CN112465714B - Multi-time-phase remote sensing image processing method, device and equipment - Google Patents
Multi-time-phase remote sensing image processing method, device and equipment Download PDFInfo
- Publication number
- CN112465714B CN112465714B CN202011330787.XA CN202011330787A CN112465714B CN 112465714 B CN112465714 B CN 112465714B CN 202011330787 A CN202011330787 A CN 202011330787A CN 112465714 B CN112465714 B CN 112465714B
- Authority
- CN
- China
- Prior art keywords
- image
- grid
- detected
- remote sensing
- reference image
- 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
- 238000003672 processing method Methods 0.000 title claims abstract description 14
- 230000008859 change Effects 0.000 claims abstract description 41
- 238000004364 calculation method Methods 0.000 claims abstract description 20
- 238000012952 Resampling Methods 0.000 claims abstract description 10
- 238000000034 method Methods 0.000 claims description 23
- 238000012545 processing Methods 0.000 claims description 18
- 238000010276 construction Methods 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 4
- 238000012937 correction Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 10
- 238000005516 engineering process Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/80—Geometric correction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
Abstract
The application relates to a multi-temporal remote sensing image processing method, which comprises the following steps: selecting remote sensing image data of one time phase from the remote sensing image data of two different time phases as a reference image, and using the remote sensing image data of the other time phase as an image to be detected; determining an overlapping area of the reference image and the image to be detected, calculating an external rectangle of the overlapping area, and dividing the external rectangle step by taking the external rectangle as the uppermost grid to construct a multi-stage grid; resampling the reference image and the image to be detected based on the multi-stage grid to obtain a regular reference image and the image to be detected, and respectively carrying out pixel-by-pixel feature description operator calculation on the regular reference image and the image to be detected to obtain feature description operators of all pixels; and determining the confidence coefficient of each layer of grids in the multi-level grids according to the obtained feature description operator, and obtaining the remote sensing image change probability according to the confidence coefficient. The accuracy of the remote sensing image change detection result can be effectively improved.
Description
Technical Field
The present application relates to the field of remote sensing image processing technologies, and in particular, to a method, an apparatus, and a device for processing a multi-temporal remote sensing image.
Background
The multi-temporal remote sensing image is an important data source in the monitoring of the change of the ground object. In the process of processing and analyzing multi-temporal remote sensing data, geometric registration among remote sensing images in different periods is generally carried out firstly, and then change detection among the remote sensing images is carried out. Thereby the change condition of the ground object is found. However, due to the influence of factors such as topography, satellite observation angles and the like, a certain geometric residual error still exists in the existing remote sensing image registration result. Meanwhile, in the related art, when the remote sensing image is processed, the image registration and the change detection of the remote sensing image are generally carried out separately in two stages, and the factors of the change of the ground object are not considered in the geometric registration, so that the geometric error of registration is increased, and the accuracy of the change detection result is lower.
Disclosure of Invention
In view of this, the application provides a multi-temporal remote sensing image processing method, which can effectively improve the accuracy of the remote sensing image change detection result.
According to an aspect of the present application, there is provided a multi-temporal remote sensing image processing method, including:
Selecting remote sensing image data of one time phase from the remote sensing image data of two different time phases as a reference image, and using the remote sensing image data of the other time phase as an image to be detected;
Determining an overlapping area of the reference image and the image to be detected, calculating an external rectangle of the overlapping area, and using the external rectangle as an uppermost grid, and dividing the external rectangle step by step to construct a multi-stage grid;
Resampling the reference image and the image to be detected based on the multi-stage grid to obtain a normalized reference image and an image to be detected, and respectively carrying out pixel-by-pixel feature description operator calculation on the normalized reference image and the image to be detected to obtain feature description operators of all pixels in the reference image and the image to be detected;
and determining the confidence coefficient of each layer of grids in the multi-stage grids according to the obtained feature description operator, and obtaining the remote sensing image change probability according to the confidence coefficient.
In one possible implementation manner, before determining the overlapping area of the reference image and the image to be detected, the method further includes:
And performing geometric correction on the image to be detected by adopting an image registration method and taking the reference image as a benchmark.
