase segmentation is specific real
Apply mode structure chart.
Fig. 3 is a kind of flow diagram of 3-D image multiphase segmentation method provided in an embodiment of the present invention；
Fig. 4 is another flow diagram for realizing 3-D image multiphase segmentation method provided in an embodiment of the present invention；
Fig. 5 is a kind of disclosure 3-D image multiphase segmentation process schematic shown according to an exemplary embodiment；
Fig. 6 is a kind of 3-D image multiphase segmentation process schematic that the disclosure is shown according to another exemplary embodiment.
Specific embodiment
In order to enable those skilled in the art to better understand the solution of the present invention, with reference to the accompanying drawings and detailed description
The present invention is described in further detail.Obviously, described embodiments are only a part of the embodiments of the present invention, rather than
Whole embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not making creative work premise
Under every other embodiment obtained, shall fall within the protection scope of the present invention.
The description and claims of this application and term " first ", " second ", " third " " in above-mentioned attached drawing
Four " etc. be for distinguishing different objects, rather than for describing specific sequence.Furthermore term " includes " and " having " and
Their any deformations, it is intended that cover and non-exclusive include.Such as contain a series of steps or units process, method,
System, product or equipment are not limited to listed step or unit, but may include the step of not listing or unit.
After describing the technical solution of the embodiment of the present invention, the various non-limiting realities of detailed description below the application
Apply mode.
Referring first to Fig. 1, Fig. 1 is a kind of computer dress for realizing 3-D image multiphase segmentation provided in an embodiment of the present invention
Structural schematic diagram in one embodiment is set, the embodiment of the present invention may include the following contents:
Image is divided original image according to certain area division scheme it is understood that multiphase image is divided into
At the process of multiple phases, an effective multiphase image parted pattern problem to be solved includes the expression, more of energy functional
The partition strategy of alpha region and the numerical value of EVOLUTION EQUATION calculate.Wherein, the selection of area division scheme and provincial characteristics function
The core for being designed as image multiphase segmentation, to prevent overlapping between region and leakage point, multiphase image segmentation must be resolved
Competitive strategy between region.In consideration of it, the computer installation for realizing 3-D image multiphase segmentation may include region division module 1
With image segmentation module 2.
It is understood that level set is a kind of effective evolution curve/curved surface implicit representation, i.e., by partitioning boundary or table
Face implicit representation is the level set of the more high-dimension function to develop in Euler's frame, the high-dimension function, that is, level set function.Due to
Level Set Method automatically processes change in topology, to two and three dimensions even more higher-dimension image segmentation expression shape having the same
Formula.Therefore, it during three-dimensional image segmentation is realized, can be used and the energy model of two dimensional image segmentation and region division side
Method.Variation level set method can be used for energy functional E (C) minimization problem for solving to develop for closed curve C, by drawing
Enter imbedding function φ (x) and Heaviside function, E (C) is transformed into E (φ (x)), the calculus of variations is recycled to acquire about φ (x)
Partial differential equation.Wherein, it is desirable that guarantee that level set function remains the feature of symbolic measurement in evolutionary process | ▽
φ |=1.C=(x) | φ (x)=c0}.C is to meet function phi (x) equal to constant value c0Point set, be function phi (x)A water
Flat collection.Function phi (x) is the imbedding function of curve C, referred to as level set function.As constant c0When=0, i.e. C={ (x)|φ (x)=
0 } it is referred to as the zero level collection of level set function φ (x).Based on this, the application can be horizontal using the n-layer of multiple-layer horizontal set function
The surface evolution of collection in the picture is split image, and the energy of surface evolution then can be calculated by variation level set method
The extreme value of functional.Area division scheme based on multiple-layer horizontal collection can be realized by region division module 1, the module can be used for for
