CN110415262B - Computer device for realizing three-dimensional image multiphase segmentation and three-dimensional segmentation method and equipment - Google Patents
Computer device for realizing three-dimensional image multiphase segmentation and three-dimensional segmentation method and equipment Download PDFInfo
- Publication number
- CN110415262B CN110415262B CN201910726082.0A CN201910726082A CN110415262B CN 110415262 B CN110415262 B CN 110415262B CN 201910726082 A CN201910726082 A CN 201910726082A CN 110415262 B CN110415262 B CN 110415262B
- Authority
- CN
- China
- Prior art keywords
- dimensional image
- segmentation
- level set
- image
- function
- 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.)
- Active
Links
- 230000011218 segmentation Effects 0.000 title claims abstract description 113
- 238000000034 method Methods 0.000 title claims abstract description 73
- 238000003709 image segmentation Methods 0.000 claims abstract description 35
- 238000004364 calculation method Methods 0.000 claims abstract description 29
- 238000010276 construction Methods 0.000 claims abstract description 11
- 238000002924 energy minimization method Methods 0.000 claims abstract description 10
- 230000006870 function Effects 0.000 claims description 120
- 230000014509 gene expression Effects 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 8
- 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
- 238000000354 decomposition reaction Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 description 18
- 238000010586 diagram Methods 0.000 description 10
- 210000004373 mandible Anatomy 0.000 description 5
- 238000002591 computed tomography Methods 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 238000000638 solvent extraction Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000005516 engineering process Methods 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
- 238000011160 research Methods 0.000 description 1
Images
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 computer device for realizing three-dimensional image multiphase segmentation, a three-dimensional segmentation method, three-dimensional segmentation equipment and a computer readable storage medium. The device comprises a region dividing module and an image dividing module, wherein the region dividing module is used for setting a characteristic function based on a Heaviside function for each region to be divided so as to determine the boundary of each region, so that each pixel point can automatically fall into a unique corresponding region, and the image dividing module is used for dividing the three-dimensional image to be processed into a plurality of disjoint regions by using a three-dimensional image multiphase dividing model. The image segmentation module comprises a model pre-construction sub-module and a parameter calculation sub-module; the model pre-construction sub-module divides the three-dimensional image into a plurality of non-overlapping and independent areas based on the multi-layer level set of 1 continuous level set function to construct a three-dimensional image multiphase division model; and the parameter calculation submodule calculates various parameter values of the three-dimensional image multiphase segmentation model through multiple iterations by using an energy minimization method. The method and the device efficiently and accurately realize the multi-phase segmentation of the three-dimensional image.
Description
Technical Field
The embodiment of the invention relates to the technical field of image processing, in particular to a computer device for realizing three-dimensional image multiphase segmentation, a three-dimensional segmentation method, three-dimensional segmentation equipment and a computer readable storage medium.
Background
The multi-phase image segmentation is widely applied to the fields of image processing, image analysis, computer vision and the like, and particularly has important research values in the aspects of pattern recognition, medical diagnosis, geophysical exploration, three-dimensional image reconstruction and the like.
The multiphase image segmentation is a process of segmenting an image into a plurality of phases according to a certain region segmentation scheme, and the problems to be solved by an effective multiphase image segmentation model comprise expression of an energy functional, a partitioning strategy of a multiphase region and numerical calculation of an evolution equation. The selection of the region division scheme and the design of the region feature function are the core of image multi-phase segmentation, and in order to prevent overlapping and missing division between regions, the multi-phase image segmentation must solve a competition strategy between good regions. The present multi-phase image segmentation method has a good effect on multi-phase segmentation of two-dimensional images, but the multi-phase segmentation method applied to two-dimensional images cannot be directly applied to multi-phase segmentation of three-dimensional images due to the difference between the two-dimensional images and the three-dimensional images.
In view of this, how to implement efficient and accurate multi-phase segmentation on a three-dimensional image is an urgent problem to be solved by those skilled in the art.
Disclosure of Invention
The embodiment of the disclosure provides a computer device, a three-dimensional segmentation method and equipment, and a computer readable storage medium for realizing three-dimensional image multiphase segmentation, and the multiphase segmentation of a three-dimensional image is efficiently and accurately realized.
