CN107392951A - Remote sensing images high accuracy rapid registering method - Google Patents
Remote sensing images high accuracy rapid registering method Download PDFInfo
- Publication number
- CN107392951A CN107392951A CN201710420175.1A CN201710420175A CN107392951A CN 107392951 A CN107392951 A CN 107392951A CN 201710420175 A CN201710420175 A CN 201710420175A CN 107392951 A CN107392951 A CN 107392951A
- Authority
- CN
- China
- Prior art keywords
- registration
- image
- region
- satellite
- error
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/38—Registration of image sequences
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of remote sensing images high accuracy rapid registering method, it comprises the following steps：It is connected the nominal grid that establishment of coordinate system is ideally imaged over the ground in the earth, according to the satellite orbit of actual measurement, posture, camera angle data, the every frame scan image passed down is projected to nominal grid according to time series according to coordinate system transformational relation by pixel；In projection process, the visual field of the cold space segment beyond earth background region can not map to nominal grid, and earth background region and cold area of space are distinguished by earth edge, and the registration calculating based on projected image is carried out to earth background area；The second frame original image is received according to sequential, when receiving first piece of region subject to registration, according to camera angle position, attitude of satellite etc. corresponding to the region；Present invention demonstrates that in-orbit image registration accuracy can be effectively improved, meet index request, accuracy rate can ensure more than 5%.
Description
Technical field
The present invention relates to satellite image to handle application field, more particularly to a kind of remote sensing images high accuracy rapid registering side
Method.
Background technology
The detector scale of the satellite carried camera of remote sensing of the earth increasingly increases with technology development, and camera is in-orbit to be imaged number over the ground
According to the corresponding increase of amount, the real time handling requirement on star and ground is also improved therewith., can by the registration to continuous multiple frames image
Identify the dynamic and static target in image.Under the conditions of certain registration accuracy, the motion rail of moving-target in image can be further estimated
Mark, and the movement locus of moving-target and time are closely related.Therefore, the high-precision rapid registering of remote sensing images and its effectively reality
Apply, it is most important for giving full play to for the in-orbit effectiveness of satellite.
The content of the invention
The technical problems to be solved by the invention are to provide a kind of remote sensing images high accuracy rapid registering method, which demonstrate
In-orbit image registration accuracy can be effectively improved, meets index request.
The present invention is that solve above-mentioned technical problem by following technical proposals：A kind of remote sensing images are quickly matched somebody with somebody in high precision
Quasi- method, it comprises the following steps：
Be connected the nominal grid that establishment of coordinate system is ideally imaged over the ground in the earth, according to the satellite orbit of actual measurement,
Posture, camera angle data, the every frame scan image passed down is thrown according to time series according to coordinate system transformational relation by pixel
Shadow is to nominal grid；
In projection process, the visual field of the cold space segment beyond earth background region can not map to nominal grid, lead to
Cross earth edge and distinguish earth background region and cold area of space, the registration meter based on projected image is carried out to earth background area
Calculate；
The second frame original image is received according to sequential, when receiving first piece of region subject to registration, according to corresponding to the region
Camera angle position, the attitude of satellite；Track, thermal change graphic data calculate the region in the position of nominal grid, then by area subject to registration
Domain is projected to nominal grid, the Region Matching subject to registration with relevant position in first two field picture；
The index of its in-orbit image registration accuracy is calculated using the remote sensing images high accuracy rapid registering model emulation of foundation
Satisfaction, by contrasting Index Theory analyze data；Image registration accuracy is between the shifting of picture centre picture in registration accuracy and region
Pixel picture shifting amount sum；Influenceing the error term of image registration mainly includes camera angle measurement error, attitude of satellite error, track
Error, time synchronization error, imaging model calibration error factor.
Preferably, the camera angle measurement error includes angle measurement elemental error and processing noise in electronic circuits.
Preferably, the registration based on projected image, which calculates, comprises the following specific steps that：
According to blit under satellite as packet sequence according to ranks number is spliced into original image；
Image is carried out to wait area dividing；
Attitude of satellite data are surveyed according to first frame image data first trip, establish the registering frame of reference；
According to camera angle position and the attitude of satellite is corresponded in first two field picture, each regional center point coordinates is calculated；
Follow-up every two field picture is carried out waiting area dividing；
Scanning mirror angle position and each regional center point coordinates of attitude of satellite calculating are corresponded to according in follow-up every two field picture；
According to regional center point coordinates position, by area image overall transformation to the registering frame of reference, in registering benchmark
Under coordinate system, according to first two field picture with it is follow-up per two field picture in relevant range position coordinates, calculate the area size of lap,
And the region is associated, the region formed after registration.
The positive effect of the present invention is：The present invention can effectively improve in-orbit image registration accuracy, meet index
It is required that accuracy rate can ensure the present invention more than 5%.Effectively quantificational description image slices move change, are connected coordinate in the earth
Conventional coordinates is established under system as benchmark, by carrying out the positioning based on nominal coordinate system to every two field picture, realizes that the earth is carried on the back
The multiframe registration of scenic spot area image.
Brief description of the drawings
Fig. 1 is the image-region rapid registering flow chart of the present invention.
Embodiment
Present pre-ferred embodiments are provided below in conjunction with the accompanying drawings, to describe technical scheme in detail.
Remote sensing images high accuracy rapid registering method of the present invention comprises the following steps：
Step 1: it is connected the nominal grid that establishment of coordinate system is ideally imaged over the ground in the earth, according to defending for actual measurement
The data such as star orbital road, posture, camera angle, by the every frame scan image passed down according to time series according to coordinate system transformational relation
Projected by pixel to nominal grid.
Step 2: in projection process, the visual field of the cold space segment beyond earth background region can not map to nominally
Grid, earth background region and cold area of space can be distinguished by earth edge, earth background area is carried out to be based on perspective view
The registration calculating of picture.
Step 3: the second frame original image is received according to sequential, when receiving first piece of region subject to registration, according to the region
Corresponding camera angle position, the attitude of satellite.Track, thermal change graphic data calculate the region in the position of nominal grid, then will treat
Registration region projection (translation, rotation, scaling) Region Matching subject to registration of relevant position into nominal grid, with first two field picture.
Step 4: calculate its in-orbit image registration accuracy using the remote sensing images high accuracy rapid registering model emulation of foundation
Index satisfaction, by contrasting Index Theory analyze data；Adjacent two field picture, which is influenceed, in satellite orbit motion scales smaller situation
Under, only considering the translation and rotation between image, different frame inter-image transformations model can be described as V=RU+T, V=in formula (vx,
Vy) t is the point in image f1 (x, y), and U=(ux, uy) t is the point in image f2 (x, y), R=[cos θ, sin θ；-sinθ,
Cos θ] it is spin matrix, θ is the two images anglec of rotation, and T=(tx, ty) t is the translational movement on y directions along x between image.Figure
As registration accuracy is equal to registration accuracy between the shifting of picture centre picture and pixel picture shifting amount sum in region.Influence the error of image registration
Item mainly includes camera angle measurement error, attitude of satellite error, orbit error, time synchronization error, imaging model calibration mistake
The factors such as difference.
The relative change in earth background region is smaller in the multiple image being imaged over the ground for satellite, and adjacent two field picture corresponds to picture
The relative picture of member moves less observation condition, considers image slices shifting and ground in satellite frame and calculates disposal ability, proposition one
Kind remote sensing images high accuracy rapid registering method.
As shown in figure 1, the registration based on projected image is calculated and comprised the following specific steps that：
(1) according to blit under satellite as packet sequence according to ranks number is spliced into original image；
(2) image is carried out waiting area dividing (N*N)；
(3) attitude of satellite data are surveyed according to first frame image data first trip, establishes the registering frame of reference (X0, Y0)；
(4) according to camera angle (scanning mirror corner) position and the attitude of satellite is corresponded in first two field picture, calculate in each region
Heart point coordinates (Xi, Yi)；
(5) follow-up every two field picture is carried out waiting area dividing (N*N)；
(6) scanning mirror angle position and each regional center point coordinates of attitude of satellite calculating are corresponded to according in subsequently every two field picture
(Xj, Yj)；
(7) according to regional center point coordinates position, by area image overall transformation to the registering frame of reference, in registering base
It is big according to first two field picture and relevant range position coordinates, the region for calculating lap in follow-up every two field picture under conventional coordinates
It is small, and the region is associated, the region formed after registration.
The rapid registering of remote sensing images can be realized according to the method described above, and image registration accuracy is equal between picture centre picture moves and matched somebody with somebody
Quasi- precision and pixel picture shifting amount sum in region.Because region registration combined measurement mode need to obtain each pixel in real time by star
It is connected position corresponding to coordinate system in the earth, the measurement error for having an impact the position is the error source of image registration.Image
Heat distortion amount can be approximately the same in correlation time, therefore the error term for influenceing image registration mainly includes camera angle measurement and missed
The factor such as difference, attitude of satellite error, orbit error, time synchronization error, imaging model calibration error.
Camera angle measurement error includes angle measurement elemental error and processing noise in electronic circuits etc., it is necessary in laboratory condition
Under demarcated.
By the way of star sensor and Gyro determine the attitude of satellite, its error term includes star sensor installation matrix
Deviation, measurement constant value deviation, low frequency aberration, noise equivalent angle；The installation matrix deviation of Gyro, constant value drift, random drift
Shifting, random walk and quantizing noise etc..High-frequency noise can be eliminated by filtering algorithm, obtain the real-time attitude of satellite.
High-precision navigation receiving equipment is configured on orbital position and star using ground multistation combined measurement satellite, can be improved
Satellite orbital position determines precision.
On star time reference and star time synchronization error influence the precision of image registration simultaneously.The star of remote sensing satellite at present
Mechanism disclosure satisfy that the time synchronized requirement on satellite and ground when ground time service and school.
There is distortion in the camera being imaged using long alignment or large area array, optical imagery, need to be in laboratory conditions to camera
Imaging model carries out calibration.
It is 8000 yuan × 12000 rows to consider original remote sensing images size, and inter frame image is first two field picture with reference of reference, i.e.,
Coordinate system benchmark using the attitude of satellite corresponding to the first trip of first two field picture as registration, by first two field picture, according to 512nd × 512 area
Domain is divided into 16 (alignment direction) × 24 (scanning direction) block regions subject to registration.The second frame original image is received according to sequential, when
When receiving first piece of region (512 × 512 yuan) subject to registration, according to camera angle position, Satellite Attitude corresponding to the regional center point
State increment information calculates the region in the position with conventional coordinates, then region entirety subject to registration is changed to conventional coordinates, with
The Region Matching subject to registration of relevant position in first two field picture.
Certain satellite background beforehand research model remote sensing images high accuracy rapid registering model is established in the method for patent of the present invention,
The index satisfaction of its in-orbit image registration accuracy of simulation calculation, background beforehand research model is helped smoothly successfully to set up the project.According to index
Theory analysis data, rapid registering index and the simulation result contrast simulation of continuous multiple frames image calculate data, such as table 1 below institute
Show, data show that this method calculates the error of remote sensing images rapid registering no more than 5% in table, demonstrate the accurate of this method
Property.
Table 1
Passing through the present invention, effectively quantificational description image slices move change, it was demonstrated that can effectively improve in-orbit image registration essence
Degree, meets index request, accuracy rate can ensure more than 5%.
Particular embodiments described above, technical problem, technical scheme and the beneficial effect of the solution to the present invention are carried out
It is further described, should be understood that the specific embodiment that the foregoing is only of the invention, be not limited to
The present invention, within the spirit and principles of the invention, any modification, equivalent substitution and improvements done etc., it should be included in this
Within the protection domain of invention.
Claims (3)
1. a kind of remote sensing images high accuracy rapid registering method, it is characterised in that it comprises the following steps：
Be connected the nominal grid that establishment of coordinate system is ideally imaged over the ground in the earth, according to the satellite orbit of actual measurement, posture,
Camera angle data, the every frame scan image passed down is projected to mark according to time series according to coordinate system transformational relation by pixel
Claim grid；
In projection process, the visual field of the cold space segment beyond earth background region can not map to nominal grid, pass through ground
Earth background region and cold area of space are distinguished in ball edge, and the registration calculating based on projected image is carried out to earth background area；
The second frame original image is received according to sequential, when receiving first piece of region subject to registration, according to camera corresponding to the region
Angle position, the attitude of satellite；Track, thermal change graphic data calculate the region and thrown in the position of nominal grid, then by region subject to registration
The Region Matching subject to registration of shadow relevant position into nominal grid, with first two field picture；
The index that its in-orbit image registration accuracy is calculated using the remote sensing images high accuracy rapid registering model emulation of foundation is met
Degree, by contrasting Index Theory analyze data；Image registration accuracy registration accuracy and pixel in region between the shifting of picture centre picture
As shifting amount sum；Influenceing the error term of image registration mainly includes camera angle measurement error, attitude of satellite error, track mistake
Difference, time synchronization error, imaging model calibration error factor.
2. remote sensing images high accuracy rapid registering method as claimed in claim 1, it is characterised in that the camera angle measurement
Error includes angle measurement elemental error and processing noise in electronic circuits.
