CN109903246B - Method and device for detecting image change - Google Patents
Method and device for detecting image change Download PDFInfo
- Publication number
- CN109903246B CN109903246B CN201910133541.4A CN201910133541A CN109903246B CN 109903246 B CN109903246 B CN 109903246B CN 201910133541 A CN201910133541 A CN 201910133541A CN 109903246 B CN109903246 B CN 109903246B
- Authority
- CN
- China
- Prior art keywords
- image
- band
- matrix
- optimal
- projection vector
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method and a device for detecting image change, relates to the technical field of image processing, and aims to improve the accuracy of a change detection result. The method of the invention comprises the following steps: acquiring a first multispectral image and a second multispectral image; performing single-band image extraction processing on the first multispectral image and the second multispectral image to obtain a plurality of first single-band images corresponding to the first multispectral image and a plurality of second single-band images corresponding to the second multispectral image; performing iterative weighting processing on the first single-waveband image and the second single-waveband image corresponding to each waveband to obtain an optimal characteristic difference image corresponding to each waveband; performing fusion processing on the optimal feature difference images to obtain variation intensity images corresponding to the optimal feature difference images; and carrying out binary clustering analysis on the change intensity image according to a preset clustering algorithm to obtain a change detection result. The method is suitable for the process of change detection of the two multispectral images.
Description
Technical Field
The present invention relates to the field of image processing technologies, and in particular, to a method and an apparatus for detecting image changes.
Background
With the continuous development of scientific technology, the remote sensing image change detection technology is applied to many fields, and by detecting the change of a plurality of remote sensing images which are different from each other and correspond to the same region, the change characteristics and the change process of all ground objects in the region can be obtained. Because different ground objects have different structures and composition components, different ground objects have different spectral characteristics, namely, reflection spectral curves of different ground objects are different, so that when the reflection spectra of different ground objects on certain wave bands are similar, the reflection spectra of the ground objects on other wave bands have larger difference, and a plurality of wave bands of the multispectral image can reflect the characteristics of the ground objects under different wave bands, so that a plurality of multispectral images corresponding to the same region in different directions are subjected to change detection, and the change condition of the ground objects in the region can be reflected more truly.
At present, when two multispectral images corresponding to different regions in the same region are subjected to change detection, an IR-MAD algorithm is usually adopted to carry out change detection on the two multispectral images. However, radiation differences generated by factors such as an external environment affect the correlation between two single-band images corresponding to certain bands in the two multispectral images to be detected, and when the IR-MAD algorithm is used for detecting changes of the two multispectral images, it cannot be guaranteed that the two single-band images corresponding to each band in the two multispectral images have high correlation, so that when the IR-MAD algorithm is used for detecting changes of two multispectral images corresponding to different regions in the same region, the accuracy of a change detection result is low.
Disclosure of Invention
In view of this, the present invention provides a method and an apparatus for detecting image changes, and mainly aims to improve the accuracy of a change detection result when two multispectral images corresponding to different regions are subjected to change detection.
In order to achieve the above purpose, the present invention mainly provides the following technical solutions:
in a first aspect, the present invention provides a method of detecting image changes, the method comprising:
acquiring a first multispectral image and a second multispectral image, wherein the first multispectral image and the second multispectral image are multispectral images which are different and correspond to each other in the same region;
performing single-band image extraction processing on the first multispectral image and the second multispectral image to obtain a plurality of first single-band images corresponding to the first multispectral image and a plurality of second single-band images corresponding to the second multispectral image;
performing iterative weighting processing on a first single-waveband image and a second single-waveband image corresponding to each waveband to obtain an optimal characteristic difference image corresponding to each waveband;
performing fusion processing on the optimal feature difference images to obtain variation intensity images corresponding to the optimal feature difference images;
and carrying out binary clustering analysis on the change intensity image according to a preset clustering algorithm to obtain a change detection result.
Optionally, after the acquiring the first multispectral image and the second multispectral image, the method further comprises:
and carrying out registration processing on the first multispectral image and the second multispectral image.
Optionally, the performing iterative weighting processing on the first single-band image and the second single-band image corresponding to each band to obtain an optimal feature difference image corresponding to each band includes:
performing iterative weighting processing on a first single-band image and a second single-band image corresponding to each band to obtain a first optimal projection vector and a second optimal projection vector corresponding to each band;
and generating an optimal feature difference image corresponding to each wave band according to the first optimal projection vector and the second optimal projection vector corresponding to each wave band.
Optionally, after the iterative weighting processing is performed on the first single-band image and the second single-band image corresponding to each band to obtain an optimal feature difference image corresponding to each band, the method further includes:
and performing Gaussian filtering processing on the optimal characteristic difference image corresponding to each wave band.
Optionally, the performing fusion processing on the multiple optimal feature difference images to obtain a variation intensity image corresponding to the multiple optimal feature difference images includes:
generating an optimal characteristic difference matrix corresponding to each optimal characteristic difference image;
performing fusion processing on the optimal feature difference matrixes according to an Euclidean distance formula to obtain change intensity matrixes corresponding to the optimal feature difference matrixes;
and generating the change intensity image according to the change intensity matrix.
Optionally, the preset clustering algorithm is a fuzzy C-means FCM clustering algorithm.
In a second aspect, the present invention also provides an apparatus for detecting image changes, the apparatus comprising:
the device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring a first multispectral image and a second multispectral image, and the first multispectral image and the second multispectral image are multispectral images which correspond to the same region in different times;
an extracting unit, configured to perform single-band image extraction processing on the first multispectral image and the second multispectral image acquired by the acquiring unit to obtain a plurality of first single-band images corresponding to the first multispectral image and a plurality of second single-band images corresponding to the second multispectral image;
the iterative weighting unit is used for performing iterative weighting processing on the first single-waveband image and the second single-waveband image corresponding to each waveband to obtain an optimal characteristic difference image corresponding to each waveband;
the fusion unit is used for performing fusion processing on the optimal feature difference images obtained by the iteration weighting unit to obtain variation intensity images corresponding to the optimal feature difference images;
and the analysis unit is used for carrying out binary clustering analysis on the change intensity image obtained by the fusion unit according to a preset clustering algorithm so as to obtain a change detection result.
Optionally, the apparatus further comprises:
and the registration unit is used for performing registration processing on the first multispectral image and the second multispectral image after the acquisition unit acquires the first multispectral image and the second multispectral image.
Optionally, the iterative weighting unit includes:
the iterative weighting module is used for performing iterative weighting processing on the first single-waveband image and the second single-waveband image corresponding to each waveband to obtain a first optimal projection vector and a second optimal projection vector corresponding to each waveband;
and the first generation module is used for generating an optimal feature difference image corresponding to each wave band according to the first optimal projection vector and the second optimal projection vector corresponding to each wave band, which are obtained by the iterative weighting module.
Optionally, the apparatus further comprises:
and the filtering unit is used for performing Gaussian filtering on the optimal characteristic difference image corresponding to each wave band after the iterative weighting unit performs iterative weighting processing on the first single-wave-band image and the second single-wave-band image corresponding to each wave band to obtain the optimal characteristic difference image corresponding to each wave band.
Optionally, the fusion unit includes:
the second generation module is used for generating an optimal characteristic difference matrix corresponding to each optimal characteristic difference image;
the fusion module is used for performing fusion processing on the optimal feature difference matrixes generated by the second generation module according to an Euclidean distance formula so as to obtain change intensity matrixes corresponding to the optimal feature difference matrixes;
and the third generation module is used for generating the change intensity image according to the change intensity matrix obtained by the fusion module.
Optionally, the preset clustering algorithm is a fuzzy C-means FCM clustering algorithm.
In order to achieve the above object, according to a third aspect of the present invention, there is provided a storage medium including a stored program, wherein when the program runs, an apparatus in which the storage medium is located is controlled to execute the above method for detecting an image change.
In order to achieve the above object, according to a fourth aspect of the present invention, a processor for executing a program is provided, wherein the program executes the method for detecting image change.
By the technical scheme, the technical scheme provided by the invention at least has the following advantages:
the invention provides a method and a device for detecting image change, compared with the prior art that an IR-MAD algorithm is adopted to detect the change of two multispectral images which are different and correspond to the same region, the invention can extract a first single-band image corresponding to each wave band from a first multispectral image and a second single-band image corresponding to each wave band from a second multispectral image after a first multispectral image and a second multispectral image which are different and correspond to the same region are obtained, iterative weighting processing is carried out on the first single-band image and the second single-band image corresponding to each wave band, so as to obtain an optimal characteristic difference image corresponding to each wave band, fusion processing is carried out on a plurality of optimal characteristic difference images, so as to obtain a plurality of variable intensity images corresponding to the optimal characteristic difference images, binary cluster analysis is carried out on the variable intensity images according to a preset clustering algorithm, and obtaining a change detection result. Because the iterative weighting processing is respectively carried out on the first single-waveband image and the second single-waveband image corresponding to each waveband, the correlation between the first single-waveband image and the second single-waveband image corresponding to each waveband can be effectively improved, so that the optimal characteristic difference image corresponding to each waveband can be obtained, and the accuracy of the obtained change detection result is higher by carrying out binary cluster analysis on the change intensity image obtained by fusing the optimal characteristic difference images.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
Drawings
Various additional advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
FIG. 1 is a flow chart of a method for detecting image changes according to an embodiment of the present invention;
FIG. 2 is a flow chart illustrating another method for detecting image changes provided by embodiments of the present invention;
FIG. 3 is a block diagram of an apparatus for detecting image changes according to an embodiment of the present invention;
fig. 4 is a block diagram illustrating another apparatus for detecting image changes according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
An embodiment of the present invention provides a method for detecting an image change, as shown in fig. 1, where the method includes:
101. and acquiring a first multispectral image and a second multispectral image.
The first multispectral image and the second multispectral image are multispectral images which correspond to different regions in the same region; the multispectral image is an image synthesized by a plurality of single-band images of different bands, and the plurality of bands may be, but is not limited to: red band, blue band, green band, near infrared band, and the like.
In the embodiment of the present invention, first, a first multispectral image and a second multispectral image, which correspond to different regions in the same region, need to be obtained first, so as to detect image changes between the first multispectral image and the second multispectral image subsequently.
102. And performing single-band image extraction processing on the first multispectral image and the second multispectral image to obtain a plurality of first single-band images corresponding to the first multispectral image and a plurality of second single-band images corresponding to the second multispectral image.
In the embodiment of the present invention, since the multispectral image is an image synthesized from a plurality of single-band images of different bands, after the first multispectral image and the second multispectral image are obtained, the single-band image extraction processing may be performed on the first multispectral image and the second multispectral image, so as to obtain a plurality of first single-band images corresponding to the first multispectral image and a plurality of second single-band images corresponding to the second multispectral image, that is, a first single-band image corresponding to each band is extracted from the first multispectral image and a second single-band image corresponding to each band is extracted from the second multispectral image.
103. And performing iterative weighting processing on the first single-band image and the second single-band image corresponding to each band to obtain an optimal characteristic difference image corresponding to each band.
The optimal characteristic difference image corresponding to any one waveband comprises the change intensity information between corresponding pixel points in the first single-waveband image and the second single-waveband image corresponding to the waveband.
In the embodiment of the present invention, after obtaining a plurality of first single-band images corresponding to the first multispectral image and a plurality of second single-band images corresponding to the second multispectral image, the first single-band image and the second single-band image corresponding to each band are respectively weighted iteratively to obtain an optimal feature difference image corresponding to each band, for example, in the foregoing steps, the first single-band image corresponding to the red band, the first single-band image corresponding to the blue band, the first single-band image corresponding to the green band, and the first single-band image corresponding to the near-infrared band are extracted from the first multispectral image, and the second single-band image corresponding to the red band, the second single-band image corresponding to the blue band, the second single-band image corresponding to the green band, and the second single-band image corresponding to the near-infrared band are extracted from the second multispectral image, at the moment, iterative weighting processing can be carried out on the first single-band image and the second single-band image corresponding to the red band, so that an optimal characteristic difference image corresponding to the red band is obtained; performing iterative weighting processing on a first single-waveband image and a second single-waveband image corresponding to a blue waveband to obtain an optimal characteristic difference image corresponding to the blue waveband; performing iterative weighting processing on a first single-band image and a second single-band image corresponding to a green band to obtain an optimal characteristic difference image corresponding to the green band; and carrying out iterative weighting processing on the first single-band image and the second single-band image corresponding to the near-infrared band so as to obtain the optimal characteristic difference image corresponding to the near-infrared band.
It should be noted that, by performing iterative weighting processing on the first single-band image and the second single-band image corresponding to each band, the correlation between the first single-band image and the second single-band image corresponding to each band can be effectively improved, so that the optimal feature difference image corresponding to each band can be obtained.
104. And carrying out fusion processing on the optimal characteristic difference images to obtain the change intensity images corresponding to the optimal characteristic difference images.
In the embodiment of the invention, after the optimal characteristic difference image corresponding to each wave band is obtained, the optimal characteristic difference images can be fused, so that the change intensity images corresponding to the optimal characteristic difference images are obtained, namely the change intensity images corresponding to the first multispectral image and the second multispectral image are obtained, wherein the change intensity images comprise change intensity information between corresponding pixel points in the first multispectral image and the second multispectral image. Specifically, in this step, the plurality of optimal feature difference images may be fused according to the euclidean distance formula, so as to obtain the variation intensity images corresponding to the plurality of optimal feature difference images, but the present invention is not limited thereto.
105. And carrying out binary clustering analysis on the change intensity image according to a preset clustering algorithm to obtain a change detection result.
The preset clustering algorithm may specifically be an FCM clustering algorithm.
In the embodiment of the present invention, after obtaining the variation intensity images corresponding to the first multispectral image and the second multispectral image, the variation intensity images are subjected to binary clustering analysis according to a preset clustering algorithm (i.e., the preset clustering algorithm is used to perform binary clustering analysis on a plurality of variation intensity information included in the variation intensity images to determine whether there is a variation between corresponding pixel points in the first multispectral image and the second multispectral image), and at this time, variation detection results corresponding to the first multispectral image and the second multispectral image can be obtained.
The embodiment of the invention provides a method for detecting image change, compared with the prior art that an IR-MAD algorithm is adopted to detect the change of two multispectral images which correspond to the same region in different times, the embodiment of the invention can extract a first single-band image corresponding to each wave band from a first multispectral image and a second single-band image corresponding to each wave band from a second multispectral image after a first multispectral image and a second multispectral image which correspond to the same region in different times are obtained, iterative weighting processing is carried out on the first single-band image and the second single-band image corresponding to each wave band, so as to obtain an optimal characteristic difference image corresponding to each wave band, binary clustering analysis is carried out on the change intensity image according to a preset clustering algorithm after fusion processing is carried out on a plurality of optimal characteristic difference images, so as to obtain the change intensity images corresponding to the optimal characteristic difference images, and obtaining a change detection result. Because the iterative weighting processing is respectively carried out on the first single-band image and the second single-band image corresponding to each band, the correlation between the first single-band image and the second single-band image corresponding to each band can be effectively improved, so that the optimal characteristic difference image corresponding to each band can be obtained, and the accuracy of the obtained change detection result is higher by carrying out binary cluster analysis on the change intensity image obtained by fusing the optimal characteristic difference images.
To be described in more detail below, an embodiment of the present invention provides another method for detecting image changes, and in particular, a specific method for performing iterative weighting processing on a first single-band image and a second single-band image corresponding to each band to obtain an optimal feature difference image corresponding to each band, and a specific method for performing fusion processing on a plurality of optimal feature difference images to obtain change intensity images corresponding to the plurality of optimal feature difference images, as shown in fig. 2 specifically, the method includes:
201. and acquiring a first multispectral image and a second multispectral image.
In step 201, the first multispectral image and the second multispectral image may be obtained by referring to the description of the corresponding part in fig. 1, and the details of the embodiment of the present invention will not be repeated herein.
202. And carrying out registration processing on the first multispectral image and the second multispectral image.
In the embodiment of the present invention, in order to ensure that a plurality of pixel points included in the first multispectral image correspond to a plurality of pixel points included in the second multispectral image one to one, after the first multispectral image and the second multispectral image are obtained, the first multispectral image and the second multispectral image need to be registered. Specifically, in this step, the remote sensing image processing software ENVI may be used to perform registration processing on the first multispectral image and the second multispectral image, but is not limited thereto.
203. And performing single-band image extraction processing on the first multispectral image and the second multispectral image to obtain a plurality of first single-band images corresponding to the first multispectral image and a plurality of second single-band images corresponding to the second multispectral image.
In step 203, the first multispectral image and the second multispectral image are subjected to a single-band image extraction process to obtain a plurality of first single-band images corresponding to the first multispectral image and a plurality of second single-band images corresponding to the second multispectral image, which may refer to the description of the corresponding part in fig. 1, and the details of the embodiment of the present invention will not be repeated herein.
204. And performing iterative weighting processing on the first single-band image and the second single-band image corresponding to each band to obtain an optimal characteristic difference image corresponding to each band.
In the embodiment of the present invention, after obtaining a plurality of first single-band images corresponding to the first multispectral image and a plurality of second single-band images corresponding to the second multispectral image, iterative weighting processing may be performed on the first single-band image and the second single-band image corresponding to each band, so as to obtain an optimal feature difference image corresponding to each band. The following describes how to perform iterative weighting processing on the first single-band image and the second single-band image corresponding to each band to obtain an optimal feature difference image corresponding to each band.
(1) And performing iterative weighting processing on the first single-band image and the second single-band image corresponding to each band to obtain a first optimal projection vector and a second optimal projection vector corresponding to each band.
In the embodiment of the present invention, after obtaining a plurality of first single-band images corresponding to the first multispectral image and a plurality of second single-band images corresponding to the second multispectral image, iterative weighting processing may be performed on the first single-band image and the second single-band image corresponding to each band, respectively, so as to obtain a first optimal projection vector and a second optimal projection vector corresponding to each band.
Specifically, in this step, the iterative weighting process for the first single-band image and the second single-band image corresponding to each band is as follows:
1. generating a first image matrix Fp corresponding to the first single-waveband image and a second image matrix Gp corresponding to the second single-waveband image according to the first single-waveband image and the second single-waveband image corresponding to the P-th waveband;
2. respectively calculating covariance matrixes corresponding to the first image matrix Fp according to the first image matrix Fp and the second image matrix Gp
Covariance matrix corresponding to the second image matrix Gp
And a first image matrix F p And a second image matrix G p Corresponding cross covariance matrix
And
3. the covariance matrix
Covariance matrix
Cross covariance matrix
Sum cross covariance matrix
Respectively substituted into a first preset formula and a second preset formula to calculate a first projection vector a corresponding to the first image matrix Fp p A second projection vector b corresponding to the second image matrix Gp p And the correlation between the first image matrix Fp and the second image matrix Gp
The first preset formula is specifically as follows:
the second preset formula is specifically as follows:
4. according to the first image matrix Fp, the second image matrix Gp and the first projection vector a p And a second projection vector b p Calculating a feature difference matrix Mp corresponding to the first image matrix Fp and the second image matrix Gp, i.e.
5. Calculating chi-square distance T corresponding to the characteristic difference matrix Mp according to the characteristic difference matrix Mp ij I.e. T ij ＝(Mp ij /σ p ) 2 ∈x 2 (n) wherein, chi-square distance T ij Satisfy the chi-square distribution with the degree of freedom n, sigma p The variances corresponding to the first image matrix Fp and the second image matrix Gp;
6. according to chi-square distance T ij Probability density quantile point calculation weight value omega of chi-square distribution ij I.e. omega ij ＝P(T ij >t)＝P(x 2 (n)>T ij )；
7. Repeating the steps 2-6 to perform multiple iterative weighted calculations, wherein in the process of performing the second iterative weighted calculation and performing any one iterative weighted calculation after the second iterative weighted calculation, the covariance matrix is calculated
Covariance matrix
Cross covariance matrix
Sum cross covariance matrix
In the process, any calculation related to variance and mean value needs to use the weight value omega obtained in the last iteration weighting calculation process ij Carrying out weighting processing; when the preset iteration times are reached or the difference value of the obtained correlation rho in the processes of two adjacent iteration weighting calculation is smaller than a preset threshold value, the iteration weighting processing is finished, and the first projection vector a obtained in the process of the last iteration weighting calculation is used p And a second projection vector b p Determining a first optimal projection vector and a second optimal projection vector corresponding to the pth band, where the preset iteration number may be, but is not limited to: 9 times, 10 times, 11 times, etc., and the preset threshold may be, but is not limited to: 0.0001, 0.0002, 0.0003, and the like;
8. and (3) performing iterative weighting processing on the first single-band image and the second single-band image corresponding to other bands by adopting the method in the step (1-7), so as to obtain a first optimal projection vector and a second optimal projection vector corresponding to each band.
(2) And generating an optimal characteristic difference image corresponding to each wave band according to the first optimal projection vector and the second optimal projection vector corresponding to each wave band.
In the embodiment of the invention, after the first optimal projection vector and the second optimal projection vector corresponding to each band are respectively obtained, the optimal feature difference image corresponding to each band can be generated according to the first optimal projection vector and the second optimal projection vector corresponding to each band.
Specifically, in this step, a process of generating an optimal feature difference image corresponding to each band according to the first optimal projection vector and the second optimal projection vector corresponding to each band is as follows:
1. acquiring a first image matrix Fp, a second image matrix Gp and a first optimal projection vector a corresponding to the P wave band p And a second optimal projection vector b p ；
2. According to the first image matrix Fp, the second image matrix Gp and the first optimal projection vector a p And a second optimal projection vector b p Calculating the optimal characteristic difference matrix corresponding to the first image matrix Fp and the second image matrix GpMp, i.e.
3. Generating an optimal characteristic difference image corresponding to the P wave band according to the optimal characteristic difference matrix Mp;
4. and (4) generating an optimal characteristic difference image corresponding to each wave band according to the first optimal projection vector and the second optimal projection vector corresponding to other wave bands by adopting the method in the step (1-3).
205. And carrying out Gaussian filtering processing on the optimal characteristic difference image corresponding to each wave band.
In the embodiment of the present invention, in order to reduce the influence of factors such as an external environment on the plurality of optimal feature difference images, after the plurality of optimal feature difference images are obtained, gaussian filtering processing needs to be performed on each optimal feature difference image (that is, the optimal feature difference image corresponding to each band).
206. And carrying out fusion processing on the optimal characteristic difference images to obtain the change intensity images corresponding to the optimal characteristic difference images.
In the embodiment of the invention, after the optimal feature difference images corresponding to each band are subjected to the gaussian filtering, the optimal feature difference images subjected to the gaussian filtering are subjected to fusion processing, so that the change intensity images corresponding to the optimal feature difference images are obtained, namely the change intensity images corresponding to the first multispectral image and the second multispectral image are obtained. How to perform the fusion processing on the plurality of optimal feature difference images to obtain the variation intensity images corresponding to the plurality of optimal feature difference images will be described in detail below.
(1) And generating an optimal characteristic difference matrix corresponding to each optimal characteristic difference image.
In the embodiment of the present invention, in order to fuse a plurality of optimal feature difference images into a variation intensity image, an optimal feature difference matrix corresponding to each optimal feature difference image needs to be generated.
(2) And performing fusion processing on the optimal feature difference matrixes according to an Euclidean distance formula to obtain the change intensity matrixes corresponding to the optimal feature difference matrixes.
In the embodiment of the invention, after the optimal feature difference matrix corresponding to each optimal feature difference image is generated, the multiple optimal feature difference matrices can be fused according to the Euclidean distance formula, so that the change intensity matrices corresponding to the multiple optimal feature difference matrices are obtained. Specifically, in this step, a plurality of optimal feature difference matrices may be adjusted to column vectors; then, writing a plurality of column vectors into a preset empty matrix, thereby obtaining a summary feature difference matrix; and finally, converting the summary feature difference matrix into a variation intensity matrix by using an Euclidean distance formula.
(3) A varying intensity image is generated from the varying intensity matrix.
In the embodiment of the invention, after the change intensity matrixes corresponding to the optimal feature difference matrixes are obtained, the change intensity images corresponding to the first multispectral image and the second multispectral image can be generated according to the change intensity matrixes.
207. And carrying out binary clustering analysis on the change intensity image according to a preset clustering algorithm to obtain a change detection result.
In step 207, performing binary clustering analysis on the variation intensity image according to a preset clustering algorithm to obtain a variation detection result may refer to the description of the corresponding part in fig. 1, and will not be described herein again in the embodiments of the present invention.
In order to achieve the above object, according to another aspect of the present invention, an embodiment of the present invention further provides a storage medium, where the storage medium includes a stored program, and when the program runs, a device on which the storage medium is located is controlled to execute the method for detecting an image change described above.
In order to achieve the above object, according to another aspect of the present invention, an embodiment of the present invention further provides a processor for running a program, where the program runs to execute the method for detecting an image change described above.
Further, as an implementation of the method shown in fig. 1 and fig. 2, another embodiment of the present invention further provides an apparatus for detecting an image change. The embodiment of the apparatus corresponds to the embodiment of the method, and for convenience of reading, details in the embodiment of the apparatus are not repeated one by one, but it should be clear that the apparatus in the embodiment can correspondingly implement all the contents in the embodiment of the method. The device is applied to the change detection of two multispectral images which are different and corresponding in the same region, and the accuracy of the change detection result is improved, and particularly as shown in fig. 3, the device comprises:
the acquiring unit 31 is configured to acquire a first multispectral image and a second multispectral image, where the first multispectral image and the second multispectral image are multispectral images corresponding to different regions of the same region;
an extracting unit 32, configured to perform single-band image extraction processing on the first multispectral image and the second multispectral image acquired by the acquiring unit 31 to obtain a plurality of first single-band images corresponding to the first multispectral image and a plurality of second single-band images corresponding to the second multispectral image;
the iterative weighting unit 33 is configured to perform iterative weighting processing on the first single-band image and the second single-band image corresponding to each band to obtain an optimal feature difference image corresponding to each band;
a fusion unit 34, configured to perform fusion processing on the multiple optimal feature difference images obtained by the iterative weighting unit 33 to obtain change intensity images corresponding to the multiple optimal feature difference images;
and the analysis unit 35 is configured to perform binary clustering analysis on the change intensity image obtained by the fusion unit 34 according to a preset clustering algorithm to obtain a change detection result.
Further, as shown in fig. 4, the apparatus further includes:
the registration unit 36 is configured to perform registration processing on the first multispectral image and the second multispectral image after the acquisition unit 31 acquires the first multispectral image and the second multispectral image.
Further, as shown in fig. 4, the iterative weighting unit 33 includes:
the iterative weighting module 331 is configured to perform iterative weighting processing on the first single-band image and the second single-band image corresponding to each band to obtain a first optimal projection vector and a second optimal projection vector corresponding to each band;
the first generating module 332 is configured to generate an optimal feature difference image corresponding to each of the wavelength bands according to the first optimal projection vector and the second optimal projection vector corresponding to each of the wavelength bands obtained by the iterative weighting module 331.
Further, as shown in fig. 4, the apparatus further includes:
the filtering unit 37 is configured to perform, after the iterative weighting unit 33 performs iterative weighting processing on the first single-band image and the second single-band image corresponding to each band to obtain an optimal feature difference image corresponding to each band, perform gaussian filtering processing on the optimal feature difference image corresponding to each band.
Further, as shown in fig. 4, the fusion unit 34 includes:
a second generating module 341, configured to generate an optimal feature difference matrix corresponding to each optimal feature difference image;
a fusion module 342, configured to perform fusion processing on the multiple optimal feature difference matrices generated by the second generation module 341 according to an euclidean distance formula to obtain a variation strength matrix corresponding to the multiple optimal feature difference matrices;
a third generating module 343, configured to generate the varying intensity image according to the varying intensity matrix obtained by the fusing module 342.
Further, as shown in fig. 4, the preset clustering algorithm is a fuzzy C-means FCM clustering algorithm.
The embodiment of the invention provides a method and a device for detecting image change, compared with the prior art that an IR-MAD algorithm is adopted to detect the change of two multispectral images which correspond to the same region in different times, the embodiment of the invention can extract a first single-band image corresponding to each wave band from a first multispectral image and a second single-band image corresponding to each wave band from a second multispectral image after acquiring a first multispectral image and a second multispectral image which correspond to the same region in different times, and perform iterative weighting processing on the first single-band image and the second single-band image corresponding to each wave band so as to acquire an optimal characteristic difference image corresponding to each wave band, and after performing fusion processing on a plurality of optimal characteristic difference images so as to acquire a change intensity image corresponding to the optimal characteristic difference images, and carrying out binary clustering analysis on the change intensity image according to a preset clustering algorithm to obtain a change detection result. Because the iterative weighting processing is respectively carried out on the first single-band image and the second single-band image corresponding to each band, the correlation between the first single-band image and the second single-band image corresponding to each band can be effectively improved, so that the optimal characteristic difference image corresponding to each band can be obtained, and the accuracy of the obtained change detection result is higher by carrying out binary cluster analysis on the change intensity image obtained by fusing the optimal characteristic difference images.
The device for detecting the image change comprises a processor and a memory, wherein the acquisition unit, the extraction unit, the iteration weighting unit, the fusion unit, the analysis unit and the like are stored in the memory as program units, and the processor executes the program units stored in the memory to realize corresponding functions.
The processor comprises a kernel, and the kernel calls the corresponding program unit from the memory. One or more than one kernel can be set, and the accuracy of a change detection result is improved by adjusting kernel parameters when two multispectral images which are different and corresponding to the same region are subjected to change detection.
The memory may include volatile memory in a computer readable medium, Random Access Memory (RAM) and/or nonvolatile memory such as Read Only Memory (ROM) or flash memory (flash RAM), and the memory includes at least one memory chip.
An embodiment of the present invention provides a storage medium having a program stored thereon, where the program is executed by a processor to implement the method for detecting image change described in any one of the above embodiments.
An embodiment of the present invention provides a processor, where the processor is configured to execute a program, where the program executes the method for detecting an image change in any one of the above embodiments when the program is executed.
The embodiment of the invention provides equipment, which comprises a processor, a memory and a program which is stored on the memory and can run on the processor, wherein the processor executes the program and realizes the following steps:
acquiring a first multispectral image and a second multispectral image, wherein the first multispectral image and the second multispectral image are multispectral images which are different and correspond to each other in the same region;
performing single-band image extraction processing on the first multispectral image and the second multispectral image to obtain a plurality of first single-band images corresponding to the first multispectral image and a plurality of second single-band images corresponding to the second multispectral image;
performing iterative weighting processing on a first single-band image and a second single-band image corresponding to each band to obtain an optimal characteristic difference image corresponding to each band;
performing fusion processing on the optimal feature difference images to obtain variation intensity images corresponding to the optimal feature difference images;
and carrying out binary clustering analysis on the change intensity image according to a preset clustering algorithm to obtain a change detection result.
Further, after the acquiring the first and second multispectral images, the method further comprises:
and carrying out registration processing on the first multispectral image and the second multispectral image.
Further, the performing iterative weighting processing on the first single-band image and the second single-band image corresponding to each band to obtain an optimal feature difference image corresponding to each band includes:
performing iterative weighting processing on a first single-band image and a second single-band image corresponding to each band to obtain a first optimal projection vector and a second optimal projection vector corresponding to each band;
and generating an optimal feature difference image corresponding to each wave band according to the first optimal projection vector and the second optimal projection vector corresponding to each wave band.
Further, after the iterative weighting processing is performed on the first single-band image and the second single-band image corresponding to each band to obtain an optimal feature difference image corresponding to each band, the method further includes:
and performing Gaussian filtering processing on the optimal characteristic difference image corresponding to each wave band.
Further, the performing fusion processing on the optimal feature difference images to obtain the variation intensity images corresponding to the optimal feature difference images includes:
generating an optimal characteristic difference matrix corresponding to each optimal characteristic difference image;
performing fusion processing on the optimal feature difference matrixes according to an Euclidean distance formula to obtain change intensity matrixes corresponding to the optimal feature difference matrixes;
and generating the change intensity image according to the change intensity matrix.
Further, the preset clustering algorithm is a fuzzy C-means FCM clustering algorithm.
The present application further provides a computer program product adapted to perform program code for initializing the following method steps when executed on a data processing device: acquiring a first multispectral image and a second multispectral image, wherein the first multispectral image and the second multispectral image are multispectral images which are different and correspond to each other in the same region; performing single-band image extraction processing on the first multispectral image and the second multispectral image to obtain a plurality of first single-band images corresponding to the first multispectral image and a plurality of second single-band images corresponding to the second multispectral image; performing iterative weighting processing on a first single-band image and a second single-band image corresponding to each band to obtain an optimal characteristic difference image corresponding to each band; performing fusion processing on the optimal feature difference images to obtain variation intensity images corresponding to the optimal feature difference images; and carrying out binary clustering analysis on the change intensity image according to a preset clustering algorithm to obtain a change detection result.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and so forth) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). The memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in the process, method, article, or apparatus that comprises the element.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (9)
1. A method of detecting image changes, comprising:
acquiring a first multispectral image and a second multispectral image, wherein the first multispectral image and the second multispectral image are multispectral images which are different and correspond to each other in the same region;
performing single-band image extraction processing on the first multispectral image and the second multispectral image to obtain a plurality of first single-band images corresponding to the first multispectral image and a plurality of second single-band images corresponding to the second multispectral image;
performing iterative weighting processing on a first single-band image and a second single-band image corresponding to each band to obtain an optimal characteristic difference image corresponding to each band;
performing fusion processing on the optimal feature difference images to obtain variation intensity images corresponding to the optimal feature difference images;
performing binary clustering analysis on the change intensity image according to a preset clustering algorithm to obtain a change detection result;
the iterative weighting processing is performed on the first single-band image and the second single-band image corresponding to each band to obtain an optimal characteristic difference image corresponding to each band, and the iterative weighting processing includes:
performing iterative weighting processing on a first single-band image and a second single-band image corresponding to each band to obtain a first optimal projection vector and a second optimal projection vector corresponding to each band;
generating an optimal feature difference image corresponding to each wave band according to the first optimal projection vector and the second optimal projection vector corresponding to each wave band;
the iterative weighting processing is performed on the first single-band image and the second single-band image corresponding to each band to obtain a first optimal projection vector and a second optimal projection vector corresponding to each band, and the iterative weighting processing includes:
(1) generating a first image matrix Fp corresponding to the first single-waveband image and a second image matrix Gp corresponding to the second single-waveband image according to the first single-waveband image and the second single-waveband image corresponding to the P-th waveband;
(2) respectively calculating covariance matrixes corresponding to the first image matrix Fp according to the first image matrix Fp and the second image matrix Gp
Covariance matrix corresponding to the second image matrix Gp
And a first image matrix F p And a second image matrix G p Corresponding cross covariance matrix
And
(3) the covariance matrix
Covariance matrix
Cross covariance matrix
Sum cross covariance matrix
Respectively substituted into a first preset formula and a second preset formula to calculate a first projection vector a corresponding to the first image matrix Fp p A second projection vector b corresponding to the second image matrix Gp p And the correlation between the first image matrix Fp and the second image matrix Gp
The first preset formula is specifically as follows:
the second preset formula is specifically as follows:
(4) according to the first image matrix Fp, the second image matrix Gp and the first projection vector a p And a second projection vector b p Calculating a feature difference matrix Mp corresponding to the first image matrix Fp and the second image matrix Gp, i.e.
(5) Calculating the chi-square distance T corresponding to the characteristic difference matrix Mp according to the characteristic difference matrix Mp ij I.e. T ij ＝(Mp ij /σ p ) 2 ∈x 2 (n) wherein, chi-square distance T ij Satisfy the chi-square distribution with the degree of freedom n, sigma p The variances corresponding to the first image matrix Fp and the second image matrix Gp;
(6) according to chi-square distance T ij Probability density quantile point calculation weight value omega of chi-square distribution ij I.e. omega ij ＝P(T ij >t)＝P(x 2 (n)>T ij )；
(7) Repeating the steps 2-6 to perform multiple iterative weighted calculations, wherein in the process of performing the second iterative weighted calculation and performing any one iterative weighted calculation after the second iterative weighted calculation, the covariance matrix is calculated
Covariance matrix
Cross covariance matrix
Sum cross covariance matrix
In time, any calculation involving variance and mean requires the use of the weight value ω found in the last iterative weighting calculation ij Carrying out weighting processing; when the preset iteration times are reached or the difference value of the obtained correlation rho in the processes of two adjacent iteration weighting calculation is smaller than a preset threshold value, the iteration weighting processing is finished, and the first projection vector a obtained in the process of the last iteration weighting calculation is used p And a second projection vector b p Determining a first optimal projection vector and a second optimal projection vector corresponding to the P wave band;
(8) performing iterative weighting processing on the first single-band image and the second single-band image corresponding to other bands by adopting the method in the step 1-7, so as to obtain a first optimal projection vector and a second optimal projection vector corresponding to each band;
the generating an optimal feature difference image corresponding to each of the bands according to the first optimal projection vector and the second optimal projection vector corresponding to each of the bands includes:
(1) acquiring a first image matrix Fp, a second image matrix Gp and a first optimal projection vector a corresponding to the P wave band p And a second optimal projection vector b p ；
(2) According to the first image matrix Fp, the second image matrix Gp and the first optimal projection vector a p And a second optimal projection vector b p Calculating the optimal feature difference matrix Mp corresponding to the first image matrix Fp and the second image matrix Gp, i.e. calculating the optimal feature difference matrix Mp
(3) Generating an optimal characteristic difference image corresponding to the P-th wave band according to the optimal characteristic difference matrix Mp;
(4) and (4) generating an optimal characteristic difference image corresponding to each wave band according to the first optimal projection vector and the second optimal projection vector corresponding to other wave bands by adopting the method in the step (1-3).
2. The method according to claim 1, wherein after said acquiring a first multispectral image and a second multispectral image, said method further comprises:
and carrying out registration processing on the first multispectral image and the second multispectral image.
3. The method of claim 1, wherein after the iteratively weighting the first and second one-band images for each band to obtain the optimal feature difference image for each band, the method further comprises:
and performing Gaussian filtering processing on the optimal characteristic difference image corresponding to each wave band.
4. The method according to claim 1, wherein the fusing the optimal feature difference images to obtain the variation intensity images corresponding to the optimal feature difference images comprises:
generating an optimal characteristic difference matrix corresponding to each optimal characteristic difference image;
performing fusion processing on the optimal feature difference matrixes according to an Euclidean distance formula to obtain change intensity matrixes corresponding to the optimal feature difference matrixes;
and generating the change intensity image according to the change intensity matrix.
5. The method according to any of claims 1-4, wherein the pre-set clustering algorithm is a fuzzy C-means FCM clustering algorithm.
6. An apparatus for detecting image changes, comprising:
the device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring a first multispectral image and a second multispectral image, and the first multispectral image and the second multispectral image are multispectral images which correspond to each other in the same region when the same region is different;
an extracting unit, configured to perform single-band image extraction processing on the first multispectral image and the second multispectral image acquired by the acquiring unit to obtain a plurality of first single-band images corresponding to the first multispectral image and a plurality of second single-band images corresponding to the second multispectral image;
the iterative weighting unit is used for performing iterative weighting processing on the first single-waveband image and the second single-waveband image corresponding to each waveband to obtain an optimal characteristic difference image corresponding to each waveband;
the fusion unit is used for performing fusion processing on the optimal feature difference images obtained by the iteration weighting unit to obtain variation intensity images corresponding to the optimal feature difference images;
the analysis unit is used for carrying out binary clustering analysis on the change intensity image obtained by the fusion unit according to a preset clustering algorithm so as to obtain a change detection result;
the iterative weighting unit includes:
the iterative weighting module is used for performing iterative weighting processing on a first single-band image and a second single-band image corresponding to each band so as to obtain a first optimal projection vector and a second optimal projection vector corresponding to each band;
the first generation module is used for generating an optimal feature difference image corresponding to each wave band according to the first optimal projection vector and the second optimal projection vector corresponding to each wave band obtained by the iterative weighting module;
the iterative weighting module is specifically configured to:
(1) generating a first image matrix Fp corresponding to the first single-waveband image and a second image matrix Gp corresponding to the second single-waveband image according to the first single-waveband image and the second single-waveband image corresponding to the P-th waveband;
(2) respectively calculating covariance matrixes corresponding to the first image matrix Fp according to the first image matrix Fp and the second image matrix Gp
Covariance matrix corresponding to the second image matrix Gp
And a first image matrix F p And a second image matrix G p Corresponding cross covariance matrix
And
(3) the covariance matrix
Covariance matrix
Cross covariance matrix
Sum cross covariance matrix
Respectively substituted into a first preset formula and a second preset formula to calculate a first projection vector a corresponding to the first image matrix Fp p A second projection vector b corresponding to the second image matrix Gp p And the correlation between the first image matrix Fp and the second image matrix Gp
The first preset formula is specifically as follows:
the second preset formula is specifically as follows:
(4) according to the first image matrix Fp, the second image matrix Gp and the first projection vector a p And a second projection vector b p Calculating a feature difference matrix Mp corresponding to the first image matrix Fp and the second image matrix Gp, i.e.
(5) Calculating the correspondence of the characteristic difference matrix Mp according to the characteristic difference matrix MpChi-square distance T of ij I.e. T ij ＝(Mp ij /σ p ) 2 ∈x 2 (n) wherein, chi-square distance T ij Satisfy the chi-square distribution with the degree of freedom n, sigma p The variances corresponding to the first image matrix Fp and the second image matrix Gp;
(6) according to chi-square distance T ij Calculating weight value omega by using probability density quantile points of chi-square distribution ij I.e. omega ij ＝P(T ij >t)＝P(x 2 (n)>T ij )；
(7) Repeating the steps 2-6 to perform multiple iterative weighted calculations, wherein in the process of performing the second iterative weighted calculation and performing any one iterative weighted calculation after the second iterative weighted calculation, the covariance matrix is calculated
Covariance matrix
Cross covariance matrix
Sum cross covariance matrix
In time, any calculation involving variance and mean requires the use of the weight value ω found in the last iterative weighting calculation ij Carrying out weighting processing; when the preset iteration times are reached or the difference value of the obtained correlation rho in the processes of two adjacent iteration weighting calculation is smaller than a preset threshold value, the iteration weighting processing is finished, and the first projection vector a obtained in the process of the last iteration weighting calculation is used p And a second projection vector b p Determining a first optimal projection vector and a second optimal projection vector corresponding to the P wave band;
(8) performing iterative weighting processing on the first single-band image and the second single-band image corresponding to other bands by adopting the method in the step 1-7, so as to obtain a first optimal projection vector and a second optimal projection vector corresponding to each band;
the first generating module is specifically configured to:
(1) acquiring a first image matrix Fp, a second image matrix Gp and a first optimal projection vector a corresponding to the P wave band p And a second optimal projection vector b p ；
(2) According to the first image matrix Fp, the second image matrix Gp and the first optimal projection vector a p And a second optimal projection vector b p Calculating the optimal feature difference matrix Mp corresponding to the first image matrix Fp and the second image matrix Gp, i.e. calculating the optimal feature difference matrix Mp
(3) Generating an optimal characteristic difference image corresponding to the P-th wave band according to the optimal characteristic difference matrix Mp;
(4) and (4) generating an optimal characteristic difference image corresponding to each wave band according to the first optimal projection vector and the second optimal projection vector corresponding to other wave bands by adopting the method in the step (1-3).
7. The apparatus of claim 6, further comprising:
and the registration unit is used for performing registration processing on the first multispectral image and the second multispectral image after the acquisition unit acquires the first multispectral image and the second multispectral image.
8. A storage medium, characterized in that the storage medium comprises a stored program, wherein when the program runs, a device where the storage medium is located is controlled to execute the method for detecting image change according to any one of claims 1 to 5.
9. A processor, characterized in that the processor is configured to run a program, wherein the program when running performs the method of detecting image changes of any one of claims 1 to 5.
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 CN109903246A (en) | 2019-06-18 |
| CN109903246B true 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) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112633155B (en) * | 2020-12-22 | 2021-08-31 | 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 |
| 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 |
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 |
|---|
| The Regularized Iteratively Reweighted MAD Method for Change Detection in Multi- and Hyperspectral Data;Allan Aasbjerg Nielsen;《IEEE TRANSACTIONS ON IMAGE PROCESSING》;20070228;463-478 * |
| 基于组合差异图和FCM聚类的SAR图像变化检测;周文艳;《激光杂志》;20180325;89-93 * |
| 基于选权迭代估计与非监督分类的多光谱图像变化检测;李莎;《国土资源遥感》;20140917;34-40 * |
| 多光谱遥感图像的变化检测算法研究;马利媛;《中国优秀硕士学位论文全文数据库 (工程科技Ⅱ辑)》;20210731;C028-71 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109903246A (en) | 2019-06-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114092820B (en) | Target detection method and moving target tracking method applying same | |
| US10896351B2 (en) | Active machine learning for training an event classification | |
| US9600746B2 (en) | Image processing apparatus and image processing method | |
| CN106355188B (en) | Image detection method and device | |
| CN109446889B (en) | Object tracking method and device based on twin matching network | |
| US9292745B2 (en) | Object detection apparatus and method therefor | |
| CN109903246B (en) | Method and device for detecting image change | |
| CN110023989B (en) | Sketch image generation method and device | |
| US20180089537A1 (en) | Learning apparatus, image identification apparatus, learning method, image identification method, and storage medium | |
| CN113361495A (en) | Face image similarity calculation method, device, equipment and storage medium | |
| Klein et al. | Salient pattern detection using W 2 on multivariate normal distributions | |
| CN112950685A (en) | Infrared and visible light image registration method, system and storage medium | |
| CN107316296B (en) | Remote sensing image change detection method and device based on logarithmic transformation | |
| CN112364916A (en) | Image classification method based on transfer learning, related equipment and storage medium | |
| CN113723352A (en) | Text detection method, system, storage medium and electronic equipment | |
| Paul et al. | Multimodal optical image registration using modified SIFT | |
| Favorskaya et al. | Fast salient object detection in non-stationary video sequences based on spatial saliency maps | |
| CN114119970B (en) | Target tracking method and device | |
| CN113963236A (en) | Target detection method and device | |
| Negri et al. | A general and extensible framework for assessing change detection techniques | |
| Aria et al. | Spectral region identification versus individual channel selection in supervised dimensionality reduction of hyperspectral image data | |
| Papadomanolaki et al. | Stacked encoder-decoders for accurate semantic segmentation of very high resolution satellite datasets | |
| CN111967365A (en) | Method and device for extracting image connection points | |
| Himabindu et al. | Detection and Removal of RainDrop from Images Using DeepLearning | |
| KR20200106111A (en) | Face landmark detection apparatus and method using gaussian landmark map with regression scheme |
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 |