In order to solve the above technical problems, embodiments of the present invention provide the following technical solutions:
an embodiment of the present invention provides a device for implementing multi-phase segmentation of a three-dimensional image, including:
the area division module is used for setting a characteristic function based on a Heaviside function for each area to be divided of the three-dimensional image to be processed so as to determine the boundary of each area, so that each pixel point of the three-dimensional image to be processed can automatically fall into a unique area;
the image segmentation module is used for segmenting the three-dimensional image to be processed into a plurality of disjoint areas by utilizing a pre-constructed three-dimensional image multiphase segmentation model;
the image segmentation module comprises a model pre-construction sub-module and a parameter calculation sub-module; the model pre-construction sub-module is used for constructing the three-dimensional image multiphase segmentation model by segmenting the three-dimensional image into a plurality of mutually non-overlapped and respectively independent areas based on the multilayer level set of 1 continuous level set function; and the parameter calculation submodule is used for calculating various parameter values of the three-dimensional image multi-phase segmentation model through multiple iterations by using an energy minimization method.
Optionally, the region division module uses a normalized Heaviside function relation calculation formula as an nth region ΩnSetting a characteristic function χn(phi), the regularized Heaviside function relation calculation formula is as follows:
χn(φ)=H(φ-ln-1)(1-H(φ-ln)),ln-1<φ(x)≤ln,n=1,2,3……;
wherein phi (x) is an element of [ l ∈ [ ]0,ln],(0=l0<l1<…<ln) And is andn is the number of regions obtained by segmentation, H () is the Heaviside function, lnThe nth level set is a multi-level set function.
Optionally, the model pre-construction sub-module is configured to construct the three-dimensional image multi-phase segmentation model based on a multi-phase image segmentation variation level set relational computation formula, where the multi-phase image segmentation variation level set relational computation formula is:
and phi (x) belongs to [ l ∈ ]0,ln],(0=l0<l1<…<ln),|▽φ|=1,li∈Z;
Wherein f is the image intensity of the three-dimensional image to be processed, phi is a multi-layer level set function, and chii(phi) is the characteristic function of the ith region, H () is the Heaviside function, ui=(u1,u2,…,un) A constant value l for each region of the image intensity of the three-dimensional image to be processednFor the nth level set, γ, of said multi-level set functioniIs an edge term-Ω|▽H(φ-li)|dxN is the number of divided regions.
Optionally, the system further comprises an image denoising module; the image denoising module is used for carrying out geometric denoising processing on the segmentation result image of the three-dimensional image to be processed so as to obtain a plurality of smooth sub-images.
Optionally, the parameter calculation sub-module includes:
an energy functional computing unit, configured to compute an energy functional of the multi-layer level set function based on an energy functional equivalence relation computational expression, where the energy functional equivalence relation computational expression is:
wherein f is the image intensity of the three-dimensional image to be processed, phi is a multi-layer level set function, and chii(phi) is a characteristic function of the ith region, ui=(u1,u2,…,un) A constant value l for each region of the image intensity of the three-dimensional image to be processednIs the nth level set of the multi-level set function, n is the number of the divided regions, gammaiδ () is a Dirac function, being a weight parameter;
and the level set function calculation unit is used for calculating the extreme value of the energy functional by utilizing a Split-Bregman projection algorithm and an alternative optimization method.
Another aspect of the embodiments of the present invention provides a three-dimensional image multi-phase segmentation method, including:
setting a characteristic function based on a Heaviside function for each region to be segmented of the three-dimensional image to be processed to determine the boundary of each region so as to ensure that each pixel point of the three-dimensional image to be processed automatically falls into a unique region;
based on the characteristic function of each region, segmenting the three-dimensional image to be processed into a plurality of disjoint regions by utilizing a pre-constructed three-dimensional image multiphase segmentation model;
the three-dimensional image multiphase segmentation model is obtained by segmenting a three-dimensional image into a plurality of non-overlapping and independent areas based on a multi-layer level set of 1 continuous level set function and calculating various parameter values through multiple iterations by using an energy minimization method.
Optionally, the three-dimensional image multiphase segmentation model is obtained by segmenting the three-dimensional image into a plurality of non-overlapping and independent regions based on a multi-layer level set of 1 continuous level set function, and calculating each parameter value through multiple iterations by using an energy minimization method, and includes:
constructing the three-dimensional image multi-phase segmentation model based on a multi-phase image segmentation variation level set relational calculation formula, wherein the multi-phase image segmentation variation level set relational calculation formula is as follows:
and phi (x) belongs to [ l ∈ ]0,ln],(0=l0<l1<…<ln),|▽φ|=1,li∈Z;
Wherein f is the image intensity of the three-dimensional image to be processed, phi is a multi-layer level set function, and chii(phi) is the characteristic function of the ith region, H () is the Heaviside function, ui=(u1,u2,…,un) A constant value l for each region of the image intensity of the three-dimensional image to be processednFor the nth level set, gamma, of a multi-level set functioniIs an edge term-Ω|▽H(φ-li) And | dx is a weight parameter, and n is the number of the divided areas.
