CN109903246A - The method and device of detection image variation - Google Patents
The method and device of detection image variation Download PDFInfo
- Publication number
- CN109903246A CN109903246A CN201910133541.4A CN201910133541A CN109903246A CN 109903246 A CN109903246 A CN 109903246A CN 201910133541 A CN201910133541 A CN 201910133541A CN 109903246 A CN109903246 A CN 109903246A
- Authority
- CN
- China
- Prior art keywords
- image
- multispectral
- single band
- optimal characteristics
- wave band
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The present invention discloses a kind of method and device of detection image variation, is related to technical field of image processing, its object is to improve the accuracy of variation testing result.The method comprise the steps that obtaining the first multispectral image and the second multispectral image；The processing of single band image zooming-out is carried out to the first multispectral image and the second multispectral image, to obtain the corresponding multiple first single band images of the first multispectral image and the corresponding multiple second single band images of the second multispectral image；The first single band image corresponding to each wave band and the second single band image are iterated weighting processing, to obtain the corresponding optimal characteristics differential image of each wave band；Fusion treatment is carried out to multiple optimal characteristics differential images, to obtain the corresponding change intensity image of multiple optimal characteristics differential images；Binary clusters analysis is carried out to change intensity image according to default clustering algorithm, to obtain variation testing result.During the present invention is suitable for being changed detection to two width multispectral images.
Description
Technical field
The present invention relates to technical field of image processing, more particularly to a kind of method and device of detection image variation.
Background technique
With the continuous development of science and technology, Remote Sensing Imagery Change Detection technology has been applied to numerous areas, by right
Several corresponding remote sensing images are changed detection when the same area difference, just can know the change of all atural objects in the region
Change feature and change procedure.Due to, different atural object has different structure and composition ingredients, therefore, different atural object tools
There is different Spectral Characteristics, i.e., the reflection spectrum curve of different atural object is different, thus when different atural object is certain
When reflectance spectrum on wave band is similar, then the reflectance spectrum of these atural objects has biggish difference on its all band, and it is more
Multiple wave bands of spectrum picture can embody feature of the atural object under different-waveband, thus corresponding when to the same area difference
Several multispectral images are changed detection, can more really reflect the situation of change of the atural object in the region.
Currently, when corresponding two width multispectral image is changed detection when to the same area difference, it will usually adopt
Detection is changed to this two width multispectral image with IR-MAD algorithm.However, due to the radiation of the factors such as external environment generation
Difference can be to the correlation between the corresponding two width single band image of certain wave bands under the two width multispectral images detected
It impacts, and when being changed detection to this two width multispectral image using IR-MAD algorithm, do not ensure that this two width
Higher correlation is all had between the corresponding two single band images of each wave band under multispectral image, so as to cause adopting
When corresponding two width multispectral image is changed detection when with IR-MAD algorithm to the same area difference, variation detection is obtained
As a result accuracy is lower.
Summary of the invention
In view of this, the present invention provides a kind of method and device of detection image variation, main purpose is to same
When corresponding two width multispectral image is changed detection when the difference of region, the accuracy of variation testing result is improved.
In order to achieve the above object, present invention generally provides following technical solutions:
In a first aspect, the present invention provides a kind of methods of detection image variation, this method comprises:
Obtain the first multispectral image and the second multispectral image, wherein first multispectral image and described second
Multispectral image corresponding multispectral image when being the same area difference；
The processing of single band image zooming-out is carried out to first multispectral image and second multispectral image, to obtain
The corresponding multiple first single band images of first multispectral image and second multispectral image corresponding multiple second
Single band image；
The first single band image corresponding to each wave band and the second single band image are iterated weighting processing, to obtain
The corresponding optimal characteristics differential image of each wave band；
Fusion treatment is carried out to multiple optimal characteristics differential images, to obtain multiple optimal characteristics differential images
Corresponding change intensity image；
Binary clusters analysis is carried out to the change intensity image according to default clustering algorithm, changes detection knot to obtain
Fruit.
Optionally, after the first multispectral image of the acquisition and the second multispectral image, the method also includes:
Registration process is carried out to first multispectral image and second multispectral image.
Optionally, the first single band image corresponding to each wave band and the second single band image are iterated weighting
Processing, to obtain the corresponding optimal characteristics differential image of each wave band, comprising:
The first single band image corresponding to each wave band and the second single band image are iterated weighting processing, to obtain
The corresponding first optimal projection vector of each wave band and the second optimal projection vector；
It is generated according to the corresponding first optimal projection vector of each wave band and the second optimal projection vector each described
The corresponding optimal characteristics differential image of wave band.
Optionally, it is iterated and adds in the first single band image corresponding to each wave band and the second single band image
Power processing, after obtaining the corresponding optimal characteristics differential image of each wave band, the method also includes:
Gaussian filtering process is carried out to the corresponding optimal characteristics differential image of each wave band.
Optionally, described that fusion treatment is carried out to multiple optimal characteristics differential images, it is multiple described optimal to obtain
The corresponding change intensity image of feature difference image, comprising:
Generate the corresponding optimal characteristics difference matrix of each optimal characteristics differential image；
Fusion treatment is carried out to multiple optimal characteristics difference matrix according to Euclidean distance formula, it is multiple described to obtain
The corresponding change intensity matrix of optimal characteristics difference matrix；
The change intensity image is generated according to the change intensity matrix.
Optionally, the default clustering algorithm is fuzzy C-mean algorithm FCM clustering algorithm.
Second aspect, the present invention also provides a kind of device of detection image variation, which includes:
Acquiring unit, for obtaining the first multispectral image and the second multispectral image, wherein the first multispectral figure
Picture and second multispectral image corresponding multispectral image when being the same area difference；
Extraction unit, first multispectral image and the second multispectral figure for being obtained to the acquiring unit
As carrying out the processing of single band image zooming-out, to obtain the corresponding multiple first single band images of first multispectral image and institute
State the corresponding multiple second single band images of the second multispectral image；
Iteration weighted units, for changing to the corresponding first single band image of each wave band and the second single band image
It is handled for weighting, to obtain the corresponding optimal characteristics differential image of each wave band；
Integrated unit, for being merged to multiple optimal characteristics differential images that the iteration weighted units obtain
Processing, to obtain the corresponding change intensity image of multiple optimal characteristics differential images；
Analytical unit, the change intensity image progress for being obtained according to clustering algorithm is preset to the integrated unit
Binary clusters analysis, to obtain variation testing result.