In one possible implementation manner, the external rectangle is taken as the uppermost grid, the external rectangle is divided step by step, and a multi-stage grid is constructed, including:
Dividing the circumscribed rectangle into two sub-rectangles by taking a perpendicular bisector of a longer side of the circumscribed rectangle as a dividing line to obtain a next-stage grid positioned on the uppermost-layer grid;
Dividing each sub-rectangle step by taking a perpendicular bisector of the longer side of each sub-rectangle as a dividing line until the longer side of the rectangle obtained by dividing is smaller than the pixel size of the reference image;
And the circumscribed rectangle is used as the uppermost grid, and each division result is used as a level and is arranged from top to bottom according to the generation sequence, so that the corresponding multi-stage grid is generated.
In one possible implementation, the resampling of the reference image and the image to be detected based on the multi-stage grid is performed by using a bilinear interpolation method.
In one possible implementation manner, the feature description operator is respectively performed on the normalized reference image and the image to be detected pixel by pixel, and when the feature description operators of the pixels in the reference image and the image to be detected are obtained, the formula is as follows:
Calculating to obtain;
Wherein i is an integer from 0 to 3, j is an integer from 0 to 31, σ is a gaussian distribution probability density function with variance of 1, fr (i, j) is a eigenvector of pixel P (x 0, y 0).
In one possible implementation manner, determining the confidence level of each layer of the multi-level mesh according to the obtained feature description operator includes:
determining the confidence degree of the uppermost grid as a first preset value, and taking the uppermost grid as a current grid;
searching the next layer of the current grid step by step according to the corresponding relation in the multi-stage grids, and calculating the confidence coefficient of the next layer of the grid according to the confidence coefficient of the current grid until the lowest layer of the grid is searched.
In one possible implementation manner, the value of the first preset value is 1;
when the confidence of the next layer of grid is calculated according to the confidence of the current grid, the following formula is adopted:
Calculating to obtain;
Wherein, Representing all resampled pixel positions corresponding to the lower layer, fr r represents the feature operator of the reference image, fr t represents the feature operator of the image to be registered, and k p represents the confidence of the current grid.
In one possible implementation manner, obtaining the remote sensing image change probability according to the confidence coefficient includes:
And carrying out logic calculation on the confidence coefficient of each pixel, and determining the remote sensing image change probability according to the calculation result.
According to another aspect of the application, there is further provided a multi-temporal remote sensing image processing device, including an image selection module, a grid construction module, a feature description operator calculation module and a change probability determination module;
The image selecting module is configured to select remote sensing image data of one time phase from remote sensing image data of two different time phases as a reference image, and remote sensing image data of the other time phase is used as an image to be detected;
the grid construction module is configured to determine an overlapping area of the reference image and the image to be detected, calculate an external rectangle of the overlapping area, and take the external rectangle as an uppermost grid, divide the external rectangle step by step, and construct a multi-stage grid;
the feature description operator calculation module is configured to resample the reference image and the image to be detected based on the multi-stage grid to obtain a normalized reference image and an image to be detected, and respectively calculate feature description operators of the normalized reference image and the image to be detected pixel by pixel to obtain feature description operators of each pixel in the reference image and the image to be detected;
The change probability determining module is configured to determine the confidence coefficient of each layer of grids in the multi-level grids according to the obtained feature description operator, and obtain the change probability of the remote sensing image according to the confidence coefficient.
According to another aspect of the present application, there is also provided a multi-temporal remote sensing image processing apparatus including:
A processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to implement any of the methods described above when executing the executable instructions.
According to the multi-temporal remote sensing image processing method, two remote sensing images with different time phases are respectively used as a reference image and an image to be detected, overlapping areas of the reference image and the image to be detected are divided step by step, corresponding multi-stage grids are constructed, and then feature description operators of pixels in the reference image and the image to be detected are respectively calculated based on the constructed multi-stage grids after resampling of the reference image and the image to be detected, after confidence degrees of the grids at each stage are determined according to the feature description operators of the pixels, finally remote sensing image change probability is obtained according to the determined confidence degrees of the grids at each stage, the purposes of integrating local registration and change detection are achieved through repeated iterative operation on the basis of calculating the feature description operators of the images, and accordingly accuracy of remote sensing image change detection results is effectively improved.