Optionally, after the pre-constructed three-dimensional image multiphase segmentation model is used to segment the three-dimensional image to be processed into a plurality of disjoint regions, the method further includes:
and carrying out geometric denoising processing on the segmentation result graph of the three-dimensional image to be processed to obtain a plurality of smooth sub-images.
An embodiment of the present invention further provides a three-dimensional image multiphase segmentation apparatus, including a processor, configured to implement the steps of the three-dimensional image multiphase segmentation method according to any one of the preceding items when executing the computer program stored in the memory.
Finally, an embodiment of the present invention provides a computer-readable storage medium, where a program for implementing a polyphase decomposition of a three-dimensional image is stored, and when the program is executed by a processor, the method implements the steps of the polyphase decomposition method of the three-dimensional image.
The technical scheme provided by the application has the advantages that the multi-phase segmentation model of the three-dimensional image is constructed by utilizing the multi-layer level set function, so that the image segmentation module realizes the curved surface evolution of n layers of level sets of the multi-layer level set function in the image to segment the image, and the parameters of the multi-phase segmentation model of the three-dimensional image can be obtained by calculation only by calculating the extreme value of the level set function, thereby being beneficial to improving the multi-phase segmentation efficiency of the three-dimensional image; in addition, the region division module utilizes the Heaviside function to construct the characteristic function of each region, so that each pixel point of the three-dimensional image to be processed can be ensured to automatically fall into a unique region, and the multiphase segmentation accuracy and precision of the three-dimensional image are ensured.
In addition, the embodiment of the invention also provides a corresponding implementation method, equipment and a computer readable storage medium for a computer device for implementing the three-dimensional image multi-phase segmentation, so that the device has higher practicability and feasibility, and the method, the equipment and the computer readable storage medium have corresponding advantages.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the related art, the drawings required to be used in the description of the embodiments or the related art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a block diagram of a computer apparatus for performing multi-phase segmentation of a three-dimensional image according to an embodiment of the present invention;
fig. 2 is a block diagram of another embodiment of a computer device for implementing polyphase segmentation of a three-dimensional image according to an embodiment of the present invention.
FIG. 3 is a flowchart illustrating a three-dimensional image multi-phase segmentation method according to an embodiment of the present invention;
FIG. 4 is a schematic flowchart of another method for implementing polyphase segmentation of a three-dimensional image according to an embodiment of the present invention;
FIG. 5 is a schematic diagram illustrating a three-dimensional image polyphase segmentation process according to an exemplary embodiment of the present disclosure;
FIG. 6 is a schematic diagram illustrating a three-dimensional image multi-phase segmentation process according to another exemplary embodiment of the present disclosure.
Detailed Description
In order that those skilled in the art will better understand the disclosure, the invention will be described in further detail with reference to the accompanying drawings and specific embodiments. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The terms "first," "second," "third," "fourth," and the like in the description and claims of this application and in the above-described drawings are used for distinguishing between different objects and not for describing a particular order. Furthermore, the terms "comprising" and "having," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements but may include other steps or elements not expressly listed.
Having described the technical solutions of the embodiments of the present invention, various non-limiting embodiments of the present application are described in detail below.
Referring to fig. 1, fig. 1 is a schematic structural diagram of a computer apparatus for implementing three-dimensional image multi-phase segmentation according to an embodiment of the present invention, where the embodiment of the present invention includes the following:
it can be understood that, the multi-phase image segmentation is a process of segmenting an original image into multiple phases according to a certain region segmentation scheme, and the problems to be solved by an effective multi-phase image segmentation model include expression of an energy functional, a partitioning strategy of a multi-phase region, and numerical calculation of an evolution equation. The selection of the region division scheme and the design of the region feature function are the core of image multi-phase segmentation, and in order to prevent overlapping and missing division between regions, the multi-phase image segmentation must solve a competition strategy between good regions. In view of this, a computer apparatus for performing polyphase segmentation of a three-dimensional image may comprise a region segmentation module 1 and an image segmentation module 2.