Optionally, described device further include:
Registration unit is used for after the acquiring unit obtains the first multispectral image and the second multispectral image, right
First multispectral image and second multispectral image carry out registration process.
Optionally, the iteration weighted units include:
Iteration weighting block, for changing to the corresponding first single band image of each wave band and the second single band image
It is handled for weighting, to obtain the corresponding first optimal projection vector of each wave band and the second optimal projection vector；
First generation module, it is corresponding first optimal for each of obtaining the wave band according to the iteration weighting block
Projection vector and the second optimal projection vector generate the corresponding optimal characteristics differential image of each wave band.
Optionally, described device further include:
Filter unit, for single in the iteration weighted units the first single band image corresponding to each wave band and second
Band image is iterated weighting processing, after obtaining the corresponding optimal characteristics differential image of each wave band, to each
The corresponding optimal characteristics differential image of the wave band carries out gaussian filtering process.
Optionally, the integrated unit includes:
Second generation module, for generating the corresponding optimal characteristics difference matrix of each optimal characteristics differential image；
Fusion Module, multiple optimal characteristics for being generated according to Euclidean distance formula to second generation module
Difference matrix carries out fusion treatment, to obtain the corresponding change intensity matrix of multiple optimal characteristics difference matrix；
Third generation module, it is strong that the change intensity matrix for being obtained according to the Fusion Module generates the variation
Spend image.
Optionally, the default clustering algorithm is fuzzy C-mean algorithm FCM clustering algorithm.
To achieve the goals above, according to the third aspect of the invention we, a kind of storage medium, the storage medium are provided
Program including storage, wherein equipment where controlling the storage medium in described program operation executes inspection described above
The method of altimetric image variation.
To achieve the goals above, according to the fourth aspect of the invention, a kind of processor is provided, the processor is used for
Run program, wherein described program executes detection image variation described above method when running.
By above-mentioned technical proposal, technical solution provided by the invention is at least had the advantage that
The present invention provides a kind of method and device of detection image variation, and uses IR-MAD algorithm to same in the prior art
Corresponding two width multispectral image is changed detection and compares when one region difference, and the present invention can acquire same area
When the difference of domain after corresponding first multispectral image and the second multispectral image, extracted from the first multispectral image each
The corresponding first single band image of wave band and the corresponding second single band figure of each wave band is extracted from the second multispectral image
Picture, and weighting processing is iterated to the corresponding first single band image of each wave band and the second single band image, to obtain often
The corresponding optimal characteristics differential image of a wave band is carrying out fusion treatment to multiple optimal characteristics differential images, to obtain more
After the corresponding change intensity image of a optimal characteristics differential image, two-value is carried out to change intensity image according to default clustering algorithm
Clustering obtains variation testing result.Due to single by the first single band image corresponding to each wave band respectively and second
Band image is iterated weighting processing, can effectively improve the corresponding first single band image of each wave band and the second single band
Correlation between image, so as to acquire the corresponding optimal characteristics differential image of each wave band, and then by by
The change intensity image that multiple optimal characteristics differential image fusions obtain carries out binary clusters analysis, the variation detection acquired
As a result accuracy is higher.
The above description is only an overview of the technical scheme of the present invention, in order to better understand the technical means of the present invention,
And it can be implemented in accordance with the contents of the specification, and in order to allow above and other objects of the present invention, feature and advantage can
It is clearer and more comprehensible, the followings are specific embodiments of the present invention.
Detailed description of the invention
By reading the following detailed description of the preferred embodiment, various other advantages and benefits are common for this field
Technical staff will become clear.The drawings are only for the purpose of illustrating a preferred embodiment, and is not considered as to the present invention
Limitation.And throughout the drawings, the same reference numbers will be used to refer to the same parts.In the accompanying drawings:
Fig. 1 shows a kind of method flow diagram of detection image variation provided in an embodiment of the present invention；
Fig. 2 shows the method flow diagrams of another detection image variation provided in an embodiment of the present invention；
Fig. 3 shows a kind of composition block diagram of the device of detection image variation provided in an embodiment of the present invention；
Fig. 4 shows the composition block diagram of the device of another detection image variation provided in an embodiment of the present invention.
Specific embodiment
Exemplary embodiments of the present disclosure are described in more detail below with reference to accompanying drawings.Although showing the disclosure in attached drawing
Exemplary embodiment, it being understood, however, that may be realized in various forms the disclosure without should be by embodiments set forth here
It is limited.On the contrary, these embodiments are provided to facilitate a more thoroughly understanding of the present invention, and can be by the scope of the present disclosure
It is fully disclosed to those skilled in the art.
The embodiment of the present invention provides a kind of method of detection image variation, as shown in Figure 1, this method comprises:
101, the first multispectral image and the second multispectral image are obtained.
Wherein, corresponding multispectral figure when the first multispectral image and the second multispectral image are the same area difference
Picture；Wherein, multispectral image is the image synthesized by multiple single band images of different-waveband, and multiple wave bands can be, but not limited to
Are as follows: red wave band, blue wave band, green wave band, near infrared band etc..
In embodiments of the present invention, it is necessary first to acquire corresponding first multispectral image when the same area difference
With the second multispectral image, so as to the image change between the first multispectral image of subsequent detection and the second multispectral image.
102, the processing of single band image zooming-out is carried out to the first multispectral image and the second multispectral image, to obtain first
The corresponding multiple first single band images of multispectral image and the corresponding multiple second single band images of the second multispectral image.