3. remote sensing images high accuracy rapid registering method as claimed in claim 1, it is characterised in that described to be based on projected image
Registration calculate comprise the following specific steps that：
According to blit under satellite as packet sequence according to ranks number is spliced into original image；
Image is carried out to wait area dividing；
Attitude of satellite data are surveyed according to first frame image data first trip, establish the registering frame of reference；
According to camera angle position and the attitude of satellite is corresponded in first two field picture, each regional center point coordinates is calculated；
Follow-up every two field picture is carried out waiting area dividing；
Scanning mirror angle position and each regional center point coordinates of attitude of satellite calculating are corresponded to according in follow-up every two field picture；
According to regional center point coordinates position, by area image overall transformation to the registering frame of reference, in registering reference coordinate
Under system, according to first two field picture and relevant range position coordinates in follow-up often two field picture, the area size of lap is calculated, and incite somebody to action
The region is associated, the region formed after registration.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710420175.1A CN107392951A (en) | 2017-06-06 | 2017-06-06 | Remote sensing images high accuracy rapid registering method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710420175.1A CN107392951A (en) | 2017-06-06 | 2017-06-06 | Remote sensing images high accuracy rapid registering method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107392951A true CN107392951A (en) | 2017-11-24 |
Family
ID=60333206
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710420175.1A Pending CN107392951A (en) | 2017-06-06 | 2017-06-06 | Remote sensing images high accuracy rapid registering method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107392951A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108986025A (en) * | 2018-06-08 | 2018-12-11 | 西北工业大学 | Based on the high-precision different time image mosaic of incomplete appearance rail information and bearing calibration |
| CN111161186A (en) * | 2019-12-30 | 2020-05-15 | 国家卫星气象中心(国家空间天气监测预警中心) | Push-broom remote sensor channel registration method and device |
| CN111435969A (en) * | 2019-01-11 | 2020-07-21 | 佳能株式会社 | Image processing apparatus, control method thereof, recording medium, and information processing system |
| CN112017217A (en) * | 2020-09-02 | 2020-12-01 | 国家卫星气象中心(国家空间天气监测预警中心) | Registration method and device between remote sensing data channels of multi-frequency terahertz detector |
| CN112199801A (en) * | 2020-04-13 | 2021-01-08 | 北京电子工程总体研究所 | Scanning area determination method and device based on satellite scanning simulation |
| CN113096163A (en) * | 2021-04-16 | 2021-07-09 | 湘潭大学 | Satellite-borne SAR image high-precision matching method without prior lifting rail information |
| CN113487676A (en) * | 2018-03-19 | 2021-10-08 | 百度在线网络技术（北京）有限公司 | Method and apparatus for determining relative pose angle between cameras mounted to an acquisition entity |
| CN114663480A (en) * | 2022-02-10 | 2022-06-24 | 上海卫星工程研究所 | Synchronous image rotation elimination and channel registration method and system for 45-degree rotary scanning space camera |
| CN117011179A (en) * | 2023-08-09 | 2023-11-07 | 北京精英路通科技有限公司 | Image conversion method and device, electronic equipment and storage medium |
| CN117745777A (en) * | 2024-01-05 | 2024-03-22 | 哈尔滨工业大学 | Remote sensing detection earth surface dense abnormal element removing method based on space-time registration |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7343051B1 (en) * | 2005-03-07 | 2008-03-11 | Hsu Shin-Yi | Method of recognizing an object in an image using multi-sensor integration through conditionally optimal geoscene generation and registration |
| CN104111456A (en) * | 2014-07-23 | 2014-10-22 | 中国国土资源航空物探遥感中心 | Ground surface deformation high-resolution interferometric synthetic aperture radar (InSAR) monitoring method along high speed railway |
| CN106709944A (en) * | 2016-12-14 | 2017-05-24 | 上海微小卫星工程中心 | Satellite remote sensing image registration method |
-
2017
- 2017-06-06 CN CN201710420175.1A patent/CN107392951A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7343051B1 (en) * | 2005-03-07 | 2008-03-11 | Hsu Shin-Yi | Method of recognizing an object in an image using multi-sensor integration through conditionally optimal geoscene generation and registration |
| CN104111456A (en) * | 2014-07-23 | 2014-10-22 | 中国国土资源航空物探遥感中心 | Ground surface deformation high-resolution interferometric synthetic aperture radar (InSAR) monitoring method along high speed railway |
| CN106709944A (en) * | 2016-12-14 | 2017-05-24 | 上海微小卫星工程中心 | Satellite remote sensing image registration method |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113487676A (en) * | 2018-03-19 | 2021-10-08 | 百度在线网络技术（北京）有限公司 | Method and apparatus for determining relative pose angle between cameras mounted to an acquisition entity |
| CN108986025A (en) * | 2018-06-08 | 2018-12-11 | 西北工业大学 | Based on the high-precision different time image mosaic of incomplete appearance rail information and bearing calibration |
| CN111435969B (en) * | 2019-01-11 | 2021-11-09 | 佳能株式会社 | Image processing apparatus, control method thereof, recording medium, and information processing system |
| CN111435969A (en) * | 2019-01-11 | 2020-07-21 | 佳能株式会社 | Image processing apparatus, control method thereof, recording medium, and information processing system |
| US11074670B2 (en) | 2019-01-11 | 2021-07-27 | Canon Kabushiki Kaisha | Image processing apparatus, control method thereof, recording medium, and information processing system |
| CN111161186A (en) * | 2019-12-30 | 2020-05-15 | 国家卫星气象中心(国家空间天气监测预警中心) | Push-broom remote sensor channel registration method and device |
| CN111161186B (en) * | 2019-12-30 | 2023-10-03 | 国家卫星气象中心(国家空间天气监测预警中心) | Push-broom type remote sensor channel registration method and device |
| CN112199801A (en) * | 2020-04-13 | 2021-01-08 | 北京电子工程总体研究所 | Scanning area determination method and device based on satellite scanning simulation |
| CN112017217A (en) * | 2020-09-02 | 2020-12-01 | 国家卫星气象中心(国家空间天气监测预警中心) | Registration method and device between remote sensing data channels of multi-frequency terahertz detector |
| CN112017217B (en) * | 2020-09-02 | 2022-05-03 | 国家卫星气象中心(国家空间天气监测预警中心) | Registration method and device between remote sensing data channels of multi-frequency terahertz detector |
| CN113096163A (en) * | 2021-04-16 | 2021-07-09 | 湘潭大学 | Satellite-borne SAR image high-precision matching method without prior lifting rail information |
| CN114663480A (en) * | 2022-02-10 | 2022-06-24 | 上海卫星工程研究所 | Synchronous image rotation elimination and channel registration method and system for 45-degree rotary scanning space camera |
| CN117011179A (en) * | 2023-08-09 | 2023-11-07 | 北京精英路通科技有限公司 | Image conversion method and device, electronic equipment and storage medium |
| CN117745777A (en) * | 2024-01-05 | 2024-03-22 | 哈尔滨工业大学 | Remote sensing detection earth surface dense abnormal element removing method based on space-time registration |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107392951A (en) | Remote sensing images high accuracy rapid registering method | |
| US20200279434A1 (en) | System and method for determining geo-location(s) in images | |
| US7751651B2 (en) | Processing architecture for automatic image registration | |
| CN103822615B (en) | A kind of multi-control point extracts and the unmanned aerial vehicle target real-time location method be polymerized automatically | |
| CN102607526B (en) | Target posture measuring method based on binocular vision under double mediums | |
| CN101002069B (en) | Method of preparing a composite image with non-uniform resolution | |
| US20060215935A1 (en) | System and architecture for automatic image registration | |
| CN106709944A (en) | Satellite remote sensing image registration method | |
| US20080050042A1 (en) | Hardware-in-the-loop simulation system and method for computer vision | |
| CN112629431B (en) | Civil structure deformation monitoring method and related equipment | |
| CN107144293A (en) | A kind of geometric calibration method of video satellite area array cameras | |
| CN110503687B (en) | Target positioning method for aerial photoelectric measurement platform | |
| EP2212858A1 (en) | Method and apparatus of taking aerial surveys | |
| CN110146030A (en) | Side slope surface DEFORMATION MONITORING SYSTEM and method based on gridiron pattern notation | |
| CN105758400B (en) | Fixed statellite imaging based navigation be registrated fixed star sensitivity thing parameter extracting method | |
| CN106971408A (en) | A kind of camera marking method based on space-time conversion thought | |
| Rüther et al. | A comparison of close-range photogrammetry to terrestrial laser scanning for heritage documentation | |
| CN105444778B (en) | A kind of star sensor based on imaging geometry inverting is in-orbit to determine appearance error acquisition methods | |
| CN106525001A (en) | Method for calculating spatial pointing of optic axis of camera of remote sensing satellite in geostationary orbit | |
| Liu et al. | Accurate mapping method for UAV photogrammetry without ground control points in the map projection frame | |
| US7768631B1 (en) | Method and system for providing a known reference point for an airborne imaging platform | |
| Zhou et al. | Automatic orthorectification and mosaicking of oblique images from a zoom lens aerial camera | |
| Muffert et al. | The estimation of spatial positions by using an omnidirectional camera system | |
| CN107329191B (en) | Imaging test system and method for image navigation and registration of static meteorological satellite | |
| CN105403886A (en) | Automatic extraction method for airborne SAR scaler image position |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171124 |