It will be appreciated that the level set is an efficient implicit representation of the evolution curve/surface, i.e. the segmentation boundary or surface is implicitly represented as a level set of a higher dimensional function, i.e. a level set function, evolving in the euler framework. Since the level set method automatically handles topological changes, it has the same expression for two-dimensional and three-dimensional and even higher-dimensional image segmentation. Therefore, in the process of realizing three-dimensional image segmentation, an energy model and a region division method which are similar to two-dimensional image segmentation can be adopted. The variation level set method can be used for solving the minimization problem of the energy functional E (C) for the evolution of the closed curve C, and the E (C) is modified into E (phi (x)) by introducing an embedding function phi (x) and a Heaviside function, and then a partial differential equation about phi (x) is obtained by using a variation method. Wherein, it is required to ensure that the level set function always keeps the feature | # | ═ 1 of the symbol distance function in the evolution process. C { (x) | phi (x) ═ C0}. C is a constant value C satisfying a function phi (x)0Is a function phi (x))A level set of (2). The function φ (x) is an embedded function of the curve C, called the level set function. When constant c0When 0, i.e. C { (x)|Phi (x) ═ 0 is called the zero level set of the level set function phi (x). Based on the method, the image can be segmented by utilizing the curved surface evolution of n layers of level sets of the multi-layer level set function in the image, and then the image can be segmented by the variation level set methodAnd (4) calculating an extreme value of the energy functional of the curved surface evolution by the method. The region division scheme based on the multilayer level set can be realized through a region division module 1, and the module can be used for setting a characteristic function based on a Heaviside function for each region to be divided so as to determine the boundary of each region, so that each pixel point of the three-dimensional image to be processed can automatically fall into a unique region. If f is the image intensity of a three-dimensional image to be processed,in the form of an area of an image,is the image boundary. N level sets ({ x | phi (x) } l) that can be operated by 1 continuous level set functioniI ═ 0,1,2,. n }), the image is divided into n regions Ω which do not overlap each other and are independent of each otheri,Wherein the multi-level set function φ (x) satisfies φ (x) e { l0,l1,…li…lnCan relax phi (x) convex to phi (x) e [ l ∈0,ln],(0=l0<l1<…<ln) To translate into a convex optimization problem. In one embodiment, the following feature function may be set for each region to be segmented:
that is, the region division module 1 can calculate the nth region Ω by using the normalized Heaviside function relationnSetting a characteristic function χn(φ), the regularized Heaviside function relationship is calculated as:
χn(φ)=H(φ-ln-1)(1-H(φ-ln)),ln-1<φ(x)≤ln,n=1,2,3……;
wherein φ (x) is e [ l0,ln],(0=l0<l1<…<ln) And is andn is the number of regions obtained by segmentation, H () is the Heaviside function, lnThe nth level set is a multi-level set function. Since phi (x) belongs to [ l ∈ ]0,ln],(0=l0<l1<…<ln) Therefore:
when xi(x) When 1, xi(x)…χi-1(x) 0, and xi+1(x)…χn(x)=0。
so that there areNaturally, each pixel point in the image is automatically ensured to fall into a unique area, and the phenomena of repetition and missing division in the image segmentation process are avoided.
In the present application, the image segmentation module 2 may be configured to segment the three-dimensional image to be processed into a plurality of disjoint regions by using a pre-constructed three-dimensional image multi-phase segmentation model. Accordingly, the image segmentation module 2 may comprise a model pre-construction sub-module 21 and a parameter calculation sub-module 12. The model pre-construction sub-module 21 is configured to construct a three-dimensional image multi-phase segmentation model by segmenting the three-dimensional image into a plurality of non-overlapping and independent regions based on the multi-level set of 1 continuous level set function. The parameter calculation sub-module 22 is configured to calculate parameter values of the three-dimensional image multi-phase segmentation model through a plurality of iterations by using an energy minimization method. And (3) carrying out multiple iterations by using an energy minimization method to determine each parameter value which enables the energy of the three-dimensional image multiphase segmentation model to be minimum, wherein the level set function phi (x) is mainly obtained by calculation.
As a preferred embodiment, the model pre-construction sub-module 12 may construct a three-dimensional image multi-phase segmentation model based on a multi-phase image segmentation variation level set relation calculation formula, which may be:
in the formula, f is the image intensity of the three-dimensional image to be processed, phi is a multi-layer level set function, and chii(phi) is the characteristic function of the ith region, H () is the Heaviside function, ui=(u1,u2,…,un) For piecewise constant values of the image intensity of the three-dimensional image to be processed in the respective region, lnFor the nth level set, gamma, of a multi-level set functioniIs an edge term-Ω|▽H(φ-li) A weight parameter of | dx, and γiGreater than 0, n is the number of the regions obtained by division, and the integral multipleΩ|▽H(φ-li) | dx is a rule term, also called edge term of region Ω i, whose value is region ΩiArea of edge curve.
Wherein, a Lipschitz continuous level set function can be adopted, and the above formula satisfies the constraint condition phi (x) is in the middle of l0,ln],(0=l0<l1<…<ln) And simultaneously satisfies | (φ |) | (1) constraint condition, liE.g. Z, to ensure that the level set function always maintains the characteristics of the symbol distance function in the evolution process. In the formula ui=(u1,u2,…,un) For a piecewise constant value of f in the region Ω, the estimated equation may be:
in the technical scheme provided by the embodiment of the invention, a multi-layer level set function is utilized to construct a three-dimensional image multiphase segmentation model, so that the image segmentation module realizes the curved surface evolution of n layers of level sets of the multi-layer level set function in an image to segment the image, and the parameters of the three-dimensional image multiphase segmentation model can be obtained by calculating only the extreme value of the level set function, thereby being beneficial to improving the three-dimensional image multiphase segmentation efficiency; in addition, the region division module constructs each region characteristic function by using the Heaviside function, so that each pixel point of the three-dimensional image to be processed can be ensured to automatically fall into a unique region, and the multi-phase segmentation precision of the three-dimensional image is ensured.
As a preferred embodiment, the energy functional can be expressed as two parts, namely a data item and a rule item, and based on the generalized characteristic function based on regularization Heaviside function as region division, the energy functional can be subjected to energy minimization calculation by using a Split-Bregman projection method. Based on this, the parameter calculation sub-module 21 may include:
an energy functional computing unit, configured to compute an energy functional of the multi-layer level set function based on an energy functional equivalence relation computational expression, where the energy functional equivalence relation computational expression may be:
in the formula, f is the image intensity of the three-dimensional image to be processed, phi is a multi-layer level set function, and chii(phi) is a characteristic function of the ith region, ui=(u1,u2,…,un) For piecewise constant values of the image intensity of the three-dimensional image to be processed in the respective region, lnIs the nth level set of the multi-level set function, n is the number of the divided areas, gammaiδ () is a Dirac function, being a weight parameter;
and the level set function calculation unit is used for calculating the extreme value of the energy functional by utilizing a Split-Bregman projection algorithm and an alternative optimization method.
The specific implementation process of the energy functional computing unit may include:
regularized Heaviside and Dirac functions may be employed to achieve functional representation of region partitioning. I.e. when ε → 0, Hε(φ) → H (φ) (ε tends to 0, HεPhi is approximated by the Heaviside function H (phi)), epsilon being a positive number with a small value. The calculation relationship may be:
for the sake of writing simplicity, the original symbolic representation is still used. The edge term can be equivalently expressed as:
∫Ω|▽H(φ-li)|dx=∫|▽φ|δ(φ-li)dx。
thus, the energy functional of the multi-level set function may be:
the process of solving the extremum for the energy functional using the efficient and stable Split-Bregman projection algorithm may include:
introducing an auxiliary variable w and a Bregman iteration parameterThe energy functional may be transformed into a variational model iteration format as follows:
wherein the content of the first and second substances,theta (theta > 0) is a punishment parameter, and an alternating optimization method is adopted to respectively obtain an Euler Lagrange equation about phi and an Euler Lagrange equation about phiThe generalized soft threshold formula (2) is calculated according to the following relationship:
for satisfying the constraint phi (x) epsilon [ l0,ln]A constraint term may be added to the function phi (x) as:
φk+1=min(max(l0,φk+1),ln)。 (3)
the iterative algorithm steps of the present application may be as follows:
2) estimate ui,i=1,2,…,n;;
5) If it satisfiesStopping iteration; if not, the iteration is continued, namely k is k +1, and the step 2) is executed.
It can be understood that, since the original data is noisy, that is, the to-be-processed three-dimensional image is noisy, the obtained segmentation structure diagram naturally contains noise, and in order to improve the quality of each obtained segmented image, after the to-be-processed three-dimensional image is successfully subjected to multi-phase segmentation, the segmentation result diagram of the to-be-processed three-dimensional image can be subjected to geometric denoising processing by using the image denoising module 3 to obtain a plurality of smooth sub-images. Any image denoising processing method can be adopted, the method is not limited in this application, and the corresponding denoising implementation process can refer to the implementation process recorded in the related technology, and is not described herein again.
The embodiment of the invention also provides a corresponding implementation method for a computer device for implementing the three-dimensional image multiphase segmentation, so that the computer device is more feasible. In the following, the three-dimensional image multi-phase segmentation method provided by the embodiment of the present invention is introduced, and the three-dimensional image multi-phase segmentation method described below and the computer device for implementing the three-dimensional image multi-phase segmentation described above may be referred to correspondingly.
Referring to fig. 3, fig. 3 is a schematic flowchart of a method for implementing multi-phase segmentation of a three-dimensional image according to an embodiment of the present invention, where the embodiment of the present invention includes the following:
s301: and setting a characteristic function based on the Heaviside function for each region to be segmented of the three-dimensional image to be processed to determine the boundary of each region, so as to ensure that each pixel point of the three-dimensional image to be processed automatically falls into a unique region.
The three-dimensional image multiphase segmentation model is obtained by segmenting a three-dimensional image into a plurality of non-overlapping and independent areas based on a multilayer level set of 1 continuous level set function and calculating various parameter values through multiple iterations by using an energy minimization method.
S302: and based on the characteristic function of each region, segmenting the three-dimensional image to be processed into a plurality of disjoint regions by utilizing a pre-constructed three-dimensional image multiphase segmentation model.
Based on the above embodiments, please refer to fig. 4, which may further include:
s303: and performing geometric denoising processing on the segmentation result graph of the three-dimensional image to be processed to obtain a plurality of smooth sub-images.
The implementation method of each step of the three-dimensional image multi-phase segmentation method according to the embodiment of the present invention can be according to the corresponding function of each functional module in the above-mentioned device embodiment, and the specific implementation process thereof can refer to the related description of the above-mentioned device embodiment, and is not described herein again.
Therefore, the embodiment of the invention efficiently and accurately realizes the multi-phase segmentation of the three-dimensional image.
In order to make the technical solutions to be protected by the present application more clearly understood by those skilled in the art, the present application further provides two schematic examples to illustrate the implementation process of the three-dimensional image multi-phase segmentation, please refer to fig. 5 and fig. 6, fig. 5 is a schematic view of mandible image sequence segmentation, where a is a mandible image under the effect of an initial level set, b is a graph of mandible image segmentation results after 3 iterations, and c is a graph of mandible image segmentation results after 10 iterations; fig. 6 is a schematic diagram of dental image sequence segmentation, where a is a mandible image under the action of an initial level set, b is a dental image segmentation result diagram iterated for 3 times, and c is a dental image segmentation result diagram iterated for 10 times. The image segmentation process may include:
the embodiment of the invention takes a real human body CT scanning image, such as a real human body lower jaw part CT scanning image sequence and a tooth part CT scanning image sequence, as a three-dimensional image to be processed. 3 level sets l using a Lipschitz level set function1=1,l 23 and l3The three-dimensional image to be processed is subjected to multiphase segmentation, and each parameter value can be set as epsilon=8、α=1、γ=0.09×2552、θ=0.3。
First, the level set function is initialized to the symbol distance function, and second, u is estimatedi(i is 1,2, …, n), and the following three calculation relations are calculated in sequence to obtain
Then updatedIf it isStopping the iteration and displaying the segmentation result, or notIf not, continuing the iteration, namely k is k +1, and repeating the steps until the iteration termination condition is met.
The embodiment of the present invention further provides a three-dimensional image multi-phase segmentation apparatus, which specifically includes:
a memory for storing a computer program;
a processor for executing a computer program to implement the steps of the three-dimensional image multi-phase segmentation method according to any one of the above embodiments.
The functions of the functional modules of the three-dimensional image multi-phase segmentation apparatus according to the embodiments of the present invention may be specifically implemented according to the method in the above method embodiments, and the specific implementation process may refer to the related description of the above method embodiments, which is not described herein again.
Therefore, the embodiment of the invention efficiently and accurately realizes the multi-phase segmentation of the three-dimensional image.
The embodiment of the present invention further provides a computer-readable storage medium, which stores a program for implementing the multiphase segmentation of the three-dimensional image, and the program for implementing the multiphase segmentation of the three-dimensional image is executed by a processor, and the steps of the multiphase segmentation method of the three-dimensional image according to any one of the above embodiments are provided.
The functions of the functional modules of the computer-readable storage medium according to the embodiment of the present invention may be specifically implemented according to the method in the foregoing method embodiment, and the specific implementation process may refer to the related description of the foregoing method embodiment, which is not described herein again.
Therefore, the embodiment of the invention efficiently and accurately realizes the multi-phase segmentation of the three-dimensional image.
The embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same or similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The present invention provides a computer device, a three-dimensional segmentation method, a three-dimensional segmentation device, and a computer-readable storage medium for implementing multi-phase segmentation of a three-dimensional image. The principles and embodiments of the present invention are explained herein using specific examples, which are presented only to assist in understanding the method and its core concepts. It should be noted that, for those skilled in the art, it is possible to make various improvements and modifications to the present invention without departing from the principle of the present invention, and those improvements and modifications also fall within the scope of the claims of the present invention.
Claims (8)
1. A computer apparatus for performing a polyphase segmentation of a three-dimensional image, comprising:
the area division module is used for setting a characteristic function based on a Heaviside function for each area to be divided of the three-dimensional image to be processed so as to determine the boundary of each area, so that each pixel point of the three-dimensional image to be processed can automatically fall into a unique area;
the image segmentation module is used for segmenting the three-dimensional image to be processed into a plurality of disjoint areas by utilizing a pre-constructed three-dimensional image multiphase segmentation model;
the image segmentation module comprises a model pre-construction sub-module and a parameter calculation sub-module; the model pre-construction submodule is used for constructing the three-dimensional image multiphase segmentation model by segmenting the three-dimensional image into a plurality of mutually non-overlapped and respectively independent areas based on a multi-layer level set of 1 continuous level set function; the parameter calculation submodule is used for calculating various parameter values of the three-dimensional image multi-phase segmentation model through multiple iterations by using an energy minimization method;
the three-dimensional image is divided into a plurality of non-overlapping and independent areas based on the multi-layer level set of the 1 continuous level set function, and the three-dimensional image multiphase division model is constructed as follows: dividing the image into n independent areas which are not overlapped with each other through n level sets of 1 continuous level set function;
the region division module is used for calculating an nth region omega by utilizing a regularized Heaviside function relationnSetting a characteristic function χn(phi), the regularized Heaviside function relation calculation formula is as follows:
χn(φ)=H(φ-ln-1)(1-H(φ-ln)),ln-1<φ(x)≤ln,n=1,2,3……;
2. The computer device for performing multi-phase segmentation of three-dimensional images according to claim 1, wherein the model pre-construction sub-module is configured to construct the three-dimensional image multi-phase segmentation model based on a multi-phase image segmentation variance level set relation calculation formula:
In the formula, f isThe image intensity of the three-dimensional image to be processed is phi which is the continuous level set function and chii(phi) is the characteristic function of the ith region, H () is the Heaviside function, ui=(u1,u2,…,un) A constant value l for each region of the image intensity of the three-dimensional image to be processednFor the nth level set of the continuous level set function, gammaiIs an edge itemN is the number of divided regions.
3. The computer apparatus for implementing polyphase segmentation of three-dimensional images according to claim 1 or 2, further comprising an image denoising module; the image denoising module is used for carrying out geometric denoising processing on the segmentation result image of the three-dimensional image to be processed so as to obtain a plurality of smooth sub-images.
4. The computer apparatus for polyphase segmentation of three-dimensional images according to claim 3, wherein said parameter computation submodule comprises:
an energy functional computing unit, configured to compute an energy functional of the continuous level set function based on an energy functional equivalence relation computational expression, where the energy functional equivalence relation computational expression is:
wherein f is the image intensity of the three-dimensional image to be processed, phi is the continuous level set function, and chii(phi) is a characteristic function of the ith region, ui=(u1,u2,…,un) A constant value l for each region of the image intensity of the three-dimensional image to be processednIs the nth level set of the continuous level set function, n is the number of the divided regions, gammaiδ () is a Dirac function, being a weight parameter;
and the level set function calculation unit is used for calculating the extreme value of the energy functional by utilizing a Split-Bregman projection algorithm and an alternative optimization method.
5. A three-dimensional image multi-phase segmentation method is characterized by comprising the following steps:
setting a characteristic function based on a Heaviside function for each region to be segmented of the three-dimensional image to be processed to determine the boundary of each region so as to ensure that each pixel point of the three-dimensional image to be processed automatically falls into a unique region;
based on the characteristic function of each region, segmenting the three-dimensional image to be processed into a plurality of disjoint regions by utilizing a pre-constructed three-dimensional image multiphase segmentation model;
the three-dimensional image multi-phase segmentation model is a multi-layer level set based on 1 continuous level set function, a three-dimensional image is segmented into a plurality of non-overlapping and independent areas, and each parameter value is calculated through multiple iterations by using an energy minimization method;
the multi-layer level set based on 1 continuous level set function divides the three-dimensional image into a plurality of non-overlapping and independent areas, and the three-dimensional image multi-phase division model is constructed as follows: dividing the image into n independent areas which are not overlapped with each other through n level sets of 1 continuous level set function;
the three-dimensional image multiphase segmentation model is a multilayer level set based on 1 continuous level set function, divides a three-dimensional image into a plurality of non-overlapping and independent areas, and calculates each parameter value through multiple iterations by using an energy minimization method, and comprises the following steps:
constructing the three-dimensional image multi-phase segmentation model based on a multi-phase image segmentation variation level set relational calculation formula, wherein the multi-phase image segmentation variation level set relational calculation formula is as follows:
Wherein f is the image intensity of the three-dimensional image to be processed, phi is the continuous level set function, and chii(phi) is the characteristic function of the ith region, H () is the Heaviside function, ui=(u1,u2,…,un) A constant value l for each region of the image intensity of the three-dimensional image to be processednFor the nth level set of the continuous level set function, gammaiIs an edge itemN is the number of divided regions.
6. The three-dimensional image polyphase segmentation method according to claim 5, wherein after segmenting the three-dimensional image to be processed into a plurality of disjoint regions by using the pre-constructed three-dimensional image polyphase segmentation model, the method further comprises:
and carrying out geometric denoising processing on the segmentation result graph of the three-dimensional image to be processed to obtain a plurality of smooth sub-images.
7. A three-dimensional image polyphase segmentation apparatus comprising a processor for implementing the steps of the three-dimensional image polyphase segmentation method according to claim 5 or 6 when executing a computer program stored in a memory.
8. A computer-readable storage medium, in which a program for performing a polyphase decomposition of a three-dimensional image is stored, which program, when being executed by a processor, performs the steps of the polyphase decomposition method of a three-dimensional image according to claim 5 or 6.
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 CN110415262A (en) | 2019-11-05 |
CN110415262B true 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) |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104835168B (en) * | 2015-05-15 | 2017-08-18 | 东华理工大学 | Quick multiphase image dividing method based on global convex optimization Variation Model |
-
2019
- 2019-08-07 CN CN201910726082.0A patent/CN110415262B/en active Active
Non-Patent Citations (3)
Title |
---|
基于一个水平集函数的多相图像分割方法;赵瑞雪;《计算机工程》;20180430;第317-321页 * |
基于变分水平集方法的通用模型和快速投影算法研究;端金鸣;《中国优秀硕士学位论文全文数据库信息科技辑》;20141115;第6-54页 * |
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》.2014, * |
Also Published As
Publication number | Publication date |
---|---|
CN110415262A (en) | 2019-11-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Li et al. | Fast guided global interpolation for depth and motion | |
US11514573B2 (en) | Estimating object thickness with neural networks | |
Trinh et al. | Novel example-based method for super-resolution and denoising of medical images | |
Miklos et al. | Discrete scale axis representations for 3D geometry | |
US9561004B2 (en) | Automated 3-D orthopedic assessments | |
Pahr et al. | From high-resolution CT data to finite element models: development of an integrated modular framework | |
Dong et al. | Image reconstruction with locally adaptive sparsity and nonlocal robust regularization | |
Ghosal et al. | A fast scalable algorithm for discontinuous optical flow estimation | |
JP4885138B2 (en) | Method and system for motion correction in a sequence of images | |
Foare et al. | Semi-linearized proximal alternating minimization for a discrete Mumford–Shah model | |
US20140161352A1 (en) | Iterative method for determining a two-dimensional or three-dimensional image on the basis of signals arising from x-ray tomography | |
US8929636B2 (en) | Method and system for image segmentation | |
Cho et al. | Effective five directional partial derivatives-based image smoothing and a parallel structure design | |
Wang et al. | Fast and accurate smoothing method using a modified Allen–Cahn equation | |
Guo et al. | Box spline wavelet frames for image edge analysis | |
Zhao et al. | Direct multiphase mesh generation from 3D images using anisotropic mesh adaptation and a redistancing equation | |
CN110415262B (en) | Computer device for realizing three-dimensional image multiphase segmentation and three-dimensional segmentation method and equipment | |
US20180040152A1 (en) | Method for real-time deformable fusion of a source multi-dimensional image and a target multi-dimensional image of an object | |
Cho et al. | Low-complexity topological derivative-based segmentation | |
Sarfraz | Generating outlines of generic shapes by mining feature points | |
CN115100115A (en) | Organ image segmentation method, device, equipment and storage medium | |
Popuri et al. | Fast FEM-based non-rigid registration | |
JP7316771B2 (en) | Learning device, parameter creation method, neural network, and information processing device using the same | |
Ye et al. | A sparsity-promoting image decomposition model for depth recovery | |
Zisler et al. | Non-binary discrete tomography by continuous non-convex optimization |
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 |