In embodiments of the present invention, since multispectral image is the figure synthesized by multiple single band images of different-waveband
Picture, therefore, after acquiring the first multispectral image and the second multispectral image, can respectively to the first multispectral image and
Second multispectral image carries out the processing of single band image zooming-out, to obtain corresponding multiple first unicasts of the first multispectral image
Section image and the corresponding multiple second single band images of the second multispectral image, i.e., extract each from the first multispectral image
The corresponding first single band image of wave band the second single band figure corresponding with each wave band is extracted from the second multispectral image
Picture.
103, the first single band image corresponding to each wave band and the second single band image are iterated weighting processing, with
Obtain the corresponding optimal characteristics differential image of each wave band.
It wherein, include corresponding first single band of the wave band in the corresponding optimal characteristics differential image of any one wave band
Change intensity information in image and the second single band image between corresponding pixel points.
In embodiments of the present invention, the corresponding multiple first single band images of the first multispectral image and more than second are being obtained
It, can the first single band image corresponding to each wave band and respectively after the corresponding multiple second single band images of spectrum picture
Two single band images are iterated weighting processing, so that the corresponding optimal characteristics differential image of each wave band is obtained, for example, preceding
It states in step, the corresponding first single band image of red wave band, blue wave band corresponding first is extracted from the first multispectral image
The corresponding first single band image of single band image, green wave band and the corresponding first single band image of near infrared band, and
The corresponding second single band image of red wave band, the corresponding second single band figure of blue wave band are extracted from the second multispectral image
Picture, the corresponding second single band image of green wave band and the corresponding second single band image of near infrared band, at this point, can be to red
The corresponding first single band image of wave band and the second single band image are iterated weighting processing, so that it is corresponding to obtain red wave band
Optimal characteristics differential image；The first single band image corresponding to blue wave band and the second single band image are iterated at weighting
Reason, to obtain the corresponding optimal characteristics differential image of blue wave band；The first single band image corresponding to green wave band and the second list
Band image is iterated weighting processing, to obtain the corresponding optimal characteristics differential image of green wave band；To near infrared band pair
The the first single band image and the second single band image answered are iterated weighting processing, so that it is corresponding most to obtain near infrared band
Excellent feature difference image.
It needs to be illustrated, passes through the first single band image corresponding to each wave band and the second single band figure respectively
As being iterated weighting processing, can effectively improve the corresponding first single band image of each wave band and the second single band image it
Between correlation, so as to acquire the corresponding optimal characteristics differential image of each wave band.
104, fusion treatment is carried out to multiple optimal characteristics differential images, it is corresponding to obtain multiple optimal characteristics differential images
Change intensity image.
It in embodiments of the present invention, can be to more after acquiring the corresponding optimal characteristics differential image of each wave band
A optimal characteristics differential image carries out fusion treatment, to obtain the corresponding change intensity figure of multiple optimal characteristics differential images
Picture obtains the first multispectral image and the corresponding change intensity image of the second multispectral image, wherein in change intensity image
It include the change intensity information in the first multispectral image and the second multispectral image between corresponding pixel points.Specifically,
In this step, fusion treatment can be carried out to multiple optimal characteristics differential images according to Euclidean distance formula, to obtain multiple
The corresponding change intensity image of optimal characteristics differential image, but not limited to this.
105, according to clustering algorithm is preset to the progress binary clusters analysis of change intensity image, change detection knot to obtain
Fruit.
Wherein, default clustering algorithm is specifically as follows FCM clustering algorithm.
In embodiments of the present invention, the first multispectral image and the corresponding variation of the second multispectral image are being acquired by force
After spending image, binary clusters analysis is carried out (i.e. using default clustering algorithm to the change intensity image according to default clustering algorithm
Binary clusters analysis is carried out to the multiple change intensity information for including in the change intensity image, to determine the first multispectral image
With in the second multispectral image between corresponding pixel points with the presence or absence of variation), at this point, the first multispectral image and the just can be obtained
The corresponding variation testing result of two multispectral images.
The embodiment of the present invention provides a kind of method of detection image variation, and uses IR-MAD algorithm to same in the prior art
Corresponding two width multispectral image is changed detection and compares when one region difference, and the embodiment of the present invention can acquire
When the same area difference after corresponding first multispectral image and the second multispectral image, extracted from the first multispectral image
The corresponding first single band image of each wave band and that each wave band is extracted from the second multispectral image is corresponding second single out
Band image, and weighting processing is iterated to the corresponding first single band image of each wave band and the second single band image, from
And the corresponding optimal characteristics differential image of each wave band is obtained, fusion treatment is being carried out to multiple optimal characteristics differential images, thus
After obtaining the corresponding change intensity image of multiple optimal characteristics differential images, according to default clustering algorithm to change intensity image into
The analysis of row binary clusters obtains variation testing result.Due to, by the first single band image corresponding to each wave band respectively and
Second single band image is iterated weighting processing, can effectively improve the corresponding first single band image of each wave band and second
Correlation between single band image, so as to acquire the corresponding optimal characteristics differential image of each wave band, Jin Ertong
It crosses and binary clusters analysis, the change acquired is carried out to the change intensity image obtained by the fusion of multiple optimal characteristics differential images
It is higher to change testing result accuracy.
Below in order to be explained in more detail, the embodiment of the invention provides the methods of another detection image variation, special
It is not that the first single band image corresponding to each wave band and the second single band image are iterated weighting processing, it is each to obtain
The specific method of the corresponding optimal characteristics differential image of wave band and to multiple optimal characteristics differential images carry out fusion treatment, with
The specific method of the corresponding change intensity image of multiple optimal characteristics differential images is obtained, specifically as shown in Fig. 2, this method packet
It includes:
201, the first multispectral image and the second multispectral image are obtained.
Wherein, about step 201, the first multispectral image of acquisition and the second multispectral image, portion can be corresponded to reference to Fig. 1
The description divided, the embodiment of the present invention will not be described in great detail herein.
202, registration process is carried out to the first multispectral image and the second multispectral image.