The characteristic function based on Heaviside function is arranged with each zone boundary of determination, to ensure to be processed three in each region to be split
Each pixel of dimension image is automatically fallen into a unique region.If f is the image intensity of a 3-D image to be processed,For image-region,For image boundary.Can by n level set of 1 continuous horizontal set function (x | φ (x)=
li, i=0,1,2 ... n }), it divides an image into n and does not overlap region Ω independenti,Wherein, multiple-layer horizontal set function φ (x) meets φ (x) ∈ { l0,l1,…li…
ln, it can be φ (x) ∈ [l by the convex relaxation of φ (x)0,ln], (0=l0< l1< ... < ln) to be converted into convex optimization problem.It is a kind of
Under embodiment, following characteristic function can be set to each region to be split:
That is, it is n-th of area that region division module 1, which can be using Regularization Heaviside functional relation calculating formula,
Domain ΩnCharacteristic function χ is setn(φ), Regularization Heaviside functional relation calculating formula are as follows:
χn(φ)=H (φ-ln-1)(1-H(φ-ln)),ln-1< φ (x)≤ln, n=1,2,3 ...；
Wherein, φ (x) ∈ [l0,ln], (0=l0< l1< ... < ln), andN is the region that segmentation obtains
Number, H () are Heaviside function, lnFor n-th of level set of multiple-layer horizontal set function.Due to φ (x) ∈ [l0,ln], (0=l0
< l1< ... < ln), therefore:
Work as χi(x)=1 when, χi(x)…χi-1, and χ (x)=0i+1(x)…χn(x)=0.
That is,
So havingNaturally it ensure that each pixel automatically falls into a unique region in image, keeps away
Exempt from that repetition and leakage point phenomenon occur in image segmentation process.
In this application, can be used for will be to using the 3-D image multiphase segmentation model that constructs in advance for image segmentation module 2
Processing three-dimensional image segmentation is multiple disjoint regions.Correspondingly, image segmentation module 2 may include model prebuild submodule
21 and parameter computation module 12.Model prebuild submodule 21 can be used for the multiple-layer horizontal based on 1 continuous horizontal set function
Collection by three-dimensional image segmentation be it is multiple do not overlap and region independent construct 3-D image multiphase segmentation model.Parameter meter
Operator module 22 can be used for calculating the items of 3-D image multiphase segmentation model by successive ignition using energy minimization method
Parameter value.Using energy minimization method successive ignition to determine so that the smallest each ginseng of 3-D image multiphase segmentation model energy
Numerical value, that be mainly calculated is level set function i.e. φ (x).
As a preferred embodiment, model prebuild submodule 12 may be based on multiphase image segmentation variation level
Collection relationship calculating formula constructs 3-D image multiphase segmentation model, and multiphase image divides variation level set relationship calculating formula can are as follows:
In formula, f is the image intensity of 3-D image to be processed, and φ is multiple-layer horizontal set function, χi(φ) is ith zone
Characteristic function, H () be Heaviside function, ui=(u1,u2,…,un) it is the image intensity of 3-D image to be processed each
The piecewise constant in region, lnFor n-th of level set of multiple-layer horizontal set function, γiFor edge item ∫Ω|▽H(φ-li) | the weight of dx
Parameter, and γi> 0, n are the areal that segmentation obtains, ∫Ω|▽H(φ-li) | dx is regularization term, alternatively referred to as region Ω i's
Edge item, value are region ΩiEdge surface area.
Wherein, Lipschitz continuous horizontal set function can be used, and above formula meets constraint condition φ (x) ∈ [l0,ln],(0
=l0< l1< ... < ln), and meet simultaneously | ▽ φ |=1 constraint condition, li∈ Z, to guarantee level set function in evolutionary process
In remain the feature of symbolic measurement.U in formulai=(u1,u2,…,un) it is piecewise constant of the f in the Ω of region, estimate
Meter formula can are as follows:
In technical solution provided in an embodiment of the present invention, 3-D image multiphase is constructed using a multiple-layer horizontal set function
Parted pattern, so that image segmentation module realizes the surface evolution of the n-layer level set of multiple-layer horizontal set function in the picture to figure
As being split, it is only necessary to the parameter that 3-D image multiphase segmentation model can be calculated in the level set function extreme value is calculated,
Be conducive to improve 3-D image multiphase segmentation efficiency；In addition, region division module is special using each region of Heaviside construction of function
Sign function guarantees three-dimensional figure to can ensure that each pixel of 3-D image to be processed is automatically fallen into a unique region
As multiphase segmentation precision.
