CN108305282A - A kind of method for registering images and system based on hybrid rice algorithm - Google Patents
A kind of method for registering images and system based on hybrid rice algorithm Download PDFInfo
- Publication number
- CN108305282A CN108305282A CN201810092708.2A CN201810092708A CN108305282A CN 108305282 A CN108305282 A CN 108305282A CN 201810092708 A CN201810092708 A CN 201810092708A CN 108305282 A CN108305282 A CN 108305282A
- Authority
- CN
- China
- Prior art keywords
- individual
- rice
- image
- son
- registration
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/37—Determination of transform parameters for the alignment of images, i.e. image registration using transform domain methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20048—Transform domain processing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Genetics & Genomics (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Physiology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of method for registering images and system based on hybrid rice algorithm.This method includes acquiring the image of object, obtains reference picture and multiple images subject to registration;The correct image of reference picture representation space coordinate;Each image subject to registration is pre-processed, uses bilinear interpolation method that each image procossing subject to registration to have the image of equal resolution with reference picture, is obtained the corresponding pretreated image of each image subject to registration;The first space transform models of each pretreated image are established by hybrid rice algorithm;Second space transformation model is obtained according to the first space transform models of all pretreated images;Affine transformation is carried out to the coordinate of pretreated image by second space transformation model, obtains the registration image for corresponding to the image subject to registration.This method and system can improve the precision and efficiency of multiresolution image registration.
Description
Technical field
The present invention relates to image registration field, more particularly to a kind of method for registering images based on hybrid rice algorithm and
System.
Background technology
With the rapid development of Modern digital image technology and continuing to bring out for novel sensor, people obtain the energy of image
Power is continuously improved, and image procossing is widely applied to many aspects of modern medical service and industrial circle.Image registration is at image
The basic fundamental of reason, image registration be by under different time, different sensors (imaging device) or different condition (weather, illumination,
Camera position and angle etc.) two width that obtain or the multiple image process that is matched, be superimposed.Different sensors or same sensing
Spatially often there is difference in the image that device is obtained in different time, different points of view, can make these by image registration
Image information reaches the completely the same of space geometry position in the spatial domain, reach to target image there are one more comprehensively, clearly,
The purpose of accurate understanding and cognition, in order to provide more fully information.
When differences in resolution is very big, image subject to registration is difficult to accurately match with reference picture on spatial position.It solves
A kind of preferred embodiment of the problem be using a kind of optimization algorithm come find similarity measure it is optimal when transformation parameter, however mesh
All there is defects to some extent for the preceding algorithm for solving problems.Match for example, carrying out image using particle swarm optimization algorithm
Accurate disadvantage has the dynamic regulation for being a lack of speed, is easily trapped into local optimum, causes convergence precision low and is not easy to restrain;Again
Such as, initial point must be conscientiously established using powell algorithms, otherwise it cannot be guaranteed that converging to globally optimal solution.
Invention content
The object of the present invention is to provide a kind of method for registering images and system based on hybrid rice algorithm is more to improve
The precision and efficiency of class resolution ratio image registration.
To achieve the above object, the present invention provides following schemes:
A kind of method for registering images based on hybrid rice algorithm, the method includes:
The image for acquiring object, obtains reference picture and multiple images subject to registration;The reference picture representation space is sat
Mark correct image;
Each image subject to registration is pre-processed, use bilinear interpolation method by each image procossing subject to registration for
The reference picture has the image of equal resolution, obtains the corresponding pretreated image of each image subject to registration;
The first space transform models of each pretreated image are established by hybrid rice algorithm;
Second space transformation model is obtained according to the first space transform models of all pretreated images;
Affine transformation is carried out to the coordinate of the pretreated image by the second space transformation model, is obtained pair
Should image subject to registration registration image.
Optionally, the first spatial alternation mould that each pretreated image is established by hybrid rice algorithm
Type specifically includes:
Calculate the fitness function of rice individual;The rice individual representation space transformation model, the fitness function
Indicate the similarity value of the reference picture and the pretreated image;
The rice individual is divided into maintainer, sterile line and restorer according to the fitness function;
Rice individual in the maintainer is hybridized with the rice individual in the sterile line, it is optimal to obtain first
Sub- individual;
Rice individual in the restorer is selfed, the second optimal son individual is obtained;
It is the to choose the higher individual of fitness function in the described first optimal son individual and the described second optimal son individual
Three optimal son individuals, the optimal son individual of third indicate first space transform models.
Optionally, the rice individual by the maintainer hybridizes with the rice individual in the sterile line,
The first optimal son individual is obtained, is specifically included:
Rice individual in the maintainer is subjected to randomer hybridization with the rice individual in the sterile line, is obtained random
Hybrid individual;The randomer hybridization indicates any rice individual in the maintainer and any rice in the sterile line
Body is hybridized;The male parent of the randomer hybridization individual is the rice individual in the corresponding sterile line, the randomer hybridization
The female parent of individual is the rice individual in the corresponding maintainer;
It is the to choose the higher individual of fitness function in the randomer hybridization individual and the randomer hybridization individual male parent
One son individual;
The highest first son individual of fitness function is chosen as the first optimal son individual.
Optionally, the rice individual by the maintainer hybridizes with the rice individual in the sterile line,
The first optimal son individual is obtained, is specifically included:
Rice individual in the maintainer is subjected to corresponding hybridization with the rice individual in the sterile line, is corresponded to
Hybrid individual;The corresponding hybridization indicates the rice individual in the maintainer and the rice in the corresponding sterile line
Body is hybridized;The male parent of the corresponding hybrid individual is the rice individual in the corresponding sterile line, the corresponding hybridization
The female parent of individual is the rice individual in the corresponding maintainer;
It is the to choose the higher individual of fitness function in the corresponding hybrid individual and the corresponding hybrid individual male parent
One son individual;
The highest first son individual of fitness function is chosen as the first optimal son individual.
Optionally, the rice individual by the restorer is selfed;The second optimal son individual is obtained, it is specific to wrap
It includes:
Rice individual in the restorer is selfed, selfing individual is obtained;
It is the second son to choose the higher individual of fitness function in the rice individual handed in individual and the restorer
Individual;
The highest second son individual of fitness function is chosen as the second optimal son individual.
Optionally, first space transform models according to all pretreated images obtain second space change
Mold changing type;It specifically includes:
The parameter of the first space transform models of all pretreated images is obtained, the parameter is the third
The gene of optimal son individual;
The parameter is obtained into second space transformation model as product.
The present invention also provides a kind of figure registration system based on hybrid rice algorithm, the system comprises:
Acquisition module, the image for acquiring object obtain reference picture and multiple images subject to registration;The reference chart
As the correct image of representation space coordinate;
Processing module described waits matching using bilinear interpolation method for pre-processing each image subject to registration by each
Quasi- image procossing is the image for having equal resolution with the reference picture, obtains the corresponding pre- place of each image subject to registration
Image after reason;
First modeling module, the first space for establishing each pretreated image by hybrid rice algorithm
Transformation model;
Second modeling module, for obtaining second according to the first space transform models of all pretreated images
Space transform models;
Registration module, for being imitated the coordinate of the pretreated image by the second space transformation model
Transformation is penetrated, the registration image for corresponding to the image subject to registration is obtained.
Optionally, first modeling module specifically includes:
Computing unit, the fitness function for calculating rice individual;The rice individual representation space transformation model, institute
State the similarity value that fitness function indicates the reference picture and the pretreated image;
Taxon, for the rice individual to be divided into maintainer, sterile line and extensive according to the fitness function
Multiple system;
Hybridised units, it is miscellaneous for carrying out the rice individual in the maintainer with the rice individual in the sterile line
It hands over, obtains the first optimal son individual;
It is selfed unit, for the rice individual in the restorer to be selfed, obtains the second optimal son individual;
Selection unit, for choose described first it is optimal son individual and described second it is optimal son individual in fitness function compared with
High individual is the optimal son individual of third, and the optimal son individual of third indicates the spatial model.
Optionally, the hybridised units specifically include:
Randomer hybridization subelement, for by the rice individual in rice individual and the sterile line in the maintainer into
Row randomer hybridization obtains randomer hybridization individual;The randomer hybridization indicate any rice individual in the maintainer with it is described
Any rice individual in sterile line is hybridized;The male parent of the randomer hybridization individual is the water in the corresponding sterile line
Rice individual, the female parent of the randomer hybridization individual are the rice individual in the corresponding maintainer;
First chooses subelement, for choosing fitness in the randomer hybridization individual and the randomer hybridization individual male parent
The higher individual of function is the first son individual;
First chooses subelement, for choosing the highest first son individual of fitness function as the first optimal son individual.
Optionally, the hybridised units specifically include:
Corresponding hybridization subelement, for by rice individual and the rice individual in the sterile line in the maintainer into
The corresponding hybridization of row, obtains corresponding hybrid individual;The corresponding hybridization indicate rice individual in the maintainer with it is corresponding
Rice individual in the sterile line is hybridized;The male parent of the corresponding hybrid individual is the water in the corresponding sterile line
Rice individual, the female parent of the corresponding hybrid individual are the rice individual in the corresponding maintainer;
Third chooses subelement, for choosing fitness in the corresponding hybrid individual and the corresponding hybrid individual male parent
The higher individual of function is the first son individual;
4th chooses subelement, for choosing the highest first son individual of fitness function as the first optimal son individual.
According to specific embodiment provided by the invention, the invention discloses following technique effects:The present invention is using hybridization water
Rice algorithm establishes space transform models, and hybrid rice algorithm can carry out global search, is not easy to be absorbed in local optimum, optimizing ability
By force, fast convergence rate.And hybrid rice algorithm computation complexity is low, calculating speed is fast, there is the ability for jumping out locally optimal solution.Cause
This quickly can accurately find optimal solution using hybrid rice algorithm, that is, establish space transform models, to pass through this space
Transformation model can improve the precision and efficiency of multiresolution image registration.
Description of the drawings
It in order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, below will be to institute in embodiment
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the present invention
Example, for those of ordinary skill in the art, without having to pay creative labor, can also be according to these attached drawings
Obtain other attached drawings.
Fig. 1 is a kind of flow chart of the method for registering images based on hybrid rice algorithm provided in an embodiment of the present invention;
Fig. 2 is the method flow diagram provided in an embodiment of the present invention that the first space transform models are established by hybrid rice algorithm;
Fig. 3 is a kind of block diagram of the figure registration system based on hybrid rice algorithm provided in an embodiment of the present invention;
Fig. 4 is the block diagram of the first modeling module provided in an embodiment of the present invention.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation describes, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The object of the present invention is to provide a kind of method for registering images and system based on hybrid rice algorithm is more to improve
The precision and efficiency of class resolution ratio image registration.
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, below in conjunction with the accompanying drawings and specific real
Applying mode, the present invention is described in further detail.
As shown in Figure 1, a kind of method for registering images based on hybrid rice algorithm includes the following steps:
Step 101:The image for acquiring object, obtains reference picture and multiple images subject to registration;The reference picture table
Show the correct image of space coordinate.
Step 102:Each image subject to registration is pre-processed, using bilinear interpolation method by each image subject to registration
It is corresponding pretreated to obtain each image subject to registration to have the image of equal resolution with the reference picture for processing
Image.
Step 103:The first space transform models of each pretreated image are established by hybrid rice algorithm.
As shown in Fig. 2, establishing the first spatial alternation mould of each pretreated image by hybrid rice algorithm
The method of type specifically includes following steps:
Step 1031:Calculate the fitness function of rice individual;The rice individual representation space transformation model, it is described suitable
The similarity value of reference picture described in response function representation and the pretreated image.
Fitness function indicates the similarity value of the reference picture and the pretreated image, uses mutual trust herein
Breath correlated measure weighs the quality of registration result as object function, and calculation formula is as follows:
I thereinRFor reference picture, IFFor image subject to registration, MI (IR;IF) be reference picture and image subject to registration phase
Like angle value mutual information, mutual information is a kind of typical Kullback-Leibler divergences, i.e. its form can be shown as:
MI(IR;IF)=DKL(PRF||PR*PF),
DKLAs Kullback-Leibler divergences, wherein PRFFor the joint probability of reference image R and image F subject to registration
Density Distribution, PRF(x, y) is obtained by the joint intensity profile histogram calculation of two images, and PRAnd PFIt is respectively then reference chart
As the distribution of the marginal probability density of R and image F subject to registration, PR(x) and PF(y) distinguished by the intensity profile histogram of image itself
It is calculated:
JH is the joint intensity profile histogram of two images;CijIt is each element, C in joint histogramijIndicate every
C is shared on one position (i, j)ijGroup gray scale is i to meeting the gray value in piece image, and in the second width image
Gray value be j.
Step 1032:The rice individual is divided into maintainer, sterile line and restorer according to the fitness function.
The ratio that wherein maintainer, sterile line account for rice individual N is a%, and quantity is A=N × a/100, then restorer
The ratio for accounting for group is (100-2a) %, quantity N-2A.
Step 1033:Rice individual in the maintainer is hybridized with the rice individual in the sterile line, is obtained
To the first optimal son individual.
For breeding each time, the number that hybrid process carries out is identical as the individual amount of sterile line.Hybridize each time, it will
An individual is respectively chosen from sterile line and maintainer as male parent female parent, selection mode can randomly select can also be by one by one
Corresponding mode is chosen.The mode of hybridization is that male parent and the gene of maternal corresponding position are added capable recombination according to random weight heavy phase
And obtain an individual for possessing new gene.It calculates the fitness of new individual, and is criterion by itself and his father using greedy algorithm
Sterile line individual comparison in this female parent retains fitness preferably individual to the next generation.
(1) randomer hybridization
Rice individual in the maintainer is subjected to randomer hybridization with the rice individual in the sterile line, is obtained random
Hybrid individual;The randomer hybridization indicates any rice individual in the maintainer and any rice in the sterile line
Body is hybridized;The male parent of the randomer hybridization individual is the rice individual in the corresponding sterile line, the randomer hybridization
The female parent of individual is the rice individual in the corresponding maintainer;
It is the to choose the higher individual of fitness function in the randomer hybridization individual and the randomer hybridization individual male parent
One son individual;
The highest first son individual of fitness function is chosen as the first optimal son individual.
Randomer hybridization formula is as follows:
In formulaIndicate the jth Wiki for the new individual that kth time hybridization generates in the wheel breeding process because of r1,r2For
Random number between [- 1,1], and r1+r2≠0.A, b are derived from { 1,2 ..., A } at random,Indicate a-th of individual in sterile line
Jth Wiki because,Indicate maintainer in b-th individual jth Wiki because.The gene of the new individual of generation per one-dimensional
All hybridize to obtain with random ratio by the random individual in sterile line and maintainer.
(2) corresponding hybridization
Rice individual in the maintainer is subjected to corresponding hybridization with the rice individual in the sterile line, is corresponded to
Hybrid individual;The corresponding hybridization indicates the rice individual in the maintainer and the rice in the corresponding sterile line
Body is hybridized;The male parent of the corresponding hybrid individual is the rice individual in the corresponding sterile line, the corresponding hybridization
The female parent of individual is the rice individual in the corresponding maintainer;
It is the to choose the higher individual of fitness function in the corresponding hybrid individual and the corresponding hybrid individual male parent
One son individual;
The highest first son individual of fitness function is chosen as the first optimal son individual.
Corresponding crossing formula is as follows:
A=b=k in formula,Indicate sterile line in a-th individual jth Wiki because,Indicate the b in maintainer
Individual jth Wiki because.The gene of the new individual of generation per it is one-dimensional all by k-th of sterile line individual in maintainer
K-th of individual hybridizes to obtain with random ratio.
Step 1034:Rice individual in the restorer is selfed, the second optimal son individual is obtained.
Rice individual in the restorer is selfed, selfing individual is obtained;
It is the second son to choose the higher individual of fitness function in the rice individual handed in individual and the restorer
Individual;
The highest second son individual of fitness function is chosen as the second optimal son individual.
In breeding process, the number for being selfed progress is identical as the individual amount of restorer.It is selfed each time, participates in selfing
Gene on each position of restorer individual all can be towards current optimal solution close to a random quantity.Calculate the adaptation of new individual
It spends and according to greedy algorithm compared with the restorer individual before selfing, foundation is preferably saved in the next generation.If being saved in down
The individual of a generation, which is the selfing number of individual so individual before being selfed, will add 1.It is if being saved in follow-on individual
It is selfed the new individual generated, if new individual is better than current optimum individual, number is selfed and is set as 0, otherwise keeps it certainly
Hand over number constant.If the selfing number of some restorer individual has reached limited number of times maxTime, then in next round breeding
It will not participate in selfing process instead reset process in journey.
new_Xk=XSk+rand(0,1)(Xbest-XSr)
New_X in formulakIndicate the new individual that kth time selfing generates in the wheel breeding process, XSkIndicate the kth in restorer
Individual, XbestIndicate the optimum individual currently found, XSrFor r-th in restorer individual, wherein r random values in
{ 1,2 ..., N-2A }.
When restorer individual reaches the selfing number upper limit, reset.Reset process will at random generate in solution space
One group of gene, and this group of gene is added on the gene for the individual for participating in resetting, while its selfing number will be arranged to 0.
Wherein j ∈ { 1,2 ..., D-1, D }, in formulaIndicate that kth time selfing generates new in the wheel breeding process
The jth Wiki of individual is because of maxxj,minxjRespectively indicate search space jth Wiki because maxima and minima.Indicate extensive
In multiple system the jth Wiki of k-th of individual because.Repeat more wheel breedings according to above procedure, until operation to maximum breeding generation
Number maxIteration stops less than when optimizing error.
Step 1035:It is higher to choose fitness function in the described first optimal son individual and the described second optimal son individual
Individual is the optimal son individual of third, and the optimal son individual of third indicates the spatial model.
The gene of the optimal son individual of third is the optimized parameter of space transform models, and the can be set up using optimized parameter
One space transform models establish the first space transform models using affine transformation herein, and the first space transform models are joined by 5
Number is constituted, including X-direction translational movement dX, Y-direction translational movement dy, X-direction scaling SX, Y-direction scaling SyWith rotation angle θ.Each
The matrix P that the position of rice individual in space can be tieed up by one 1 × 5 is indicated:
P=dx,dy,Sx,Sy, θ,
The general expression of the first space transform models matrix that 5 parameters are constituted is:
Step 104:Second space transformation is obtained according to the first space transform models of all pretreated images
Model.
The parameter of the first space transform models of all pretreated images is obtained, the parameter is the third
The gene of optimal son individual;
The parameter is obtained into second space transformation model as product.
Step 105:Affine change is carried out to the coordinate of the pretreated image by the second space transformation model
It changes, obtains the registration image for corresponding to the image subject to registration.
According to specific embodiment provided by the invention, the invention discloses following technique effects:Using hybrid rice algorithm
Space transform models are established, hybrid rice algorithm can carry out global search, and it is strong to be not easy to be absorbed in local optimum optimizing ability, convergence
Speed is fast.And hybrid rice algorithm computation complexity is low, calculating speed is fast, there is the ability for jumping out locally optimal solution.Therefore it uses
Hybrid rice algorithm quickly can accurately find optimal solution, that is, establish space transform models, to pass through this spatial alternation mould
Type can improve the precision and efficiency of multiresolution image registration.
As shown in figure 3, the present invention also provides a kind of figure registration systems based on hybrid rice algorithm.The system packet
It includes:
Acquisition module 301, the image for acquiring object obtain reference picture and multiple images subject to registration;The ginseng
Examine the correct image of graphical representation space coordinate.
Processing module 302 described is waited for using bilinear interpolation method by each for being pre-processed to each image subject to registration
It is the image for having equal resolution with the reference picture to be registrated image procossing, and it is corresponding pre- to obtain each image subject to registration
Treated image.
First modeling module 303, for establishing the first of each pretreated image by hybrid rice algorithm
Space transform models.
As shown in figure 4, the first modeling module 303 includes:
Computing unit 3031, the fitness function for calculating rice individual;The rice individual representation space becomes mold changing
Type, the fitness function indicate the similarity value of the reference picture and the pretreated image;
Taxon 3032, for according to the fitness function by the rice individual be divided into maintainer, sterile line with
And restorer;
Hybridised units 3033, for carrying out the rice individual in the maintainer with the rice individual in the sterile line
Hybridization obtains the first optimal son individual.
Specifically hybridised units 3033 include:
Randomer hybridization subelement, for by the rice individual in rice individual and the sterile line in the maintainer into
Row randomer hybridization obtains randomer hybridization individual;The randomer hybridization indicate any rice individual in the maintainer with it is described
Any rice individual in sterile line is hybridized;The male parent of the randomer hybridization individual is the water in the corresponding sterile line
Rice individual, the female parent of the randomer hybridization individual are the rice individual in the corresponding maintainer;
First chooses subelement, for choosing fitness in the randomer hybridization individual and the randomer hybridization individual male parent
The higher individual of function is the first son individual;
Second chooses subelement, for choosing the highest first son individual of fitness function as the first optimal son individual.
Corresponding hybridization subelement, for by rice individual and the rice individual in the sterile line in the maintainer into
The corresponding hybridization of row, obtains corresponding hybrid individual;The corresponding hybridization indicate rice individual in the maintainer with it is corresponding
Rice individual in the sterile line is hybridized;The male parent of the corresponding hybrid individual is the water in the corresponding sterile line
Rice individual, the female parent of the corresponding hybrid individual are the rice individual in the corresponding maintainer;
Third chooses subelement, for choosing fitness in the corresponding hybrid individual and the corresponding hybrid individual male parent
The higher individual of function is the first son individual;
4th chooses subelement, for choosing the highest first son individual of fitness function as the first optimal son individual.
It is selfed unit 3034, for the rice individual in the restorer to be selfed, obtains the second optimal son individual.
Specifically, selfing unit 3034 includes:
It is selfed subelement, for the rice individual in the restorer to be selfed, obtains selfing individual;
5th chooses subelement, for choosing fitness function in the rice individual handed in individual and the restorer
Higher individual is the second son individual;
6th chooses subelement, for choosing the highest second son individual of fitness function as the first optimal son individual.
Selection unit 3035, for choosing fitness letter in the described first optimal son individual and the described second optimal son individual
The higher individual of number is the optimal son individual of third, and the optimal son individual of third indicates the spatial model.
Second modeling module 304, for being obtained according to the first space transform models of all pretreated images
Second space transformation model.
Specifically, the second modeling module 304 includes:
Parameter acquisition module, the parameter of the first space transform models for obtaining all pretreated images,
The parameter is the gene of the optimal son individual of the third;
Product module, for the parameter to be obtained second space transformation model as product.
Registration module 305, for by the second space transformation model to the coordinate of the pretreated image into
Row affine transformation obtains the registration image for corresponding to the image subject to registration.
Above system establishes space transform models using hybrid rice algorithm, and hybrid rice algorithm can carry out the overall situation and search
Rope, it is strong to be not easy to be absorbed in local optimum optimizing ability, fast convergence rate.And hybrid rice algorithm computation complexity is low, calculating speed
Soon, there is the ability for jumping out locally optimal solution.Therefore optimal solution quickly can accurately be found using hybrid rice algorithm, that is, established
Space transform models, to which the precision and efficiency of multiresolution image registration can be improved by this space transform models.
Each embodiment is described by the way of progressive in this specification, the highlights of each of the examples are with other
The difference of embodiment, just to refer each other for identical similar portion between each embodiment.For system disclosed in embodiment
For, since it is corresponded to the methods disclosed in the examples, so description is fairly simple, related place is said referring to method part
It is bright.
Principle and implementation of the present invention are described for specific case used herein, and above example is said
The bright method and its core concept for being merely used to help understand the present invention;Meanwhile for those of ordinary skill in the art, foundation
The thought of the present invention, there will be changes in the specific implementation manner and application range.In conclusion the content of the present specification is not
It is interpreted as limitation of the present invention.
Claims (10)
1. a kind of method for registering images based on hybrid rice algorithm, which is characterized in that the method includes:
The image for acquiring object, obtains reference picture and multiple images subject to registration;The reference picture representation space coordinate is just
True image;
Each image subject to registration is pre-processed, use bilinear interpolation method by each image procossing subject to registration for it is described
Reference picture has the image of equal resolution, obtains the corresponding pretreated image of each image subject to registration;
The first space transform models of each pretreated image are established by hybrid rice algorithm;
Second space transformation model is obtained according to the first space transform models of all pretreated images;
Affine transformation is carried out to the coordinate of the pretreated image by the second space transformation model, obtains corresponding be somebody's turn to do
The registration image of image subject to registration.
2. according to the method described in claim 1, it is characterized in that, described establish each pre- place by hybrid rice algorithm
First space transform models of the image after reason, specifically include:
Calculate the fitness function of rice individual;The rice individual representation space transformation model, the fitness function indicate
The similarity value of the reference picture and the pretreated image;
The rice individual is divided into maintainer, sterile line and restorer according to the fitness function;
Rice individual in the maintainer is hybridized with the rice individual in the sterile line, obtains the first optimal son
Body;
Rice individual in the restorer is selfed, the second optimal son individual is obtained;
Choose described first it is optimal son individual and described second it is optimal son individual in fitness function it is higher individual be third most
Excellent son individual, the optimal son individual of third indicate first spatial model.
3. according to the method described in claim 2, it is characterized in that, the rice individual by the maintainer with it is described not
The rice individual educated in being is hybridized, and is obtained the first optimal son individual, is specifically included:
Rice individual in the maintainer is subjected to randomer hybridization with the rice individual in the sterile line, obtains randomer hybridization
Individual;The randomer hybridization indicate any rice individual in the maintainer and any rice individual in the sterile line into
Row hybridization;The male parent of the randomer hybridization individual is the rice individual in the corresponding sterile line, the randomer hybridization individual
Female parent be the corresponding maintainer in rice individual;
The higher individual of fitness function in the randomer hybridization individual and the randomer hybridization individual male parent is chosen as the first son
Individual;
The highest first son individual of fitness function is chosen as the first optimal son individual.
4. according to the method described in claim 2, it is characterized in that, the rice individual by the maintainer with it is described not
The rice individual educated in being is hybridized, and is obtained the first optimal son individual, is specifically included:
Rice individual in the maintainer is subjected to corresponding hybridization with the rice individual in the sterile line, obtains corresponding hybridization
Individual;The corresponding hybridization indicate rice individual in the maintainer and the rice individual in the corresponding sterile line into
Row hybridization;The male parent of the corresponding hybrid individual is the rice individual in the corresponding sterile line, the corresponding hybrid individual
Female parent be the corresponding maintainer in rice individual;
The higher individual of fitness function in the corresponding hybrid individual and the corresponding hybrid individual male parent is chosen as the first son
Individual;
The highest first son individual of fitness function is chosen as the first optimal son individual.
5. according to the method described in claim 2, it is characterized in that, the rice individual by the restorer carries out certainly
It hands over, obtains the second optimal son individual, specifically include:
Rice individual in the restorer is selfed, selfing individual is obtained;
It is the second son to choose the higher individual of fitness function in the rice individual in the selfing individual and the restorer
Body;
The highest second son individual of fitness function is chosen as the second optimal son individual.
6. according to the method described in claim 2, it is characterized in that, described according to the first of all pretreated images
Space transform models obtain second space transformation model;It specifically includes:
The parameter of the first space transform models of all pretreated images is obtained, the parameter is that the third is optimal
The gene of sub- individual;
The parameter is obtained into second space transformation model as product.
7. a kind of figure registration system based on hybrid rice algorithm, which is characterized in that the system comprises:
Acquisition module, the image for acquiring object obtain reference picture and multiple images subject to registration;The reference picture table
Show the correct image of space coordinate;
Processing module, for being pre-processed to each image subject to registration, using bilinear interpolation method by each figure subject to registration
As processing to have the image of equal resolution with the reference picture, after obtaining the corresponding pretreatment of each image subject to registration
Image;
First modeling module, the first spatial alternation for establishing each pretreated image by hybrid rice algorithm
Model;
Second modeling module, for obtaining second space according to the first space transform models of all pretreated images
Transformation model;
Registration module, for carrying out affine change to the coordinate of the pretreated image by the second space transformation model
It changes, obtains the registration image for corresponding to the image subject to registration.
8. system according to claim 7, which is characterized in that first modeling module specifically includes:
Computing unit, the fitness function for calculating rice individual;The rice individual representation space transformation model, it is described suitable
The similarity value of reference picture described in response function representation and the pretreated image;
Taxon, for the rice individual to be divided into maintainer, sterile line and restorer according to the fitness function;
Hybridised units are obtained for hybridizing the rice individual in the maintainer with the rice individual in the sterile line
To the first optimal son individual;
It is selfed unit, for the rice individual in the restorer to be selfed, obtains the second optimal son individual;
Selection unit, it is higher for choosing fitness function in the described first optimal son individual and the described second optimal son individual
Individual is the optimal son individual of third, and the optimal son individual of third indicates the spatial model.
9. system according to claim 8, which is characterized in that the hybridised units specifically include:
Randomer hybridization subelement, for by the rice individual in rice individual and the sterile line in the maintainer carry out with
Machine hybridizes, and obtains randomer hybridization individual;The randomer hybridization indicates any rice individual in the maintainer and the infertility
Any rice individual in system is hybridized;The male parent of the randomer hybridization individual is the rice in the corresponding sterile line
Body, the female parent of the randomer hybridization individual are the rice individual in the corresponding maintainer;
First chooses subelement, for choosing fitness function in the randomer hybridization individual and the randomer hybridization individual male parent
Higher individual is the first son individual;
Second chooses subelement, for choosing the highest first son individual of fitness function as the first optimal son individual.
10. system according to claim 8, which is characterized in that the hybridised units specifically include:
Corresponding hybridization subelement, for carrying out pair rice individual and the rice individual in the sterile line in the maintainer
It should hybridize, obtain corresponding hybrid individual;The corresponding hybridization indicate rice individual in the maintainer with it is corresponding described
Rice individual in sterile line is hybridized;The male parent of the corresponding hybrid individual is the rice in the corresponding sterile line
Body, the female parent of the corresponding hybrid individual are the rice individual in the corresponding maintainer;
Third chooses subelement, for choosing fitness function in the corresponding hybrid individual and the corresponding hybrid individual male parent
Higher individual is the first son individual;
4th chooses subelement, for choosing the highest first son individual of fitness function as the first optimal son individual.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810092708.2A CN108305282A (en) | 2018-01-31 | 2018-01-31 | A kind of method for registering images and system based on hybrid rice algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810092708.2A CN108305282A (en) | 2018-01-31 | 2018-01-31 | A kind of method for registering images and system based on hybrid rice algorithm |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108305282A true CN108305282A (en) | 2018-07-20 |
Family
ID=62867155
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810092708.2A Pending CN108305282A (en) | 2018-01-31 | 2018-01-31 | A kind of method for registering images and system based on hybrid rice algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108305282A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112862866A (en) * | 2021-04-13 | 2021-05-28 | 湖北工业大学 | Image registration method and system based on sparrow search algorithm and computing equipment |
CN113707229A (en) * | 2021-08-13 | 2021-11-26 | 湖北工业大学 | Sulfur hexafluoride buffer gas selection method based on electronic localization function |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103236059A (en) * | 2013-04-25 | 2013-08-07 | 深圳先进技术研究院 | Diffeomorphism demons image registration method and system based on mode transformation |
CN104268869A (en) * | 2014-09-23 | 2015-01-07 | 中山大学 | Automatic multilevel-resolution remote-sensing image registration method based on particle swarm optimization |
-
2018
- 2018-01-31 CN CN201810092708.2A patent/CN108305282A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103236059A (en) * | 2013-04-25 | 2013-08-07 | 深圳先进技术研究院 | Diffeomorphism demons image registration method and system based on mode transformation |
CN104268869A (en) * | 2014-09-23 | 2015-01-07 | 中山大学 | Automatic multilevel-resolution remote-sensing image registration method based on particle swarm optimization |
Non-Patent Citations (1)
Title |
---|
YE ZHIWEI, ET AL.: "A Hybrid Rice Optimization Algorithm", 《THE 11TH INTERNATIONAL CONFERENCE ON COMPUTER SCIENCE&EDUCATION(ICCSE 2016)》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112862866A (en) * | 2021-04-13 | 2021-05-28 | 湖北工业大学 | Image registration method and system based on sparrow search algorithm and computing equipment |
CN113707229A (en) * | 2021-08-13 | 2021-11-26 | 湖北工业大学 | Sulfur hexafluoride buffer gas selection method based on electronic localization function |
CN113707229B (en) * | 2021-08-13 | 2023-06-09 | 湖北工业大学 | Sulfur hexafluoride buffer gas selection method based on electronic localization function |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104484648B (en) | Robot variable visual angle obstacle detection method based on outline identification | |
CN110659565B (en) | 3D multi-person human body posture estimation method based on porous convolution | |
CN104933717A (en) | Camera intrinsic and extrinsic parameter automatic calibration method based on directional calibration target | |
CN106127740B (en) | One kind being based on the associated profile testing method of the more orientation of sensory field of visual pathway | |
CN108171249B (en) | RGBD data-based local descriptor learning method | |
CN108305282A (en) | A kind of method for registering images and system based on hybrid rice algorithm | |
CN106952223A (en) | Method for registering images and device | |
CN106303409B (en) | A kind of target object combined tracking method and target object combine tracking device | |
CN110097574A (en) | A kind of real-time pose estimation method of known rigid body | |
CN110826500B (en) | Method for estimating 3D human body posture based on antagonistic network of motion link space | |
JP2019159940A (en) | Point group feature extraction device, point group feature extraction method, and program | |
CN109255783A (en) | A kind of position of skeleton key point on more people's images is arranged detection method | |
CN108022209A (en) | Acquisition methods and device, the method and apparatus of spatial compound imaging of color value | |
CN105488754B (en) | Image Feature Matching method and system based on local linear migration and affine transformation | |
CN107369169A (en) | The approximate most like image block matching method transmitted based on direction alignment with matching that a kind of GPU accelerates | |
CN104463896B (en) | Image corner point detection method and system based on kernel similar region distribution characteristics | |
CN110288026A (en) | A kind of image partition method and device practised based on metric relation graphics | |
CN104200214B (en) | The power spectrum of Walsh conversion is used to carry out the method for Description Image local characteristic region | |
CN105700402B (en) | Embedded control system, apparatus and method based on personnel's detection | |
CN105844589A (en) | Method for realizing light field image super-resolution based on mixed imaging system | |
CN109543896B (en) | Workshop equipment double-row layout method based on absolute value and function minimization and particle swarm optimization | |
CN108460829B (en) | A kind of 3-D image register method for AR system | |
CN108765472A (en) | Image Set Registration Method Based on Sparse Directed Graph | |
CN113177901B (en) | Multi-frame moving image fusion method and system for robot vision | |
CN103871068A (en) | High-precision calibration method based on genetic algorithm |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180720 |
|
RJ01 | Rejection of invention patent application after publication |