In embodiments of the present invention, in order to guarantee multiple pixels for including in the first multispectral image and second multispectral
The multiple pixels for including in image correspond, and after acquiring the first multispectral image and the second multispectral image, need
Registration process is carried out to the first multispectral image and the second multispectral image.Specifically, in this step, remote sensing can be used
Image processing software ENVI carries out registration process to the first multispectral image and the second multispectral image, but not limited to this.
203, the processing of single band image zooming-out is carried out to the first multispectral image and the second multispectral image, to obtain first
The corresponding multiple first single band images of multispectral image and the corresponding multiple second single band images of the second multispectral image.
Wherein, it is carried out at single band image zooming-out about step 203, to the first multispectral image and the second multispectral image
Reason, to obtain the corresponding multiple first single band images of the first multispectral image and the second multispectral image corresponding multiple second
Single band image, can refer to the description of Fig. 1 corresponding part, and the embodiment of the present invention will not be described in great detail herein.
204, the first single band image corresponding to each wave band and the second single band image are iterated weighting processing, with
Obtain the corresponding optimal characteristics differential image of each wave band.
In embodiments of the present invention, the corresponding multiple first single band images of the first multispectral image and more than second are being obtained
It, can the first single band image corresponding to each wave band and respectively after the corresponding multiple second single band images of spectrum picture
Two single band images are iterated weighting processing, to obtain the corresponding optimal characteristics differential image of each wave band.It below will be right
How the first single band image corresponding to each wave band and the second single band image are iterated weighting processing, each to obtain
The corresponding optimal characteristics differential image of wave band is described in detail.
(1) the first single band image corresponding to each wave band and the second single band image are iterated weighting processing, with
Obtain the corresponding first optimal projection vector of each wave band and the second optimal projection vector.
In embodiments of the present invention, the corresponding multiple first single band images of the first multispectral image and more than second are being obtained
It, can the first single band image corresponding to each wave band and respectively after the corresponding multiple second single band images of spectrum picture
Two single band images are iterated weighting processing, to obtain the corresponding first optimal projection vector of each wave band and second optimal
Projection vector.
Specifically, in this step, the first single band image corresponding to each wave band and the second single band image carry out
The process of iteration weighting processing is as follows:
1, the first single band figure is generated according to the corresponding first single band image of the P wave band and the second single band image
As corresponding first image array Fp second image array Gp corresponding with the second single band image；
2, according to the first image array Fp and the second image array Gp, the corresponding association side the first image array Fp is calculated separately
Poor matrixThe corresponding covariance matrix of second image array GpAnd the first image array FpWith the second image moment
Battle array GpCorresponding Cross-covarianceWith
3, by covariance matrixCovariance matrixCross-covarianceWith cross covariance square
Battle arrayIt is substituted into the first preset formula and the second preset formula respectively, calculates corresponding first projection of the first image array Fp
Vector ap, the corresponding second projection vector b of the second image array GppAnd first between image array Fp and the second image array Gp
CorrelationWherein, the first preset formula is specific as follows:
Second preset formula is specific as follows:
4, according to the first image array Fp, the second image array Gp, the first projection vector apWith the second projection vector bpIt calculates
The first image array Fp and corresponding feature difference matrix Mp of the second image array Gp, i.e.,
5, the corresponding chi-Square measure T of feature difference matrix Mp is calculated according to feature difference matrix Mpij, i.e. Tij=(Mpij/
σp)2∈x2(n), wherein chi-Square measure TijMeet the chi square distribution that freedom degree is n, σpFor the first image array Fp and the second figure
As the corresponding variance of matrix Gp；
6, according to chi-Square measure TijWeighted value ω is calculated with the probability density quantile of chi square distributionij, i.e. ωij=P (Tij>
T)=P (x2(n)>Tij)；
7, it repeats step 2-6 and carries out successive ignition weighted calculation, need to be illustrated, add carrying out second of iteration
During any an iteration weighted calculation after power calculating and second of iteration weighted calculation, covariance matrix is being calculatedCovariance matrixCross-covarianceAnd Cross-covarianceWhen, it is any to be related to
The calculating of variance and mean value is required to using the weighted value ω acquired in last iteration weight computation processijIt is weighted place
Reason；It is less than default threshold when reaching the difference for acquiring correlation ρ in default the number of iterations or the adjacent weight computation process of iteration twice
When value, processing terminate for iteration weighting, and the first projection vector a that will be acquired in last time iteration weight computation processpWith
Two projection vector bpIt is determined as the corresponding first optimal projection vector of the P wave band and the second optimal projection vector, wherein pre-
If the number of iterations can be, but not limited to are as follows: 9 times, 10 times, 11 times etc., preset threshold can be, but not limited to are as follows: 0.0001,
0.0002,0.0003 etc.；
8, using method described in step 1-7, the first single band image corresponding to its all band and the second single band figure
As being iterated weighting processing, to obtain the corresponding first optimal projection vector of each wave band and the second optimal projection vector.
(2) each wave band pair is generated according to the corresponding first optimal projection vector of each wave band and the second optimal projection vector
The optimal characteristics differential image answered.
In embodiments of the present invention, the corresponding first optimal projection vector of each wave band and the second optimal throwing are being obtained respectively
After shadow vector, each wave band can be generated according to the corresponding first optimal projection vector of each wave band and the second optimal projection vector
Corresponding optimal characteristics differential image.
Specifically, in this step, according to the corresponding first optimal projection vector of each wave band and second it is optimal project to
The process that amount generates the corresponding optimal characteristics differential image of each wave band is as follows:
1, the corresponding first image array Fp of the P wave band, the second image array Gp, the first optimal projection vector a are obtainedp
With the second optimal projection vector bp；
2, according to the first image array Fp, the second image array Gp, the first optimal projection vector apWith the second optimal projection
Vector bpThe first image array Fp and the corresponding optimal characteristics difference matrix Mp of the second image array Gp are calculated, i.e.,
3, the corresponding optimal characteristics differential image of the P wave band is generated according to optimal characteristics difference matrix Mp；
4, optimal according to the corresponding first optimal projection vector of its all band and second using method described in step 1-3
Projection vector generates the corresponding optimal characteristics differential image of each wave band.