As a preferred embodiment, the application can by energy Functional expression be data item and regularization term two parts,
And the general thermodynamic characteristic function based on above-mentioned rule-basedization Heaviside function as region division, Split- can be used
Bregman projecting method carries out energy to energy functional and minimizes calculating.Based on this, parameter computation module 21 can include:
Energy functional computing unit, for calculating the energy of multiple-layer horizontal set function based on energy functional equivalence relation calculating formula
Functional is measured, energy functional equivalence relation calculating formula can are as follows:
In formula, f is the image intensity of 3-D image to be processed, and φ is multiple-layer horizontal set function, χi(φ) is ith zone
Characteristic function, ui=(u1,u2,…,un) it is piecewise constant of the image intensity of 3-D image to be processed in each region, lnIt is more
Layer n-th of level set of level set function, n are the areal that segmentation obtains, γiFor weight parameter, δ () is Dirac function；
Level set function computing unit, for calculating energy using Split-Bregman projection algorithm and alternate optimization method
Measure the extreme value of functional.
Wherein, the specific implementation process of energy functional computing unit can include:
Heaviside function and the Dirac function of Regularization can be used to realize the function representation of region division.Work as ε
When → 0, Hε(φ) → H (φ) (when ε tends to 0, Hε(φ) approximation Heaviside function H (φ)), ε is the lesser positive number of numerical value.
Its calculation relational expression can are as follows:
It is succinct to write, still using former symbolic formulation.Edge item can be with equivalent representation are as follows:
∫Ω|▽H(φ-li) | dx=∫ | ▽ φ | δ (φ-li)dx。
Therefore, the energy functional of multiple-layer horizontal set function can are as follows:
The process of extreme value is solved to energy functional using efficient, stable Split-Bregman projection algorithm can include:
Introduce auxiliary variable w and Bregman iterative parameterEnergy functional can be changed into following Variation Model iteration lattice
Formula:
Wherein,θ (θ > 0) is punishment parameter, respectively obtains the Europe about φ using alternate optimization method
Draw Lagrange's equation and aboutFor example following relationship calculating formulas of broad sense soft-threshold formula:
Wherein,
φ (x) ∈ [l is constrained to meet0,ln], a bound term can be added to function phi (x) are as follows:
φk+1=min (max (l0,φk+1),ln)。 (3)
Consider constraint | ▽ φ |=1,It can be obtained by sciagraphy:
It updates:
The iterative algorithm step of the application can be as follows:
1) initializes φ0For level set function,
2) estimates ui, i=1,2 ..., n；；
3) successively calculated relationship calculating formula (2), (3), (4), to obtainWith
4) updates
5) if meetsThen stop iteration；It is no to be unsatisfactory for, then continue iteration, i.e. k=k+1, and jump and hold
Row step 2).
It is understood that having noise by initial data, that is, 3-D image to be processed has noise, obtained segmentation
Structure chart contains noise naturally, for each plot quality after the segmentation that improves, is successfully carrying out 3-D image to be processed
After multiphase segmentation, geometry denoising also is carried out using segmentation result figure of the image denoising module 3 to 3-D image to be processed,
To obtain multiple smooth subgraphs.Any image denoising processing method can be used, the application is not limited in any way this, phase
Should denoise realization process see the relevant technologies record realization process just repeat no more herein.
The embodiment of the present invention provides corresponding realization side also directed to the computer installation for realizing 3-D image multiphase segmentation
Method, further such that the computer installation has more feasibility.Below to 3-D image multiphase provided in an embodiment of the present invention
Dividing method is introduced, 3-D image multiphase segmentation method described below and above-described realization 3-D image multiphase point
The computer installation cut can correspond to each other reference.
Referring first to Fig. 3, Fig. 3 is a kind of stream for the method for realizing 3-D image multiphase segmentation provided in an embodiment of the present invention
Journey schematic diagram, the embodiment of the present invention may include the following contents:
S301: for 3-D image to be processed each region to be split be arranged the characteristic function based on Heaviside function with
Each zone boundary is determined, to ensure that each pixel of 3-D image to be processed is automatically fallen into a unique region.
3-D image multiphase segmentation model is the multiple-layer horizontal collection based on 1 continuous horizontal set function by three-dimensional image segmentation
It is not overlapped and region independent to be multiple, and calculates various parameter values by successive ignition using energy minimization method
Gained.
S302: the characteristic function based on each region, using the 3-D image multiphase segmentation model that constructs in advance will it is described to
Processing three-dimensional image segmentation is multiple disjoint regions.
Based on the above embodiment, referring to Fig. 4, may also include that
S303: geometry denoising is carried out to the segmentation result figure of 3-D image to be processed, obtains multiple smooth subgraphs.
The implementation method of each step of 3-D image multiphase segmentation method can be according to above-mentioned apparatus described in the embodiment of the present invention
The corresponding function that each functional module of embodiment has, the correlation that specific implementation process is referred to above-mentioned apparatus embodiment are retouched
It states, details are not described herein again.
From the foregoing, it will be observed that the embodiment of the present invention efficiently and accurately realizes the multiphase segmentation of 3-D image.
In order to make those skilled in the art that the application technical solution to be protected be more clearly understood, present invention also provides
Two illustrative examples illustrate the realization process of 3-D image multiphase segmentation, please refer to Fig. 5 and Fig. 6, Fig. 5 is mandibular sequence
Column split schematic diagram, a is the mandibular under the effect of initial level collection, the mandibular segmentation result figure that b is iteration 3 times, and c is
Iteration 10 times mandibular segmentation result figures；Fig. 6 is dental imaging sequences segmentation schematic diagram, and a is under the effect of initial level collection
Mandibular, b is iteration 3 times dental imaging segmentation result figures, the dental imaging segmentation result figure that c is iteration 10 times, according to
Known to figure, present techniques, which continue scheme, can be very good to realize the multiphase segmentation of 3-D image, have versatility and validity.Figure
As cutting procedure can include:
The embodiment of the present invention is with real human body CT scan image, such as real human body lower jaw position CT scan image sequence and tooth
Teeth portion position CT scan image sequence is as 3-D image to be processed.Using 3 level sets of a Lipschitz level set function
l1=1, l2=3 and l3=7 pairs of 3-D images to be processed carry out multiphase segmentations, each parameter value may be configured as ε=8, α=1, γ=
0.09×2552, θ=0.3.
Level set function is initialized as symbolic measurement first, secondly, estimation ui(i=1,2 ..., n), is successively counted
Following three calculation relational expressions are calculated to obtain
Then it updatesIfThen stop iteration, and shows segmentation result, it is noIt is invalid, then continue iteration, i.e. k=k+1, repeats the above steps until meeting stopping criterion for iteration.
The embodiment of the invention also provides a kind of 3-D image multiphase segmentation equipment, specifically can include:
Memory, for storing computer program；
Processor realizes 3-D image multiphase segmentation side described in any one embodiment as above for executing computer program
The step of method.
The function of each functional module of 3-D image multiphase segmentation equipment can be according to the above method described in the embodiment of the present invention
Method specific implementation in embodiment, specific implementation process are referred to the associated description of above method embodiment, herein not
It repeats again.
From the foregoing, it will be observed that the embodiment of the present invention efficiently and accurately realizes the multiphase segmentation of 3-D image.
The embodiment of the invention also provides a kind of computer readable storage medium, it is stored with and realizes 3-D image multiphase segmentation
Program, the program for realizing 3-D image multiphase segmentation three-dimensional figure described in any one embodiment as above when being executed by processor
As the step of multiphase segmentation method.
The function of each functional module of computer readable storage medium described in the embodiment of the present invention can be according to above method reality
The method specific implementation in example is applied, specific implementation process is referred to the associated description of above method embodiment, herein no longer
It repeats.
From the foregoing, it will be observed that the embodiment of the present invention efficiently and accurately realizes the multiphase segmentation of 3-D image.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with it is other
The difference of embodiment, same or similar part may refer to each other between each embodiment.For being filled disclosed in embodiment
For setting, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, related place is referring to method part
Explanation.
Professional further appreciates that, unit described in conjunction with the examples disclosed in the embodiments of the present disclosure
And algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware and
The interchangeability of software generally describes each exemplary composition and step according to function in the above description.These
Function is implemented in hardware or software actually, the specific application and design constraint depending on technical solution.Profession
Technical staff can use different methods to achieve the described function each specific application, but this realization is not answered
Think beyond the scope of this invention.
The step of method described in conjunction with the examples disclosed in this document or algorithm, can directly be held with hardware, processor
The combination of capable software module or the two is implemented.Software module can be placed in random access memory (RAM), memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
Above to a kind of computer installation for realizing 3-D image multiphase segmentation provided by the present invention and three-dimensional segmentation side
Method, equipment, computer readable storage medium are described in detail.Specific case used herein to the principle of the present invention and
Embodiment is expounded, and the above description of the embodiment is only used to help understand the method for the present invention and its core ideas.
It should be pointed out that for those skilled in the art, it without departing from the principle of the present invention, can also be right
Some improvement and modification can also be carried out by the present invention, and these improvements and modifications also fall within the scope of protection of the claims of the present invention.
Claims (10)
1. a kind of computer installation for realizing 3-D image multiphase segmentation characterized by comprising
Region division module, for the spy based on Heaviside function to be arranged for 3-D image to be processed each region to be split
Function is levied with each zone boundary of determination, to ensure that each pixel of the 3-D image to be processed automatically falls into unique one
In region；
Image segmentation module, for being divided the 3-D image to be processed using the 3-D image multiphase segmentation model constructed in advance
It is segmented into multiple disjoint regions；
Wherein, described image segmentation module includes model prebuild submodule and parameter computation module；The model prebuild
Three-dimensional image segmentation is multiple do not overlap and respectively for the multiple-layer horizontal collection based on 1 continuous horizontal set function by submodule
Independent region constructs the 3-D image multiphase segmentation model；The parameter computation module is used to utilize energy minimum side
Method calculates the various parameter values of the 3-D image multiphase segmentation model by successive ignition.
2. the computer installation according to claim 1 for realizing 3-D image multiphase segmentation, which is characterized in that the region
It is n-th of region Ω that division module, which is using Regularization Heaviside functional relation calculating formula,nCharacteristic function χ is setn(φ), institute
State Regularization Heaviside functional relation calculating formula are as follows:
χn(φ)=H (φ-ln-1)(1-H(φ-ln)),ln-1< φ (x)≤ln, n=1,2,3 ...；
In formula, φ (x) ∈ [l0,ln], (0=l0< l1< ... < ln), andN is the areal that segmentation obtains, H
() is Heaviside function, lnFor n-th of level set of multiple-layer horizontal set function.
3. the computer installation according to claim 2 for realizing 3-D image multiphase segmentation, which is characterized in that the model
Prebuild submodule is to divide variation level set relationship calculating formula based on multiphase image to construct the 3-D image multiphase segmentation mould
Type, the multiphase image divide variation level set relationship calculating formula are as follows:
And φ (x) ∈ [l0,ln], (0=l0< l1< ... < ln),li∈Z；
In formula, f is the image intensity of the 3-D image to be processed, and φ is the multiple-layer horizontal set function, χi(φ) is i-th
The characteristic function in region, H () are Heaviside function, ui=(u1,u2,…,un) be the 3-D image to be processed image
Piecewise constant of the intensity in each region, lnFor described n-th of level set of multiple-layer horizontal set function, γiFor edge itemWeight parameter, n is the obtained areal of segmentation.
4. according to claim 1 to the computer installation for realizing 3-D image multiphase segmentation described in 3 any one, feature exists
In further including image denoising module；Described image denoise module be used for the segmentation result figure of the 3-D image to be processed into
Row geometry denoising, to obtain multiple smooth subgraphs.
5. 3-D image multiphase segmentation computer installation according to claim 4, which is characterized in that the parameter computation
Module includes:
Energy functional computing unit, the energy for calculating multiple-layer horizontal set function based on energy functional equivalence relation calculating formula are general
Letter, the energy functional equivalence relation calculating formula are as follows:
In formula, f is the image intensity of the 3-D image to be processed, and φ is multiple-layer horizontal set function, χi(φ) is ith zone
Characteristic function, ui=(u1,u2,…,un) it is piecewise constant of the image intensity of the 3-D image to be processed in each region, ln
For described n-th of level set of multiple-layer horizontal set function, n is the areal that segmentation obtains, γiFor weight parameter, δ () is
Dirac function；
Level set function computing unit, for calculating the energy using Split-Bregman projection algorithm and alternate optimization method
Measure the extreme value of functional.
6. a kind of 3-D image multiphase segmentation method characterized by comprising
The characteristic function based on Heaviside function is set with each region of determination for 3-D image to be processed each region to be split
Boundary, to ensure that each pixel of the 3-D image to be processed is automatically fallen into a unique region；
Based on the characteristic function in each region, using the 3-D image multiphase segmentation model constructed in advance by the three-dimensional figure to be processed
As being divided into multiple disjoint regions；
Wherein, the 3-D image multiphase segmentation model is the multiple-layer horizontal collection based on 1 continuous horizontal set function by three-dimensional figure
As be divided into it is multiple do not overlap and region independent, and calculated by successive ignition using energy minimization method every
Obtained by parameter value.
7. 3-D image multiphase segmentation method according to claim 6, which is characterized in that the 3-D image multiphase segmentation
It is multiple do not overlap and respectively independent that model, which is the multiple-layer horizontal collection based on 1 continuous horizontal set function by three-dimensional image segmentation,
Region, and using energy minimization method by successive ignition calculate various parameter values obtained by include:
The 3-D image multiphase segmentation model, the multiphase are constructed based on multiphase image segmentation variation level set relationship calculating formula
Image segmentation variation level set relationship calculating formula are as follows:
And φ (x) ∈ [l0,ln], (0=l0< l1< ... < ln),li∈Z；
In formula, f is the image intensity of the 3-D image to be processed, and φ is multiple-layer horizontal set function, χi(φ) is ith zone
Characteristic function, H () be Heaviside function, ui=(u1,u2,…,un) be the 3-D image to be processed image intensity
Piecewise constant in each region, lnFor n-th of level set of multiple-layer horizontal set function, γiFor edge itemWeight
Parameter, n are the areal that segmentation obtains.
8. 3-D image multiphase segmentation method according to claim 7, which is characterized in that described to utilize construct in advance three
Image multiphase segmentation model is tieed up by the three-dimensional image segmentation to be processed as after multiple disjoint regions, further includes:
Geometry denoising is carried out to the segmentation result figure of the 3-D image to be processed, obtains multiple smooth subgraphs.
9. a kind of 3-D image multiphase segmentation equipment, which is characterized in that including processor, the processor is for executing memory
It is realized when the computer program of middle storage as described in any one of claim 6 to 8 the step of 3-D image multiphase segmentation method.
10. a kind of computer readable storage medium, which is characterized in that be stored with realization three on the computer readable storage medium
The program of image multiphase segmentation is tieed up, the program for realizing 3-D image multiphase segmentation is realized when being executed by processor as right is wanted
The step of seeking any one of 6 to the 8 3-D image multiphase segmentation method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910726082.0A CN110415262B (en) | 2019-08-07 | 2019-08-07 | Computer device for realizing three-dimensional image multiphase segmentation and three-dimensional segmentation method and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910726082.0A CN110415262B (en) | 2019-08-07 | 2019-08-07 | Computer device for realizing three-dimensional image multiphase segmentation and three-dimensional segmentation method and equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110415262A true CN110415262A (en) | 2019-11-05 |
| CN110415262B CN110415262B (en) | 2022-01-14 |
Family
ID=68366331
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910726082.0A Active CN110415262B (en) | 2019-08-07 | 2019-08-07 | Computer device for realizing three-dimensional image multiphase segmentation and three-dimensional segmentation method and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110415262B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104835168A (en) * | 2015-05-15 | 2015-08-12 | 东华理工大学 | Fast multi-phase image segmentation method based on global convex variational model |
-
2019
- 2019-08-07 CN CN201910726082.0A patent/CN110415262B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104835168A (en) * | 2015-05-15 | 2015-08-12 | 东华理工大学 | Fast multi-phase image segmentation method based on global convex variational model |
Non-Patent Citations (2)
| Title |
|---|
| 端金鸣: "基于变分水平集方法的通用模型和快速投影算法研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 赵瑞雪: "基于一个水平集函数的多相图像分割方法", 《计算机工程》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110415262B (en) | 2022-01-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104424629B (en) | A kind of x-ray chest radiograph lung segmentation method and apparatus | |
| US8724881B2 (en) | Method and system for precise segmentation of the left atrium in C-arm computed tomography volumes | |
| US9398855B2 (en) | System and method for magnetic resonance imaging based respiratory motion correction for PET/MRI | |
| JP5749735B2 (en) | Bone suppression in X-ray radiographs | |
| JPH05285125A (en) | Method and device for extracting contour in studying multi-sliced and multi-phased heart mri by transmitting seed contour between images | |
| EP2144199B1 (en) | Method and apparatus for multimodal visualization of volume data sets | |
| US9508140B2 (en) | Quantifying curvature of biological structures from imaging data | |
| US9177417B2 (en) | Variable-depth stereotactic surface projections | |
| CN107545579A (en) | A kind of cardiac segmentation method, equipment and storage medium | |
| Chun et al. | Post-reconstruction non-local means filtering methods using CT side information for quantitative SPECT | |
| Krishnan et al. | Analysis of time-dependent flow-sensitive PC-MRI data | |
| WO2011030274A2 (en) | Improvements to curved planar reformation | |
| Biguri et al. | A general method for motion compensation in x-ray computed tomography | |
| CN103903255A (en) | Ultrasound image segmentation method and system | |
| CN110415262A (en) | Realize 3-D image multiphase segmentation computer installation and three-dimensional dividing method, equipment | |
| Jing et al. | A novel 3D reconstruction algorithm of motion-blurred CT image | |
| CN106408541B (en) | Industrial part 3-D image smooth surface method based on Cone-Beam CT | |
| Gibson | Calculating the distance map for binary sampled data | |
| Mackay | Robust contour based surface reconstruction algorithms for applications in medical imaging | |
| Du et al. | 3d skull registration based on registration points automatic correspondence | |
| Hong et al. | Ventricular shape visualization using selective volume rendering of cardiac datasets | |
| EP4336441A1 (en) | Method for the evaluation of the thickness of a tissue of an organ | |
| Brankov et al. | Numerical observer for cardiac motion assessment | |
| Jing et al. | Research Article A Novel 3D Reconstruction Algorithm of Motion-Blurred CT Image | |
| Nadeem et al. | Novel theory and algorithms of path-gradient and their application to image segmentation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |