Optical remote sensing camera image registration precision index compliance calculation system and method

Abstract

The present application relates to a kind of multispectral space optical remote sensing camera image registration precision index compliance degree calculation system and method, the device is according to multispectral space optical remote sensing camera design index, the projection difference of the same ground object imaging on each spectrum is calculated, and then the matching error between the remote sensing image obtained by each spectrum on orbit is calculated, compared with the required registration precision index, the compliance degree of registration precision index is analyzed.The method is for the design of multispectral space optical remote sensing camera, can quickly and accurately estimate the registration precision of the remote sensing image obtained by multispectral space optical remote sensing camera on orbit after, analyze the compliance degree with registration precision index, guide the design of multispectral space optical remote sensing camera.

Description

Technical Field

[0001] This invention relates to a device and method for calculating the conformity of image registration accuracy indicators of a multi-band aerospace optical remote sensing camera, belonging to the field of aerospace remote sensing manufacturing. Background Technology

[0002] Multispectral spaceborne optical remote sensing images have seen significant improvements in spectral and spatial resolution. Various applications, such as remote sensing image fusion and interpretation, place increasingly higher demands on the spatial registration accuracy of image data across different spectral bands. Therefore, multispectral spaceborne optical remote sensing camera image band registration has become a crucial step in the preprocessing of data from multispectral cameras on high-resolution optical remote sensing satellites.

[0003] During the design process, the multi-band aerospace optical remote sensing camera sequentially images the same ground object in different spectral bands within the focal plane field of view. This results in a certain base-to-height ratio between each pair of spectral bands. When the projection difference caused by the height of the observed object in different spectral bands exceeds the index requirements, the object cannot be spatially registered in images of different spectral bands.

[0004] To avoid design flaws in multi-band aerospace optical remote sensing cameras causing product performance deviations and posing challenges to the preprocessing of multispectral images from high-resolution remote sensing satellites, the registration accuracy between different spectral bands of the multi-band aerospace optical remote sensing camera should be analyzed and verified during the design phase. This is necessary to correct design defects and calculate the degree to which the image registration accuracy of the aerospace optical remote sensing camera meets the required standards. Summary of the Invention

[0005] The technical problem solved by this invention is to overcome the shortcomings of existing technologies and fill the gaps in current methods for analyzing the registration accuracy of multi-spectral space optical remote sensing cameras. Specifically, this invention provides a system and method for calculating the degree of conformity of registration accuracy indicators for multi-spectral space optical remote sensing cameras. This system can quickly and accurately estimate the registration accuracy of remote sensing images acquired by multi-spectral space optical remote sensing cameras after they are in orbit, analyze the degree of conformity with registration accuracy indicators, and guide the design of multi-spectral space optical remote sensing cameras.

[0006] The technical solution provided by this invention is as follows:

[0007] A system for calculating the conformity of registration accuracy indicators for multi-band aerospace optical remote sensing camera images includes: an information calculation module and an indicator analysis module; wherein, the information calculation module includes a data preparation unit and a data calculation unit, and the indicator analysis module includes an accuracy analysis unit and a result output unit;

[0008] The data preparation unit reads relevant parameters from the design specifications of the multi-band spaceborne optical remote sensing camera and sends them to the data calculation unit. The data calculation unit calculates the ground resolution of the on-orbit image for each spectral band and the pointing angle of the detector for each spectral band based on the relevant parameters, and provides this information to the accuracy analysis unit. The accuracy analysis unit combines each spectral band in pairs, calculates the degree of conformity of the registration accuracy index for each spectral band combination, and transmits this information to the result output unit. The result output unit generates a file containing the degree of conformity of the registration accuracy between each pair of spectral bands of the multi-band spaceborne optical remote sensing camera and displays it.

[0009] Furthermore, the data preparation unit reads relevant parameters from the design specifications of the multi-band spaceborne optical remote sensing camera, including: imaging object distance H, principal distance f, focal plane principal point coordinates (x0, y0), and the size p of a single detector element for each spectral band. i and the center coordinates (x) of each spectral band detector i y i ); where i is a positive integer representing the number of spectral segments.

[0010] Furthermore, the data calculation unit calculates the image object distance H, principal distance f, and detector element size p for each spectral band. i Calculate the ground resolution (GSD) of the on-orbit image for each spectral band. i :

[0011] GSD i =H×p i / f.

[0012] Furthermore, based on the center coordinates (x, y) of each spectral band detector... i y i ), and calculate the pointing angle θ of the detector for each spectral band using the principal point coordinates (x0, y0) and principal distance f on the focal plane. i :

[0013]

[0014] Furthermore, the accuracy analysis unit combines each spectral band in pairs and calculates the degree of conformity of the registration accuracy index for each spectral band combination, specifically including:

[0015] The precision analysis unit combines each spectral band in pairs, and calculates the difference Δθ between the pointing angles of each band for each combination. mn And select the GSD value with the larger ground resolution between the two spectral bands. mn m and n represent the number of the two spectral segments in the combination;

[0016] Set the maximum relative height dH of the observation area, and calculate the maximum matching error E between the m-band and n-band spectral ranges. mn ;

[0017] Traverse all spectral band combinations to obtain the maximum matching error E for each combination mn , the maximum matching error E for each spectral band combination mn Compare with the design registration accuracy index T to obtain the compliance degree of the registration accuracy index for each spectral band combination.

[0018] Furthermore, the maximum matching error E between the m spectral band and the n spectral band mn is

[0019] E mn = dH × tan(Δθ mn ) / GSD mn .

[0020] Furthermore, if E mn < T, it meets the index. If E mn > T, it does not meet the index. Obtain the compliance degree of the registration accuracy index for each spectral band combination and transmit it to the result output unit.

[0021] Furthermore, the present invention also proposes a method for calculating the compliance degree of the image registration accuracy index of a multi-spectral spaceborne optical remote sensing camera, including the following steps:

[0022] (1) Read the imaging object distance H, principal distance f, focal plane principal point (x0, y0), the size p of a single detector element of each spectral band detector i , the center (x i , y i ) of each spectral band detector, where i is a positive integer representing the number of spectral bands;

[0023] (2) Calculate the ground resolution GSD of the in-orbit image of each spectral band i ;

[0024] ×(3) Calculate the pointing angle θ of each spectral band detector i ;

[0025] (4) Set the maximum relative height dH and the registration accuracy index T of the observation area;

[0026] (5) Calculate the difference Δθ between the pointing angles of the m spectral band and the n spectral band mn , and select the ground resolution GSD for accuracy analysis mn ; [[ID=*59]]

[0027] (6) Calculate the maximum matching error E between the m spectral band and the n spectral band mn as:

[0028] E mn = dH × tan(Δθ mn ) / GSD mn ;

[0029] (7) Judge the compliance degree of the registration accuracy indexes of the m spectral band and the n spectral band. If E mn <T, it meets the indexes. If E mn >T, it does not meet the indexes;

[0030] (8) Repeat steps (5)-(7) until all spectral band combinations are traversed, and obtain the compliance degree of the registration accuracy indexes of every two spectral bands;

[0031] (9) Output the compliance degree of the registration accuracy of every two spectral bands obtained in step (8).

[0032] Further, the ground resolution GSD of the in-orbit images of each spectral band i is GSD i = H×p i / f;

[0033] The pointing angle θ of the detector of each spectral band i is:

[0034] Further, the difference Δθ between the pointing angles of the m spectral band and the n spectral band mn is specifically: Δθ mn = fabs(θ m - θ n );

[0035] Select the ground resolution GSD for accuracy analysis mn , specifically:

[0036] If GSD m > GSD n , then GSD mn takes the value of GSD m ;

[0037] If GSD m < GSD n , then GSD mn takes the value of GSD n .

[0038] The advantages of the present invention compared with the prior art are as follows:

[0039] (1) Estimate the registration accuracy of the remotely sensed images obtained after being in orbit during the design stage of the multi-spectral spaceborne optical remote sensing camera.

[0040] (2) Analyze the compliance degree of the registration accuracy indexes of the in-orbit multi-spectral images during the design stage of the multi-spectral spaceborne optical remote sensing camera, and guide the design of the multi-spectral spaceborne optical remote sensing camera.

[0041] (3) Based on the design specifications of a multi-band spaceborne optical remote sensing camera, this invention calculates the projection differences of the same ground object in different spectral bands, then calculates the matching error between remote sensing images acquired in orbit for each spectral band based on the projection differences, compares it with the required registration accuracy indicators, and analyzes the degree of compliance with the registration accuracy indicators. This method is designed for multi-band spaceborne optical remote sensing cameras and can quickly and accurately predict the registration accuracy of remote sensing images acquired after the multi-band spaceborne optical remote sensing camera is in orbit, analyze the degree of compliance with the registration accuracy indicators, and guide the design of multi-band spaceborne optical remote sensing cameras. Attached Figure Description

[0042] Figure 1 This is a flowchart illustrating an embodiment of the present invention;

[0043] Figure 2 This is a schematic diagram of the device structure of the present invention. Detailed Implementation

[0044] The basic idea of ​​this invention is to provide a device for calculating the degree of conformity of image registration accuracy indicators for multi-spectral spaceborne optical remote sensing cameras. This device calculates the projection differences of the same ground object in different spectral bands based on the design specifications of the multi-spectral spaceborne optical remote sensing camera. Then, it calculates the matching error between remote sensing images acquired in orbit for each spectral band based on the projection differences, compares this error with the required registration accuracy indicators, and analyzes the degree of conformity. This method, designed for multi-spectral spaceborne optical remote sensing camera design, can quickly and accurately predict the registration accuracy of remote sensing images acquired after the multi-spectral spaceborne optical remote sensing camera is in orbit, analyze the degree of conformity with the registration accuracy indicators, and guide the design of multi-spectral spaceborne optical remote sensing cameras.

[0045] The present invention will now be described in further detail with reference to the accompanying drawings.

[0046] like Figure 2 As shown, the present invention proposes a system for calculating the conformity of image registration accuracy indicators of aerospace optical remote sensing cameras, which includes an information calculation module and an indicator analysis module. The information calculation module includes a data preparation unit and a data calculation unit, and the indicator analysis module includes an accuracy analysis unit and a result output unit.

[0047] The data preparation unit reads the imaging object distance H, principal distance f, focal plane principal point (x0, y0), and individual detector element size p of each spectral band from the design specifications of the multi-band spaceborne optical remote sensing camera. i The center of each spectral band detector (x i y i ), where i is a positive integer representing the number of spectral segments.

[0048] Let H, f, p i , (x0, y0), (x i y i)Transmitted to the data calculation unit;

[0049] The data calculation unit calculates the ground resolution GSD of the on-orbit image for each spectral band according to the imaging object distance H, the principal distance f, and the detector element size p of each spectral band detector i Calculates the ground resolution GSD of the on-orbit image for each spectral band i = H × p i / f;

[0050] According to the center (x i , y i ) of each spectral band detector, the principal point (x0, y0), and the principal distance f, calculate the pointing angle of each spectral band detector

[0051] Transmit GSD i and θ i to the accuracy analysis unit; The accuracy analysis unit combines each pair of spectral bands, and for each combination, calculates the difference Δθ in the spectral band pointing angles mn , and selects the larger value GSD of the ground resolution in the two spectral bands mn , where m and n represent the two spectral band numbers in the combination;

[0052] Set the maximum relative height dH of the observation area, and calculate the maximum matching error E between the m spectral band and the n spectral band mn = dH × tan(Δθ mn ) / GSD mn , traverse all spectral band combinations, and obtain the maximum matching error of each combination;

[0053] Compare the maximum matching error of each spectral band combination with the designed registration accuracy index T. If E mn < T, it meets the index. If E mn > T, it does not meet the index, obtain the compliance degree of the registration accuracy index for each spectral band combination, and transmit it to the result output unit;

[0054] The result output unit forms a file output and displays the compliance degree of the registration accuracy of every two spectral bands of the multi-spectral spaceborne optical remote sensing camera.

[0055] Using the above system for calculating the compliance degree of the registration accuracy index of the multi-spectral spaceborne optical remote sensing camera image, the present invention also provides a method for calculating the compliance degree of the registration accuracy index of the spaceborne optical remote sensing camera image, as Figure 1 shown. The specific steps of this method are as follows:

[0056] (1) Read the imaging object distance H, the principal distance f, the principal point (x0, y0) of the focal plane, the size p of a single detector element of each spectral band detector, i in the design specifications of the multi-spectral spaceborne optical remote sensing camera, the center (x i , y i), where \(i\) is a positive integer representing the number of spectral bands;

[0057] (2) Calculate the ground resolution GSD of the on-orbit images for each spectral band i ;

[0058] GSD i is GSD i = H × p i / f;

[0059] (3) Calculate the pointing angle \(\theta\) of the detector for each spectral band i ;

[0060]

[0061] (4) Set the maximum relative height \(dH\) of the observation area and the registration accuracy index \(T\);

[0062] (5) Calculate the difference \(\Delta\theta\) in the pointing angles between the \(m\) -th and \(n\) -th spectral bands mn , and select the ground resolution GSD for accuracy analysis mn ;

[0063] \(\Delta\theta\) mn = fabs(\(\theta\) m - \(\theta\) n );

[0064] (6) Calculate the maximum matching error \(E\) between the \(m\) -th and \(n\) -th spectral bands mn is:

[0065] \(E\) mn = \(dH\times\tan(\Delta\theta\) mn ) / GSD mn ;

[0066] (7) Judge whether the registration accuracy index of the \(m\) -th and \(n\) -th spectral bands meets the requirement. If \(E\) mn < T, it meets the index; if \(E\) mn > T, it does not meet the index;

[0067] Select the ground resolution GSD for accuracy analysis mn , specifically:

[0068] If GSD m > GSD n , then GSD mn takes the value of GSD m ;

[0069] If GSD m < GSD​​​​​​​​(8) Repeat steps (5)-(7) until all combinations of spectral bands are traversed, and obtain the compliance of the registration accuracy index for each pair of spectral bands.

[0071] (9) Output the compliance of the registration accuracy for each pair of spectral bands obtained in step (8).

[0072] Example:

[0073] The calculation of the compliance of the image registration accuracy index of a spaceborne optical remote sensing camera is carried out as follows:

[0074] (1) The data preparation unit reads the imaging object distance H = 500000m, the principal distance f = 0.25m, the principal point of the focal plane (x0, y0) = (0,0), a total of 3 spectral bands, the size of a single detector element of each spectral band p1 = 5μm, p2 = p3 = 10μm, and the center of each spectral band detector (x1, y1) = (0,0.0001m), (x2, y2) = (0,0.0002m), (x3, y3) = (0,0.0003m);

[0075] (2) The data calculation unit calculates the ground resolution of the in-orbit images of each spectral band

[0076] GSD1 = H × p1 / f = 10m, GSD2 = GSD3 = H × p2 / f = 20m;

[0077] (3) The data calculation unit calculates the pointing angle of each spectral band detector

[0078]

[0079]

[0080]

[0081] (4) The accuracy analysis unit sets the maximum relative height dH = 8848m of the observation area and the registration accuracy index T = 0.3 pixel;

[0082] (5) The accuracy analysis unit calculates the difference in pointing angles Δθ between the first spectral band and the second spectral band 12 = fabs(θ1 - θ2) = 0.0229°, because GSD1 = 10m < GSD2 = 20m, select the accuracy analysis ground resolution GSD 12 = 20m, calculate the difference in pointing angles Δθ between the first spectral band and the third spectral band 13 = fabs(θ1 - θ3) = 0.0459°, because GSD1 = 10m < GSD3 = 20m, select the accuracy analysis ground resolution GSD 13 = 20m, calculate the difference in pointing angles Δθ between the second spectral band and the third spectral band23 = fabs(θ2 - θ3) = 0.0230°, because GSD2 = GSD3 = 20m, select the ground resolution GSD for accuracy analysis 23 = 20m;

[0083] (6) The accuracy analysis unit calculates the maximum matching error between the first spectral band and the second spectral band

[0084] E 12 = dH × tan(Δθ 12 ) / GSD 12 = 0.1768 pixels, calculate the maximum matching error E between the first spectral band and the third spectral band 13 = dH × tan(Δθ 13 ) / GSD 13 = 0.3544 pixels, calculate the maximum matching error E between the second spectral band and the third spectral band 23 = dH × tan(Δθ 23 ) / GSD 23 = 0.1776 pixels;

[0085] (7) The accuracy analysis unit calculates the compliance degree of the registration accuracy index between the first spectral band and the second spectral band, E 12 < T, meeting the index requirements, calculate the compliance degree of the registration accuracy index between the first spectral band and the third spectral band, E 13 > T, not meeting the index requirements, calculate the compliance degree of the registration accuracy index between the second spectral band and the third spectral band, E 23 < T, meeting the index requirements;

[0086] (8) The result output unit outputs the compliance degree of the registration accuracy of every two spectral bands obtained in step (7).

[0087] According to the design indexes of the multi - spectral space - borne optical remote - sensing camera, the present invention calculates the projection differences of the same ground objects imaged in each spectral band, then calculates the matching errors between the remote - sensing images obtained in each spectral band in orbit from the projection differences, compares with the required registration accuracy indexes, and analyzes the compliance degree of the registration accuracy indexes. This method is designed for multi - spectral space - borne optical remote - sensing cameras, and can quickly and accurately estimate the registration accuracy of the remote - sensing images obtained after the multi - spectral space - borne optical remote - sensing camera is in orbit, analyze the compliance degree with the registration accuracy indexes, and guide the design of multi - spectral space - borne optical remote - sensing cameras.

[0088] The parts not elaborated in detail in the present invention belong to the well - known technologies in the field.

Claims

1. A system for calculating the degree of conformity of image registration accuracy indicators from a multi-band aerospace optical remote sensing camera, characterized in that... include: The system includes an information calculation module and an indicator analysis module; the information calculation module includes a data preparation unit and a data calculation unit, and the indicator analysis module includes an accuracy analysis unit and a result output unit. The data preparation unit reads relevant parameters from the design specifications of the multi-band spaceborne optical remote sensing camera and sends them to the data calculation unit. The data calculation unit calculates the ground resolution of the on-orbit image for each spectral band and the pointing angle of the detector for each spectral band based on the relevant parameters, and provides this information to the accuracy analysis unit. The accuracy analysis unit combines each spectral band in pairs, calculates the degree of conformity of the registration accuracy index for each spectral band combination, and transmits this information to the result output unit. The result output unit generates a file containing the degree of conformity of the registration accuracy between each pair of spectral bands of the multi-band spaceborne optical remote sensing camera and displays it. The accuracy analysis unit combines each spectral band in pairs and calculates the degree of conformity of the registration accuracy index for each spectral band combination, specifically including: The precision analysis unit combines each spectral band in pairs, and calculates the difference Δθ between the pointing angles of each band for each combination. mn And select the GSD value with the larger ground resolution between the two spectral bands. mn m and n represent the number of the two spectral segments in the combination; Set the maximum relative height dH of the observation area, and calculate the maximum matching error E between the m-band and n-band spectral ranges. mn ; Iterate through all spectral combinations to obtain the maximum matching error E for each combination. mn The maximum matching error E for each spectral band combination mn By comparing with the designed registration accuracy index T, the degree of conformity of the registration accuracy index for each spectral band combination is obtained; The maximum matching error E between the m-band and n-band spectral bands mn for E mn =dH×tan(Δθ mn ) / GSD mn 。 2. The system for calculating the conformity of multi-band aerospace optical remote sensing camera image registration accuracy index according to claim 1, characterized in that: The data preparation unit reads relevant parameters from the design specifications of the multi-band aerospace optical remote sensing camera, including: imaging object distance H, principal distance f, focal plane principal point coordinates (x0, y0), and the size p of a single detector element for each spectral band. i and the center coordinates (x) of each spectral band detector i y i ); where i is a positive integer representing the number of spectral segments.

3. The system for calculating the conformity of multi-band aerospace optical remote sensing camera image registration accuracy index according to claim 2, characterized in that: The data calculation unit calculates the image object distance H, principal distance f, and detector element size p for each spectral band. i Calculate the ground resolution (GSD) of the on-orbit image for each spectral band. i : GSD i =H×p i / f。 4. The system for calculating the degree of conformity of image registration accuracy index of a multi-band aerospace optical remote sensing camera according to claim 3, characterized in that: Based on the center coordinates (x) of each spectral band detector i y i ), and calculate the pointing angle θ of the detector for each spectral band using the principal point coordinates (x0, y0) and principal distance f on the focal plane. i :

5. The system for calculating the conformity of multi-band aerospace optical remote sensing camera image registration accuracy index according to claim 1, characterized in that: If E mn <T, it meets the criteria. If E mn >T, it does not meet the criteria. Obtain the compliance of the registration accuracy criteria for each spectral band combination and transmit it to the result output unit.

6. A method for calculating the degree of conformity of image registration accuracy index of a multi-band aerospace optical remote sensing camera, characterized in that... Includes the following steps: (1) Read the imaging object distance H, principal distance f, focal plane principal point (x0, y0), and individual detector element size p of each spectral band in the design specifications of the multi-band aerospace optical remote sensing camera. i , the center of each spectral band detector (x i y i ), where i is a positive integer representing the number of spectral segments; (2) Calculate the ground resolution (GSD) of the on-orbit image for each spectral band. i ; (3) Calculate the pointing angle θ of the detector for each spectral band. i ; (4) Set the maximum relative height dH of the observation area and the registration accuracy index T; (5) Calculate the difference Δθ between the pointing angles of the m-band and the n-band. mn And select the ground resolution GSD for precision analysis mn ; (6) Calculate the maximum matching error E between the m-band and the n-band spectral band. mn for: E mn =dH×tan(Δθ mn ) / GSD mn ; (7) Judge the compliance degree of the registration accuracy indexes for the m spectral band and the n spectral band. If E mn < T, it meets the indexes. If E mn > T, it does not meet the indexes; (8) Repeat steps (5)-(7) until all spectral segments are combined and traversed to obtain the degree of conformity of the registration accuracy index of each pair of spectral segments; (9) Output the degree of conformity of the registration accuracy of each pair of spectral bands obtained from step (8).

7. The method for calculating the conformity of image registration accuracy index of a multi-band aerospace optical remote sensing camera according to claim 6, characterized in that: Ground resolution (GSD) of on-orbit images in various spectral bands i For GSD i =H×p i / f; The pointing angle θ of the detectors for each spectral band i for:

8. The method for calculating the conformity of image registration accuracy index of a multi-band aerospace optical remote sensing camera according to claim 6, characterized in that: The difference Δθ between the pointing angles of the m-band and the n-band mn Specifically: Δθ mn =fabs(θ) m -θ n ); Selecting the ground resolution GSD for precision analysis mn Specifically: If GSD m >GSD n Then GSD mn The value is GSD m ; If GSD m <GSD n Then GSD mn The value is GSD n .

Images