205, gaussian filtering process is carried out to the corresponding optimal characteristics differential image of each wave band.
In embodiments of the present invention, in order to reduce the factors such as external environment shadow caused by multiple optimal characteristics differential images
It rings, after acquiring multiple optimal characteristics differential images, needs respectively to each optimal characteristics differential image (i.e. each wave band
Corresponding optimal characteristics differential image) carry out gaussian filtering process.
206, fusion treatment is carried out to multiple optimal characteristics differential images, it is corresponding to obtain multiple optimal characteristics differential images
Change intensity image.
In embodiments of the present invention, the corresponding optimal characteristics differential image of each wave band is being carried out at gaussian filtering respectively
After reason, fusion treatment is carried out to multiple optimal characteristics differential images after gaussian filtering process, to obtain multiple optimal
The corresponding change intensity image of feature difference image, that is, obtain the first multispectral image and the corresponding variation of the second multispectral image
Intensity image.Below will to how to multiple optimal characteristics differential images carry out fusion treatment, it is poor to obtain multiple optimal characteristics
The corresponding change intensity image of different image is described in detail.
(1) the corresponding optimal characteristics difference matrix of each optimal characteristics differential image is generated.
In embodiments of the present invention, it in order to which multiple optimal characteristics differential images are fused to change intensity image, needs to give birth to
At the corresponding optimal characteristics difference matrix of each optimal characteristics differential image.
(2) fusion treatment is carried out to multiple optimal characteristics difference matrix according to Euclidean distance formula, it is multiple optimal to obtain
The corresponding change intensity matrix of feature difference matrix.
In embodiments of the present invention, after generating the corresponding optimal characteristics difference matrix of each optimal characteristics differential image,
Fusion treatment can be carried out to multiple optimal characteristics difference matrix according to Euclidean distance formula, so that it is poor to obtain multiple optimal characteristics
The corresponding change intensity matrix of different matrix.Specifically, in this step, first multiple optimal characteristics difference matrix can be adjusted to
Column vector；Then, multiple column vectors are written in preset empty matrix, summarize feature difference matrix to obtain；Finally, using
It is change intensity matrix that Euclidean distance formula, which will summarize feature difference matrix conversion,.
(3) change intensity image is generated according to change intensity matrix.
In embodiments of the present invention, after acquiring the corresponding change intensity matrix of multiple optimal characteristics difference matrix,
The first multispectral image and the corresponding change intensity image of the second multispectral image can be generated according to change intensity matrix.
207, according to clustering algorithm is preset to the progress binary clusters analysis of change intensity image, change detection knot to obtain
Fruit.
Wherein, binary clusters analysis is carried out to change intensity image about step 207, according to default clustering algorithm, to obtain
Testing result must be changed, the description of Fig. 1 corresponding part can be referred to, the embodiment of the present invention will not be described in great detail herein.
To achieve the goals above, according to another aspect of the present invention, the embodiment of the invention also provides a kind of storage Jie
Matter, the storage medium include the program of storage, wherein equipment where controlling the storage medium in described program operation is held
The method of row detection image variation described above.
To achieve the goals above, according to another aspect of the present invention, the embodiment of the invention also provides a kind of processor,
The processor is for running program, wherein described program executes detection image variation described above method when running.
Further, as the realization to method shown in above-mentioned Fig. 1 and Fig. 2, another embodiment of the present invention additionally provides one
The device of kind detection image variation.The Installation practice is corresponding with preceding method embodiment, is easy to read, present apparatus embodiment
No longer the detail content in preceding method embodiment is repeated one by one, it should be understood that the device in the present embodiment can
The corresponding full content realized in preceding method embodiment.The device is applied to corresponding two width when to the same area difference
When multispectral image is changed detection, the accuracy of variation testing result is improved, specifically as shown in figure 3, the device includes:
Acquiring unit 31, for obtaining the first multispectral image and the second multispectral image, wherein described first is multispectral
Image and second multispectral image corresponding multispectral image when being the same area difference；
Extraction unit 32, first multispectral image and the second multispectral figure for being obtained to acquiring unit 31
As carrying out the processing of single band image zooming-out, to obtain the corresponding multiple first single band images of first multispectral image and institute
State the corresponding multiple second single band images of the second multispectral image；
Iteration weighted units 33 are carried out for the first single band image corresponding to each wave band and the second single band image
Iteration weighting processing, to obtain the corresponding optimal characteristics differential image of each wave band；
Integrated unit 34, for being merged to multiple optimal characteristics differential images that iteration weighted units 33 obtain
Processing, to obtain the corresponding change intensity image of multiple optimal characteristics differential images；
Analytical unit 35, the change intensity image progress for being obtained according to clustering algorithm is preset to integrated unit 34
Binary clusters analysis, to obtain variation testing result.
Further, as shown in figure 4, the device further include:
Registration unit 36 is used for after acquiring unit 31 obtains the first multispectral image and the second multispectral image, right
First multispectral image and second multispectral image carry out registration process.
Further, as shown in figure 4, iteration weighted units 33 include:
Iteration weighting block 331, for the first single band image corresponding to each wave band and the second single band image into
Row iteration weighting processing, to obtain the corresponding first optimal projection vector of each wave band and the second optimal projection vector；
First generation module 332, for according to iteration weighting block 331 obtain each of the wave band corresponding first most
Excellent projection vector and the second optimal projection vector generate the corresponding optimal characteristics differential image of each wave band.
Further, as shown in figure 4, the device further include:
Filter unit 37, for single in the first single band image corresponding to each wave band of iteration weighted units 33 and second
Band image is iterated weighting processing, after obtaining the corresponding optimal characteristics differential image of each wave band, to each
The corresponding optimal characteristics differential image of the wave band carries out gaussian filtering process.
Further, as shown in figure 4, integrated unit 34 includes:
Second generation module 341, for generating the corresponding optimal characteristics difference square of each optimal characteristics differential image
Battle array；
Fusion Module 342, multiple optimal spies for being generated according to Euclidean distance formula to the second generation module 341
It levies difference matrix and carries out fusion treatment, to obtain the corresponding change intensity matrix of multiple optimal characteristics difference matrix；
Third generation module 343, the change intensity matrix for being obtained according to Fusion Module 342 generate the variation
Intensity image.
Further, as shown in figure 4, the default clustering algorithm is fuzzy C-mean algorithm FCM clustering algorithm.
The embodiment of the present invention provides a kind of method and device of detection image variation, and uses IR-MAD to calculate in the prior art
Corresponding two width multispectral image is changed detection and compares when method is to the same area difference, and the embodiment of the present invention can obtain
It obtains when the same area difference after corresponding first multispectral image and the second multispectral image, from the first multispectral image
In extract the corresponding first single band image of each wave band and that each wave band is extracted from the second multispectral image is corresponding
Second single band image, and the corresponding first single band image of each wave band and the second single band image are iterated at weighting
Reason is carrying out fusion treatment to multiple optimal characteristics differential images to obtain the corresponding optimal characteristics differential image of each wave band,
After obtaining the corresponding change intensity image of multiple optimal characteristics differential images, according to default clustering algorithm to change intensity figure
As carrying out binary clusters analysis, variation testing result is obtained.Due to passing through the first single band figure corresponding to each wave band respectively
Picture and the second single band image be iterated weighting processing, can effectively improve the corresponding first single band image of each wave band and
Correlation between second single band image, so as to acquire the corresponding optimal characteristics differential image of each wave band, into
And by carrying out binary clusters analysis to the change intensity image obtained by the fusion of multiple optimal characteristics differential images, it acquires
Variation testing result accuracy it is higher.
The device of the detection image variation includes processor and memory, and above-mentioned acquiring unit, extraction unit, iteration add
It weighs unit, integrated unit and analytical unit etc. to store in memory as program unit, be stored in by processor execution
Above procedure unit in reservoir realizes corresponding function.
Include kernel in processor, is gone in memory to transfer corresponding program unit by kernel.Kernel can be set one
Or more, detection is changed by adjusting kernel parameter come two width multispectral image corresponding when to the same area difference
When, improve the accuracy of variation testing result.
Memory may include the non-volatile memory in computer-readable medium, random access memory (RAM) and/
Or the forms such as Nonvolatile memory, if read-only memory (ROM) or flash memory (flash RAM), memory include that at least one is deposited
Store up chip.
The embodiment of the invention provides a kind of storage mediums, are stored thereon with program, real when which is executed by processor
The method of detection image variation described in any one of existing above embodiments.
The embodiment of the invention provides a kind of processor, the processor is for running program, wherein described program operation
The method of the variation of detection image described in any one of Shi Zhihang above embodiments.
The embodiment of the invention provides a kind of equipment, equipment include processor, memory and storage on a memory and can
The program run on a processor, processor perform the steps of when executing program
Obtain the first multispectral image and the second multispectral image, wherein first multispectral image and described second
Multispectral image corresponding multispectral image when being the same area difference；
The processing of single band image zooming-out is carried out to first multispectral image and second multispectral image, to obtain
The corresponding multiple first single band images of first multispectral image and second multispectral image corresponding multiple second
Single band image；
The first single band image corresponding to each wave band and the second single band image are iterated weighting processing, to obtain
The corresponding optimal characteristics differential image of each wave band；
Fusion treatment is carried out to multiple optimal characteristics differential images, to obtain multiple optimal characteristics differential images
Corresponding change intensity image；
Binary clusters analysis is carried out to the change intensity image according to default clustering algorithm, changes detection knot to obtain
Fruit.
Further, after the first multispectral image of the acquisition and the second multispectral image, the method also includes:
Registration process is carried out to first multispectral image and second multispectral image.
Further, the first single band image corresponding to each wave band and the second single band image, which are iterated, adds
Power processing, to obtain the corresponding optimal characteristics differential image of each wave band, comprising:
The first single band image corresponding to each wave band and the second single band image are iterated weighting processing, to obtain
The corresponding first optimal projection vector of each wave band and the second optimal projection vector；
It is generated according to the corresponding first optimal projection vector of each wave band and the second optimal projection vector each described
The corresponding optimal characteristics differential image of wave band.
Further, it is iterated in the first single band image corresponding to each wave band and the second single band image
Weighting processing, after obtaining the corresponding optimal characteristics differential image of each wave band, the method also includes:
Gaussian filtering process is carried out to the corresponding optimal characteristics differential image of each wave band.
Further, described that fusion treatments are carried out to multiple optimal characteristics differential images, with obtain it is multiple it is described most
The corresponding change intensity image of excellent feature difference image, comprising:
Generate the corresponding optimal characteristics difference matrix of each optimal characteristics differential image；
Fusion treatment is carried out to multiple optimal characteristics difference matrix according to Euclidean distance formula, it is multiple described to obtain
The corresponding change intensity matrix of optimal characteristics difference matrix；
The change intensity image is generated according to the change intensity matrix.
Further, the default clustering algorithm is fuzzy C-mean algorithm FCM clustering algorithm.
Present invention also provides a kind of computer program products, when executing on data processing equipment, are adapted for carrying out just
The program code of beginningization there are as below methods step: the first multispectral image and the second multispectral image are obtained, wherein described first
Multispectral image and second multispectral image corresponding multispectral image when being the same area difference；More than described first
Spectrum picture and second multispectral image carry out the processing of single band image zooming-out, to obtain first multispectral image pair
The corresponding multiple second single band images of multiple first single band images and second multispectral image answered；To each wave band
Corresponding first single band image and the second single band image are iterated weighting processing, corresponding to obtain each wave band
Optimal characteristics differential image；Fusion treatment is carried out to multiple optimal characteristics differential images, to obtain multiple optimal spies
Levy the corresponding change intensity image of differential image；Binary clusters point are carried out to the change intensity image according to default clustering algorithm
Analysis, to obtain variation testing result.
It should be understood by those skilled in the art that, embodiments herein can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the application
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the application, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
The application is referring to method, the process of equipment (system) and computer program product according to the embodiment of the present application
Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
In a typical configuration, calculating equipment includes one or more processors (CPU), input/output interface, net
Network interface and memory.
Memory may include the non-volatile memory in computer-readable medium, random access memory (RAM) and/
Or the forms such as Nonvolatile memory, such as read-only memory (ROM) or flash memory (flash RAM).Memory is computer-readable Jie
The example of matter.
Computer-readable medium includes permanent and non-permanent, removable and non-removable media can be by any method
Or technology come realize information store.Information can be computer readable instructions, data structure, the module of program or other data.
The example of the storage medium of computer includes, but are not limited to phase change memory (PRAM), static random access memory (SRAM), moves
State random access memory (DRAM), other kinds of random access memory (RAM), read-only memory (ROM), electric erasable
Programmable read only memory (EEPROM), flash memory or other memory techniques, read-only disc read only memory (CD-ROM) (CD-ROM),
Digital versatile disc (DVD) or other optical storage, magnetic cassettes, tape magnetic disk storage or other magnetic storage devices
Or any other non-transmission medium, can be used for storage can be accessed by a computing device information.As defined in this article, it calculates
Machine readable medium does not include temporary computer readable media (transitory media), such as the data-signal and carrier wave of modulation.
It should also be noted that, the terms "include", "comprise" or its any other variant are intended to nonexcludability
It include so that the process, method, commodity or the equipment that include a series of elements not only include those elements, but also to wrap
Include other elements that are not explicitly listed, or further include for this process, method, commodity or equipment intrinsic want
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including element
There is also other identical elements in process, method, commodity or equipment.
It will be understood by those skilled in the art that embodiments herein can provide as method, system or computer program product.
Therefore, complete hardware embodiment, complete software embodiment or embodiment combining software and hardware aspects can be used in the application
Form.It is deposited moreover, the application can be used to can be used in the computer that one or more wherein includes computer usable program code
The shape for the computer program product implemented on storage media (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
Formula.
The above is only embodiments herein, are not intended to limit this application.To those skilled in the art,
Various changes and changes are possible in this application.It is all within the spirit and principles of the present application made by any modification, equivalent replacement,
Improve etc., it should be included within the scope of the claims of this application.
Claims (10)
1. a kind of method of detection image variation characterized by comprising
Obtain the first multispectral image and the second multispectral image, wherein first multispectral image and more than second light
Spectrogram picture corresponding multispectral image when being the same area difference；
The processing of single band image zooming-out is carried out to first multispectral image and second multispectral image, described in obtaining
The corresponding multiple first single band images of first multispectral image and corresponding multiple second unicasts of second multispectral image
Section image；
The first single band image corresponding to each wave band and the second single band image are iterated weighting processing, each to obtain
The corresponding optimal characteristics differential image of the wave band；
Fusion treatment is carried out to multiple optimal characteristics differential images, it is corresponding to obtain multiple optimal characteristics differential images
Change intensity image；
Binary clusters analysis is carried out to the change intensity image according to default clustering algorithm, to obtain variation testing result.
2. the method according to claim 1, wherein multispectral in the first multispectral image of the acquisition and second
After image, the method also includes:
Registration process is carried out to first multispectral image and second multispectral image.
3. the method according to claim 1, wherein the first single band image corresponding to each wave band and
Second single band image is iterated weighting processing, to obtain the corresponding optimal characteristics differential image of each wave band, comprising:
The first single band image corresponding to each wave band and the second single band image are iterated weighting processing, each to obtain
The corresponding first optimal projection vector of the wave band and the second optimal projection vector；
Each wave band is generated according to the corresponding first optimal projection vector of each wave band and the second optimal projection vector
Corresponding optimal characteristics differential image.
4. the method according to claim 1, wherein in the first single band image corresponding to each wave band
With the second single band image be iterated weighting processing, with obtain the corresponding optimal characteristics differential image of each wave band it
Afterwards, the method also includes:
Gaussian filtering process is carried out to the corresponding optimal characteristics differential image of each wave band.
5. the method according to claim 1, wherein described melt multiple optimal characteristics differential images
Conjunction processing, to obtain the corresponding change intensity image of multiple optimal characteristics differential images, comprising:
Generate the corresponding optimal characteristics difference matrix of each optimal characteristics differential image；
Fusion treatment is carried out to multiple optimal characteristics difference matrix according to Euclidean distance formula, it is multiple described optimal to obtain
The corresponding change intensity matrix of feature difference matrix；
The change intensity image is generated according to the change intensity matrix.
6. method according to any one of claims 1-5, which is characterized in that the default clustering algorithm is that Fuzzy C is equal
Value FCM clustering algorithm.
7. a kind of device of detection image variation characterized by comprising
Acquiring unit, for obtaining the first multispectral image and the second multispectral image, wherein first multispectral image and
Second multispectral image corresponding multispectral image when being the same area difference；
Extraction unit, first multispectral image and second multispectral image for being obtained to the acquiring unit into
Row single band image zooming-out processing, to obtain the corresponding multiple first single band images of first multispectral image and described the
The corresponding multiple second single band images of two multispectral images；
Iteration weighted units add for being iterated to the corresponding first single band image of each wave band and the second single band image
Power processing, to obtain the corresponding optimal characteristics differential image of each wave band；
Integrated unit, for being carried out at fusion to multiple optimal characteristics differential images that the iteration weighted units obtain
Reason, to obtain the corresponding change intensity image of multiple optimal characteristics differential images；
Analytical unit, the change intensity image progress two-value for being obtained according to clustering algorithm is preset to the integrated unit
Clustering, to obtain variation testing result.
8. device according to claim 7, which is characterized in that described device further include:
Registration unit is used for after the acquiring unit obtains the first multispectral image and the second multispectral image, to described
First multispectral image and second multispectral image carry out registration process.
9. a kind of storage medium, which is characterized in that the storage medium includes the program of storage, wherein run in described program
When control the storage medium where equipment perform claim require 1 to the detection image variation described in any one of claim 6
Method.
10. a kind of processor, which is characterized in that the processor is for running program, wherein right of execution when described program is run
Benefit require 1 to the detection image variation described in any one of claim 6 method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910133541.4A CN109903246B (en) | 2019-02-22 | 2019-02-22 | Method and device for detecting image change |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910133541.4A CN109903246B (en) | 2019-02-22 | 2019-02-22 | Method and device for detecting image change |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109903246A true CN109903246A (en) | 2019-06-18 |
| CN109903246B CN109903246B (en) | 2022-09-06 |
Family
ID=66945434
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910133541.4A Active CN109903246B (en) | 2019-02-22 | 2019-02-22 | Method and device for detecting image change |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109903246B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112633155A (en) * | 2020-12-22 | 2021-04-09 | 生态环境部卫星环境应用中心 | Natural conservation place human activity change detection method based on multi-scale feature fusion |
| CN112801978A (en) * | 2021-01-28 | 2021-05-14 | 新疆大学 | Multispectral remote sensing image change detection method and device and storage medium |
| CN114627087A (en) * | 2022-03-21 | 2022-06-14 | 国网江苏省电力有限公司无锡供电分公司 | Method and system for automatically detecting ground object change of multi-temporal satellite remote sensing image |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105608473A (en) * | 2015-12-31 | 2016-05-25 | 中国资源卫星应用中心 | High-precision land cover classification method based on high-resolution satellite image |
| CN106295696A (en) * | 2016-08-09 | 2017-01-04 | 中国科学院遥感与数字地球研究所 | A kind of multi-source Remote Sensing Images radiation normalization method |
-
2019
- 2019-02-22 CN CN201910133541.4A patent/CN109903246B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105608473A (en) * | 2015-12-31 | 2016-05-25 | 中国资源卫星应用中心 | High-precision land cover classification method based on high-resolution satellite image |
| CN106295696A (en) * | 2016-08-09 | 2017-01-04 | 中国科学院遥感与数字地球研究所 | A kind of multi-source Remote Sensing Images radiation normalization method |
Non-Patent Citations (4)
| Title |
|---|
| ALLAN AASBJERG NIELSEN: "The Regularized Iteratively Reweighted MAD Method for Change Detection in Multi- and Hyperspectral Data", 《IEEE TRANSACTIONS ON IMAGE PROCESSING》 * |
| 周文艳: "基于组合差异图和FCM聚类的SAR图像变化检测", 《激光杂志》 * |
| 李莎: "基于选权迭代估计与非监督分类的多光谱图像变化检测", 《国土资源遥感》 * |
| 马利媛: "多光谱遥感图像的变化检测算法研究", 《中国优秀硕士学位论文全文数据库 (工程科技Ⅱ辑)》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112633155A (en) * | 2020-12-22 | 2021-04-09 | 生态环境部卫星环境应用中心 | Natural conservation place human activity change detection method based on multi-scale feature fusion |
| CN112801978A (en) * | 2021-01-28 | 2021-05-14 | 新疆大学 | Multispectral remote sensing image change detection method and device and storage medium |
| CN114627087A (en) * | 2022-03-21 | 2022-06-14 | 国网江苏省电力有限公司无锡供电分公司 | Method and system for automatically detecting ground object change of multi-temporal satellite remote sensing image |
| CN114627087B (en) * | 2022-03-21 | 2024-04-12 | 国网江苏省电力有限公司无锡供电分公司 | Automatic detection method and system for ground feature change of multi-temporal satellite remote sensing image |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109903246B (en) | 2022-09-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Cortinhal et al. | Salsanext: Fast, uncertainty-aware semantic segmentation of lidar point clouds | |
| Uhrig et al. | Sparsity invariant cnns | |
| EP2356614B1 (en) | A constraint generator for use in image segregation | |
| Singh et al. | Single image dehazing for a variety of haze scenarios using back projected pyramid network | |
| CN109903246A (en) | The method and device of detection image variation | |
| CN112232426B (en) | Training method, device and equipment of target detection model and readable storage medium | |
| CN113159120A (en) | Contraband detection method based on multi-scale cross-image weak supervision learning | |
| CN111160140A (en) | Image detection method and device | |
| Cheng et al. | Learning to refine depth for robust stereo estimation | |
| Wilde et al. | Detecting gravitational lenses using machine learning: exploring interpretability and sensitivity to rare lensing configurations | |
| CN111860124A (en) | Remote sensing image classification method based on space spectrum capsule generation countermeasure network | |
| CN108830828A (en) | A kind of method for detecting change of remote sensing image and device | |
| Li et al. | Sfnet: Faster and accurate semantic segmentation via semantic flow | |
| Vašata et al. | Image inpainting using Wasserstein generative adversarial imputation network | |
| Gokul et al. | Refine and represent: Region-to-object representation learning | |
| Otsuzuki et al. | Regularized pooling | |
| US20220327811A1 (en) | System and method for composite training in machine learning architectures | |
| López-Fandino et al. | GPU projection of ECAS-II segmenter for hyperspectral images based on cellular automata | |
| KR102186125B1 (en) | Apparatus and method for generating land cover map | |
| Negri et al. | A general and extensible framework for assessing change detection techniques | |
| Candare et al. | Mapping of high value crops through an object-based SVM model using lidar data and orthophoto in Agusan del Norte Philippines | |
| Körting et al. | Divide And Segment-An Alternative For Parallel Segmentation. | |
| Werner et al. | Saliency-guided object candidates based on gestalt principles | |
| Li et al. | Obstructing deepfakes by disrupting face detection and facial landmarks extraction | |
| Draganov et al. | Object motion detection in video by fusion of RPCA and nmf decompositions |
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 |