Other features and aspects of the present application will become apparent from the following detailed description of exemplary embodiments, which proceeds with reference to the accompanying drawings.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate exemplary embodiments, features and aspects of the application and together with the description, serve to explain the principles of the application.
FIG. 1 is a flowchart of a multi-temporal remote sensing image processing method according to an embodiment of the present application;
FIG. 2 is a block diagram showing a multi-temporal remote sensing image processing apparatus according to an embodiment of the present application;
Fig. 3 is a block diagram illustrating a configuration of a multi-temporal remote sensing image processing apparatus according to an embodiment of the present application.
Detailed Description
Various exemplary embodiments, features and aspects of the application will be described in detail below with reference to the drawings. In the drawings, like reference numbers indicate identical or functionally similar elements. Although various aspects of the embodiments are illustrated in the accompanying drawings, the drawings are not necessarily drawn to scale unless specifically indicated.
The word "exemplary" is used herein to mean "serving as an example, embodiment, or illustration. Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments.
In addition, numerous specific details are set forth in the following description in order to provide a better illustration of the application. It will be understood by those skilled in the art that the present application may be practiced without some of these specific details. In some instances, well known methods, procedures, components, and circuits have not been described in detail so as not to obscure the present application.
Fig. 1 is a flowchart of a multi-temporal remote sensing image processing method according to an embodiment of the application. As shown in fig. 1, the method includes: in step S100, the remote sensing image data of one time phase is selected from the remote sensing image data of two different time phases as the reference image, and the remote sensing image data of the other time phase is used as the image to be detected. Here, it should be noted that the remote sensing image data of two different phases refer to two acquired remote sensing images of different phases of the same region. Meanwhile, it should be noted that the selection of the reference image and the image to be detected can be flexibly selected according to the needs. That is, the front-time-phase remote sensing image of the same region can be used as a reference image, and the rear-time-phase remote sensing image can be used as an image to be detected. The front time phase remote sensing image of the same area can be selected as an image to be detected, and the rear time phase remote sensing image can be selected as a reference image.
Step S200, determining an overlapping area of the reference image and the image to be detected, calculating an external rectangle of the overlapping area, and dividing the external rectangle step by taking the external rectangle as the uppermost grid to construct a multi-stage grid. In this step, the calculation of the overlapping area of the reference image and the image to be detected may be implemented by conventional technical means in the art, which is not described herein. The circumscribed rectangle of the overlap region refers to the four to four extent of the overlap region, and includes all pixels of the overlap region. The constructed multi-stage mesh corresponds to image areas contained in rectangles of different stage divisions.
Step S300, resampling the reference image and the image to be detected based on the multi-stage grid to obtain a normalized reference image and an image to be detected, and calculating a feature description operator pixel by pixel for the normalized reference image and the image to be detected to obtain the feature description operators of each pixel in the reference image and the image to be detected.
Finally, through step S400, the confidence coefficient of each layer of grids in the multi-level grids is determined according to the obtained feature description operator, and the remote sensing image change probability is obtained according to the confidence coefficient.
According to the multi-temporal remote sensing image processing method, two remote sensing images with different time phases are respectively used as a reference image and an image to be detected, overlapping areas of the reference image and the image to be detected are divided step by step, corresponding multi-stage grids are constructed, then feature description operators of pixels in the reference image and the image to be detected are respectively calculated based on the constructed multi-stage grids after resampling is carried out on the reference image and the image to be detected, confidence degrees of the grids at all stages are determined according to the feature description operators of the pixels, finally the change probability of the remote sensing image is obtained according to the determined confidence degrees of the grids at all stages, the purposes that local registration and change detection integrated processing is completed on the basis of calculating the feature description operators of the image pixels are achieved through repeated iterative operation, and therefore accuracy of remote sensing image change detection results is effectively improved.
It should be noted that, in the method of the embodiment of the present application, after determining the reference image and the image to be detected, before calculating the overlapping area of the reference image and the image to be detected, a general image registration method is further included, and the reference image is used as a reference, so that the geometric error between the reference image and the image to be detected is reduced as much as possible, and a foundation is laid for ensuring the accuracy of the determined overlapping area.
Here, it should be noted that, based on the reference image, the image registration method used for performing geometric correction on the image to be detected may be a conventional technology in the art, and will not be described herein.
After determining the overlapping area between the reference image and the image to be detected, the circumscribed rectangle of the overlapping area can be calculated. The circumscribed rectangle may be implemented in a manner that calculates the four-to-range of the overlap region. Then, the split line is divided equally into two sub-rectangles by taking the perpendicular bisector of the longer side of the circumscribed rectangle as the split line. Then, the dividing operation is repeated for the two divided sub-rectangles, namely, the two sub-rectangles are divided respectively by taking the perpendicular bisectors of the longer sides of the two sub-rectangles as dividing lines, so that four rectangles are obtained. And the like, dividing each rectangle obtained by current division step by step in the mode until the lengths of two sides of the sub-rectangle after current division are smaller than the pixel size of the reference image (or the length of the longer side is smaller than the pixel size of the reference image), and then the multi-stage grid can be divided.
The method comprises the steps of taking the circumscribed rectangle of the overlapped area as the uppermost grid, taking the result of each division as a level and arranging the division from top to bottom according to the generation sequence, so as to construct the multi-level grid.
Here, it is understood by those skilled in the art that the multi-stage division of the overlapping area between the reference image and the image to be detected is achieved by constructing a multi-stage mesh, and each layer and the mesh contains a part of the image in the overlapping area.
And then resampling the reference image and the image to be detected to obtain the regular reference image and the regular image to be detected. When resampling the reference image and the image to be detected, a bilinear interpolation method can be adopted to resample the image according to the lowest layer of the grid. The reference image and the image to be detected are respectively resampled according to the lowest layer of grids (namely, the lowest layer of grids) in the multi-level grids by adopting a bilinear interpolation method based on the constructed multi-level grids, so that the characteristics of the regular reference image and the characteristics of the image to be detected are more obvious.
Meanwhile, the feature description operator calculation is respectively carried out on the regular reference image and the image to be detected pixel by pixel, and when the feature description operators of the pixels in the reference image and the image to be detected are obtained, the feature description operators can be obtained through calculation according to the following formula:
Wherein i is an integer from 0 to 3, j is an integer from 0 to 31, and sigma is a gaussian distribution probability density function with variance of 1. Fr (i, j) is used as a 128-dimensional feature array, namely a feature operator of the pixel P (x 0,y0), and the feature operator is obtained by adopting a SIFT operator calculation method.
Further, determining the confidence level of each layer of grids in the multi-level grid according to the obtained feature description operator can be achieved in the following manner.
That is, the topmost grid includes only one grid, the confidence of the grid is determined to be a first preset value, and the topmost grid is taken as the current grid. And then searching the next layer of grids of the current grid step by step according to the corresponding relation in the multi-stage grids, and calculating the confidence coefficient of the next layer of grids according to the confidence coefficient of the current grid until the lowest layer of grids is searched. Here, it should be noted that the value of the first preset value may be set to 1.
That is, in determining the confidence level of each layer of grids in the multi-level grid from the obtained feature description operator, the determination of the confidence level of all the grids in each layer of grids may be performed layer by layer (step by step) starting with the uppermost layer of the constructed multi-level grid.
Meanwhile, when the confidence of each layer of grids is carried out, the confidence of the uppermost layer of grids is set to be 1, and the uppermost layer of grids is used as the current grids. Then, according to the corresponding relation of the multi-stage grids, searching the lower-stage grids corresponding to the current grids, and respectively calculating the confidence coefficient of the lower-stage grids according to the following formula:
Wherein, Representing all resampled pixel positions corresponding to the lower layer lattice, fr r represents a feature operator of the reference image, fr t represents a feature operator of the image to be registered, and k p represents the confidence of the current layer lattice.
After the confidence coefficient of the next layer of the current grid is calculated, the steps are continuously executed, the currently calculated grid is used as the current grid, and the calculation of the confidence coefficient of the next layer of the grid is continuously executed until the confidence coefficient of the bottommost grid is calculated.
After the confidence coefficient of the bottom layer lattice is calculated, the confidence coefficient of all pixels in the overlapping area of the reference image and the image to be detected is obtained by the one-to-one correspondence relationship between the bottom layer lattice and the pixels.
After the confidence coefficient of each pixel is obtained, the corresponding grid change probability is calculated according to the following formula.
Pc＝∑i∈gridki
Where P c is the grid change probability, grid is all pixels in the grid, i is the i-th pixel in the grid, and k i is the confidence of the pixel.
The lattice change probability is calculated starting from the lowest lattice. Firstly, calculating the grid change probability of all grids in the bottom grid; when P c of a certain grid is more than or equal to 0.5, the grid is considered to have ground object change, and pixels belonging to the grid are not involved in calculation of the corresponding upper grid change probability; if P c is less than 0.5, the grid change probability of the previous stage grid is continuously calculated. Until the change probability of all the layers of grids is calculated and judged, the grids with the change are identified as the change detection results.
Correspondingly, based on any of the multi-temporal remote sensing image processing methods, the application also provides a multi-temporal remote sensing image processing device. Because the working principle of the multi-temporal remote sensing image processing device provided by the application is the same as or similar to that of the multi-temporal remote sensing image processing method provided by the application, repeated parts are not repeated.
Referring to fig. 2, the multi-temporal remote sensing image processing apparatus 100 provided by the present application includes an image selecting module 110, a grid construction module 120, a feature description operator calculating module 130 and a change probability determining module 140. The image selecting module 110 is configured to select, from the two remote sensing image data of different time phases, the remote sensing image data of one time phase as a reference image, and the remote sensing image data of the other time phase as an image to be detected. The grid construction module 120 is configured to determine an overlapping area of the reference image and the image to be detected, calculate an external rectangle of the overlapping area, and divide the external rectangle step by step with the external rectangle as the uppermost grid to construct a multi-stage grid. The feature descriptor calculation module 130 is configured to resample the reference image and the image to be detected based on the multi-stage grid to obtain a normalized reference image and an image to be detected, and perform feature descriptor calculation on the normalized reference image and the image to be detected pixel by pixel to obtain feature descriptors of each pixel in the reference image and the image to be detected. The change probability determining module 140 is configured to determine a confidence coefficient of each layer of the multi-level grids according to the obtained feature description operator, and obtain a remote sensing image change probability according to the confidence coefficient.
Still further, according to another aspect of the present application, there is provided a multi-temporal remote sensing image processing apparatus 200. Referring to fig. 3, a multi-temporal remote sensing image processing apparatus 200 according to an embodiment of the present application includes a processor 210 and a memory 220 for storing instructions executable by the processor 210. The processor 210 is configured to implement any of the above multi-temporal remote sensing image processing methods when executing the executable instructions.
Here, it should be noted that the number of processors 210 may be one or more. Meanwhile, in the multi-temporal remote sensing image processing apparatus 200 according to the embodiment of the present application, an input device 230 and an output device 240 may be further included. The processor 210, the memory 220, the input device 230, and the output device 240 may be connected by a bus, or may be connected by other means, which is not specifically limited herein.
The memory 220 is a computer-readable storage medium that can be used to store software programs, computer-executable programs, and various modules, such as: the multi-temporal remote sensing image processing method of the embodiment of the application corresponds to a program or a module. The processor 210 executes various functional applications and data processing of the multi-temporal remote sensing image processing apparatus 200 by running software programs or modules stored in the memory 220.
The input device 230 may be used to receive an input digital or signal. Wherein the signal may be a key signal generated in connection with user settings of the device/terminal/server and function control. The output means 240 may comprise a display device such as a display screen.
The foregoing description of embodiments of the application has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the various embodiments described. The terminology used herein was chosen in order to best explain the principles of the embodiments, the practical application, or the technical improvement of the technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
Claims (8)
1. A multi-temporal remote sensing image processing method is characterized by comprising the following steps:
Selecting remote sensing image data of one time phase from the remote sensing image data of two different time phases as a reference image, and using the remote sensing image data of the other time phase as an image to be detected;
Determining an overlapping area of the reference image and the image to be detected, calculating an external rectangle of the overlapping area, and using the external rectangle as an uppermost grid, and dividing the external rectangle step by step to construct a multi-stage grid;
Resampling the reference image and the image to be detected based on the multi-stage grid to obtain a normalized reference image and an image to be detected, and respectively carrying out pixel-by-pixel feature description operator calculation on the normalized reference image and the image to be detected to obtain feature description operators of all pixels in the reference image and the image to be detected;
determining the confidence coefficient of each layer of grids in the multi-level grids according to the obtained feature description operator, and obtaining the remote sensing image change probability according to the confidence coefficient;
The determining the confidence coefficient of each layer of grids in the multi-level grids according to the obtained feature description operator comprises the following steps:
determining the confidence degree of the uppermost grid as a first preset value, and taking the uppermost grid as a current grid;
Searching a next layer of grids of the current grid step by step according to the corresponding relation in the multi-stage grids, and calculating the confidence coefficient of the next layer of grids according to the confidence coefficient of the current grid until the lowest layer of grids is searched;
Wherein the value of the first preset value is 1;
when the confidence of the next layer of grid is calculated according to the confidence of the current grid, the following formula is adopted:
Calculating to obtain;
Wherein, Representing all resampled pixel positions corresponding to the lower layer, fr r represents the feature operator of the reference image, fr t represents the feature operator of the image to be registered, and k p represents the confidence of the current grid.
2. The method of claim 1, further comprising, prior to determining the overlapping region of the reference image and the image to be detected:
And performing geometric correction on the image to be detected by adopting an image registration method and taking the reference image as a benchmark.
3. The method according to claim 1, wherein the step-by-step division is performed on the circumscribed rectangle with the circumscribed rectangle as an uppermost grid, so as to construct a multi-stage grid, comprising:
Dividing the circumscribed rectangle into two sub-rectangles by taking a perpendicular bisector of a longer side of the circumscribed rectangle as a dividing line to obtain a next-stage grid positioned on the uppermost-layer grid;
Dividing each sub-rectangle step by taking a perpendicular bisector of the longer side of each sub-rectangle as a dividing line until the longer side of the rectangle obtained by dividing is smaller than the pixel size of the reference image;
And the circumscribed rectangle is used as the uppermost grid, and each division result is used as a level and is arranged from top to bottom according to the generation sequence, so that the corresponding multi-stage grid is generated.
4. The method of claim 1, wherein resampling the reference image and the image to be detected based on the multi-stage grid is performed using bilinear interpolation.
5. The method of claim 1, wherein when performing a feature description operator calculation on the normalized reference image and the image to be detected pixel by pixel to obtain feature description operators of each pixel in the reference image and the image to be detected, the method is as follows:
Calculating to obtain;
wherein i is an integer from 0 to 3, j is an integer from 0 to 31, σ is a gaussian distribution probability density function with variance of 1, fr (i, j) is a eigenvector of pixel P (x 0, y 0); p (x, y) is the pixel in the normalized reference image and the image to be detected.
6. The method of claim 1, wherein deriving a probability of change of the remote sensing image based on the confidence level comprises:
And carrying out logic calculation on the confidence coefficient of each pixel, and determining the remote sensing image change probability according to the calculation result.
7. The multi-temporal remote sensing image processing device is characterized by comprising an image selecting module, a grid construction module, a characteristic description operator calculation module and a change probability determination module;
The image selecting module is configured to select remote sensing image data of one time phase from remote sensing image data of two different time phases as a reference image, and remote sensing image data of the other time phase is used as an image to be detected;
the grid construction module is configured to determine an overlapping area of the reference image and the image to be detected, calculate an external rectangle of the overlapping area, and take the external rectangle as an uppermost grid, divide the external rectangle step by step, and construct a multi-stage grid;
the feature description operator calculation module is configured to resample the reference image and the image to be detected based on the multi-stage grid to obtain a normalized reference image and an image to be detected, and respectively calculate feature description operators of the normalized reference image and the image to be detected pixel by pixel to obtain feature description operators of each pixel in the reference image and the image to be detected;
the change probability determining module is configured to determine the confidence coefficient of each layer of grids in the multi-level grids according to the obtained feature description operator, and obtain the change probability of the remote sensing image according to the confidence coefficient;
The determining the confidence coefficient of each layer of grids in the multi-level grids according to the obtained feature description operator comprises the following steps:
determining the confidence degree of the uppermost grid as a first preset value, and taking the uppermost grid as a current grid;
Searching a next layer of grids of the current grid step by step according to the corresponding relation in the multi-stage grids, and calculating the confidence coefficient of the next layer of grids according to the confidence coefficient of the current grid until the lowest layer of grids is searched;
Wherein the value of the first preset value is 1;
when the confidence of the next layer of grid is calculated according to the confidence of the current grid, the following formula is adopted:
Calculating to obtain;
Wherein, Representing all resampled pixel positions corresponding to the lower layer, fr r represents the feature operator of the reference image, fr t represents the feature operator of the image to be registered, and k p represents the confidence of the current grid.
8. A multi-temporal remote sensing image processing apparatus, comprising:
A processor;
a memory for storing processor-executable instructions;
Wherein the processor is configured to implement the method of any one of claims 1 to 6 when executing the executable instructions.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011330787.XA CN112465714B (en) | 2020-11-24 | 2020-11-24 | Multi-time-phase remote sensing image processing method, device and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011330787.XA CN112465714B (en) | 2020-11-24 | 2020-11-24 | Multi-time-phase remote sensing image processing method, device and equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112465714A CN112465714A (en) | 2021-03-09 |
| CN112465714B true CN112465714B (en) | 2024-05-07 |
Family
ID=74798765
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011330787.XA Active CN112465714B (en) | 2020-11-24 | 2020-11-24 | Multi-time-phase remote sensing image processing method, device and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112465714B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102073990A (en) * | 2010-11-26 | 2011-05-25 | 中国科学院遥感应用研究所 | System framework and method for automatic geometric correction of remote sensing images |
| CN103544711A (en) * | 2013-11-08 | 2014-01-29 | 国家测绘地理信息局卫星测绘应用中心 | Automatic registering method of remote-sensing image |
| CN105551028A (en) * | 2015-12-09 | 2016-05-04 | 中山大学 | Method and system for dynamically updating geographic space data based on remote sensing image |
| CN105913435A (en) * | 2016-04-13 | 2016-08-31 | 西安航天天绘数据技术有限公司 | Multidimensional remote sensing image matching method and multidirectional remote sensing image matching system suitable for large area |
| CN106447653A (en) * | 2016-09-09 | 2017-02-22 | 河海大学 | Method for detecting change of multi-temporal and multispectral image based on spatial constraint chi-squared transform |
| CN106780355A (en) * | 2016-11-17 | 2017-05-31 | 中国科学院遥感与数字地球研究所 | A kind of Enhancement Methods about Satellite Images |
| CN109598750A (en) * | 2018-12-07 | 2019-04-09 | 中国地质大学（武汉） | One kind being based on the pyramidal large scale differential image characteristic point matching method of deformation space |
| US10325370B1 (en) * | 2016-05-31 | 2019-06-18 | University Of New Brunswick | Method and system of coregistration of remote sensing images |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102012221667A1 (en) * | 2012-11-27 | 2014-05-28 | Deutsches Zentrum für Luft- und Raumfahrt e.V. | Apparatus and method for processing remote sensing data |
-
2020
- 2020-11-24 CN CN202011330787.XA patent/CN112465714B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102073990A (en) * | 2010-11-26 | 2011-05-25 | 中国科学院遥感应用研究所 | System framework and method for automatic geometric correction of remote sensing images |
| CN103544711A (en) * | 2013-11-08 | 2014-01-29 | 国家测绘地理信息局卫星测绘应用中心 | Automatic registering method of remote-sensing image |
| CN105551028A (en) * | 2015-12-09 | 2016-05-04 | 中山大学 | Method and system for dynamically updating geographic space data based on remote sensing image |
| CN105913435A (en) * | 2016-04-13 | 2016-08-31 | 西安航天天绘数据技术有限公司 | Multidimensional remote sensing image matching method and multidirectional remote sensing image matching system suitable for large area |
| US10325370B1 (en) * | 2016-05-31 | 2019-06-18 | University Of New Brunswick | Method and system of coregistration of remote sensing images |
| CN106447653A (en) * | 2016-09-09 | 2017-02-22 | 河海大学 | Method for detecting change of multi-temporal and multispectral image based on spatial constraint chi-squared transform |
| CN106780355A (en) * | 2016-11-17 | 2017-05-31 | 中国科学院遥感与数字地球研究所 | A kind of Enhancement Methods about Satellite Images |
| CN109598750A (en) * | 2018-12-07 | 2019-04-09 | 中国地质大学（武汉） | One kind being based on the pyramidal large scale differential image characteristic point matching method of deformation space |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112465714A (en) | 2021-03-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107301402B (en) | Method, device, medium and equipment for determining key frame of real scene | |
| CN111259846B (en) | Text positioning method and system and text positioning model training method and system | |
| CN111192239B (en) | Remote sensing image change area detection method and device, storage medium and electronic equipment | |
| CN109117854B (en) | Key point matching method and device, electronic equipment and storage medium | |
| CN111091572B (en) | Image processing method and device, electronic equipment and storage medium | |
| CN110807730B (en) | Image geometric correction method and device and electronic equipment | |
| CN111709876B (en) | Image splicing method, device, equipment and storage medium | |
| CN112465714B (en) | Multi-time-phase remote sensing image processing method, device and equipment | |
| CN116881515B (en) | Method and electronic equipment for comparing capacitance results solved by different algorithms | |
| CN112446951B (en) | Three-dimensional reconstruction method, three-dimensional reconstruction device, electronic equipment and computer storage medium | |
| CN111461196B (en) | Rapid robust image identification tracking method and device based on structural features | |
| CN111552751B (en) | Three-dimensional landmark control point generation and application method, generation and application device | |
| CN116486312B (en) | Video image processing method and device, electronic equipment and storage medium | |
| CN111275616B (en) | Low-altitude aerial image splicing method and device | |
| CN116977671A (en) | Target tracking method, device, equipment and storage medium based on image space positioning | |
| CN112508832A (en) | Object-oriented remote sensing image data space-time fusion method, system and equipment | |
| CN108765277B (en) | Image splicing method and device, computer equipment and storage medium | |
| CN109242894B (en) | Image alignment method and system based on mobile least square method | |
| CN114723792A (en) | Method and device for updating land parcel information, electronic equipment and storage medium | |
| CN109816613B (en) | Image completion method and device | |
| CN113344994A (en) | Image registration method and device, electronic equipment and storage medium | |
| KR102637913B1 (en) | Method and apparatus for composing images | |
| CN111754397B (en) | Remote sensing image embedding method and device, electronic equipment and storage medium | |
| CN117274310A (en) | Multi-target tracking method, device, equipment and computer readable storage medium | |
| WO2024016368A1 (en) | Mask-guided image mosaic line generation method and apparatus, computer device, and storage medium |
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 | ||
| CB02 | Change of applicant information |
Country or region after: China Address after: 100075 room 1107, Haitai building, 229 beisihuanzhong Road, Haidian District, Beijing Applicant after: Aerospace Science and Technology (Beijing) Space Information Application Co.,Ltd. Address before: 100075 room 1107, Haitai building, 229 beisihuanzhong Road, Haidian District, Beijing Applicant before: BEIJING AEROSPACE TITAN TECHNOLOGY Co.,Ltd. Country or region before: China |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |