CN112381769B - Oversampling correction method and system based on functional fitting of lunar phase graph - Google Patents
Oversampling correction method and system based on functional fitting of lunar phase graph Download PDFInfo
- Publication number
- CN112381769B CN112381769B CN202011200555.2A CN202011200555A CN112381769B CN 112381769 B CN112381769 B CN 112381769B CN 202011200555 A CN202011200555 A CN 202011200555A CN 112381769 B CN112381769 B CN 112381769B
- Authority
- CN
- China
- Prior art keywords
- moon
- satellite
- oversampling
- lunar
- axis
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- 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)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an oversampling correction method and system based on functional fitting of a lunar phase graph, wherein the method comprises the following steps: based on contrast or edge identification, extracting the pixel lengths of the pre-acquired lunar image in the satellite scanning direction and the satellite movement direction; calculating satellite parameters and moon parameters according to the ephemeris; the satellite parameters and the lunar parameters are respectively the scanning speed of the remote sensor along the satellite motion direction and the relative position of the moon, the satellite and the sun; calculating a lunar phase angle and an orbit angle based on the extracted pixel length, and constructing a theoretical lunar phase graphic function for obtaining a satellite view angle; determining a conversion coefficient of a pre-acquired lunar image and a theoretical lunar phase graphic function through function fitting to obtain an oversampling coefficient; and correcting the moon shape stretching caused by the time sequence recombination of the satellite observation moon images based on the oversampling coefficient to realize oversampling correction. The method can accurately eliminate the influence of oversampling on the lunar radiation correction.
Description
Technical Field
The invention belongs to the technical field of satellite remote sensing and lunar calibration, and particularly relates to an oversampling correction method and system based on functional fitting of a lunar phase graph.
Background
The remote sensor consists of an optical part and an electrical part, and the two parts together complete the remote sensing work. The scaling factor is used to quantify the conversion between the electrical count information transmitted back to the earth and the optical information actually detected by the optical detector.
After the remote sensing satellite enters the orbit, the remote sensing satellite is influenced by aspects such as space environment, instrument state and the like, the actual working environment of the remote sensing satellite is greatly different from laboratory calibration, and the calibration coefficient of the remote sensor in the orbit is different from that before emission. Therefore, on-track radiometric calibration is required to modify the calibration coefficients of remote sensors arriving at the track in real time. In the solar radiation band, the moon is commonly used as a stable external radioactive calibration source. The remote sensing satellite makes a detector observe a complete two-dimensional image through transverse scanning and longitudinal satellite movement. However, when a satellite for earth observation observes the moon, because the earth-satellite distance is greatly different from the moon-satellite distance, the speed of scanning the moon is not matched with the moving speed of the satellite, so that two adjacent frames are not connected end to end but have repeated parts, and oversampling is caused; the method comprises the steps of causing an actually observed moon image to have stretching deformation in the satellite moving direction, and actually repeatedly shooting a plurality of moon by one moon image; when the lunar radiation is calibrated, the influence of the factor needs to be eliminated by calculating an oversampling coefficient (reflecting the repetition rate).
At present, there are two main types of oversampling removal methods commonly used in engineering. One of the methods is that the satellite parameters, the remote sensor scanning speed and the moon parameters are calculated by using ephemeris adopted by MODIS, and oversampling coefficients are simulated. The other is a method for analyzing a moon image adopted by SeaWIFS, and oversampling is analyzed by analyzing an observed actual moon image; the influence of the actual condition of the satellite on oversampling is considered; the method has the defects that the traditional oversampling algorithm in the SeaWIFS is only suitable for the condition of low lunar phase angle, only uses a specific frame of data in a lunar image, the effect depends on the resolution ratio of the image seriously, and the practicability is poor.
In summary, a new method and system for oversampling correction based on functional fitting of a lunar phase graph is needed.
Disclosure of Invention
The present invention is directed to an oversampling correction method and system based on functional fitting of a lunar phase graph, so as to solve one or more of the above-mentioned technical problems. According to the method, the actual oversampling method is obtained by processing the actual moon image, so that the influence of the accuracy of analog calculation on oversampling analysis is reduced; the method can realize the oversampling analysis of the lunar image to obtain the oversampling coefficient containing the image repetition rate information, and can solve the problems that the traditional algorithm is small in applicable lunar phase range and seriously depends on the image resolution and the like; errors caused by the previous image segmentation and extraction on the moon can be inhibited to a certain extent, and the practicability is high; the influence of oversampling on lunar radiation correction can be accurately eliminated.
In order to achieve the purpose, the invention adopts the following technical scheme:
the invention discloses an oversampling correction method based on functional fitting of a lunar phase graph, which comprises the following steps of:
In step 2, the lunar phase angle is the included angle of the satellite, the moon and the sun, and the orbital angle is the included angle between the scanning speed along the satellite motion direction and the normal vector of the satellite, the moon and the sun plane.
The further improvement of the invention is that in the step 3, the conversion coefficient of the pre-acquired lunar image and the theoretical lunar phase graphic function is determined through function fitting, when the oversampling coefficient is obtained, the lunar phase shape is determined by the lunar outer contour and the lunar light and dark boundary line, the lunar phase shape is rotated into the observation coordinate system of the remote sensor, and the lunar phase function of the lunar phase shape observed by the remote sensor is obtained; in the remote sensor observation coordinate system, the abscissa axis is along the scanning direction of the satellite scanning mirror, and the ordinate axis is along the satellite movement direction.
The further improvement of the invention is that in step 3, the conversion coefficient of the pre-acquired lunar image and the theoretical lunar phase graphic function is determined through function fitting, and when the oversampling coefficient is obtained, the oversampling coefficient is obtained by adopting a single-line mode and a multi-line mode; the fitting algorithm uses a least squares method.
The further improvement of the present invention is that, in step 3, the step of obtaining the oversampling coefficient in the single-line mode specifically includes: the value of the lunar phase function f (x) is the length of a secant of the lunar phase cut by a straight line x ═ x; taking the maximum value of the lunar phase function f (x) to obtain an oversampling coefficient of a single line mode; the step of obtaining the oversampling coefficients using the multiline mode specifically includes: magnifying an original image with the size of M multiplied by N into M multiplied by (N multiplied by N) by cubic spline interpolation, wherein N is a scaling factor; the original image is a digitized image and is composed of discrete pixels, and the length of the secant pixels is an integer.
The further improvement of the present invention is that, in step 3, the step of obtaining the oversampling coefficient in the single-line mode specifically includes:
establishing a coordinate system to enable the Y axis to point to the satellite position, enabling the direct solar radiation direction to be located in an XY plane, enabling the included angle between the X axis and the direct solar radiation direction to be an acute angle, and enabling the Z axis to be vertical to the XY plane; in the morning and evening of the moon, onlyThere are solid lines that can be observed at satellite positions; a space rectangular coordinate system is established by taking the center of the moon as an origin, the included angle between sunlight and an X axis is theta, and the moon phase angle is
The oversampling factor of the single-line mode is expressed as
In the formula (I), the compound is shown in the specification,
is the over-sampling coefficient, xi is the angular width corresponding to one pixel, D moon Is the diameter of the moon, Y moon Length of pixel, S, which is the diameter of moon in the moon image along the direction of satellite motion sm Is the satellite to moon distance;
wherein the content of the first and second substances,
in the formula, Y obs The pixel length of the longest secant in all lunar sections along the satellite motion direction, κ (α, τ) is Y moon And Y obs The conversion relation of (1), alpha is a lunar phase angle, and tau is an included angle between the satellite motion direction and a lunar rotation axis;
wherein the content of the first and second substances,
in the formula, F is the pixel length of the longest secant in all lunar sections in the X direction, and gamma is the included angle between the satellite motion direction and the Z axis;
wherein, F is obtained by a linear equation of the morning and evening line, the moon boundary and the satellite motion direction through an approximate solution method of assignment comparison,
x 2 +z 2 ＝R 2 ；
in the formula, b is the intercept of a linear equation where the motion direction of the satellite is located on the Z axis.
The further improvement of the present invention is that, in step 3, the step of obtaining the oversampling coefficients by using the multiline mode specifically includes:
establishing a coordinate system to enable the Y axis to point to the satellite position, enabling the direct solar radiation direction to be located in an XY plane, enabling the included angle between the X axis and the direct solar radiation direction to be an acute angle, and enabling the Z axis to be vertical to the XY plane; establishing a remote sensor scanning coordinate system, and setting a mu axis along the scanning direction of a scanning mirror and a upsilon axis along the satellite motion direction; f (u) is the length of a line section of a straight line which is perpendicular to the mu axis and is cut by the moon phase, and gamma is an included angle between the upsilon axis and the Z axis; the coordinate axis direction of the image coordinate system u' is the same as u;
the coordinate system u' in the image is linear with the actual coordinate u, and is expressed as:
u' ═ Bu + C, B is a constant;
the pixel secant in the same image is in a proportional relationship with the actual secant, and is expressed as:
f′(u′)＝Af(u′)＝Af(Bu+C)；
the length correspondence, obtained by multiparameter least squares fitting, is expressed as:
in the formula: upsilon' iobs Is different u 'extracted from the actual image' i The secant length of the location;
the oversampling correction factor is defined as:
in the formula: f. of os For oversampling coefficient, upsilon' is the secant length in the actual image, theta is the resolution of the view angle, f (x) is the functional relation between the pixel secant and the actual secant, A is the proportion of the pixel secant to the actual secant, S sm Is the satellite to moon distance;
the original image of M × N size is magnified to M × (N × N) by cubic spline interpolation, and after adding a scaling factor N, the oversampling correction coefficient is expressed as
The invention relates to an oversampling correction system based on functional fitting of a lunar phase graph, which comprises:
the pixel length acquisition module is used for extracting the pixel lengths of the pre-acquired lunar image in the satellite scanning direction and the satellite motion direction according to the contrast or edge identification;
the theoretical lunar phase graph function acquisition module is used for calculating satellite parameters and lunar parameters according to ephemeris; the satellite parameters and the lunar parameters are respectively the scanning speed of the remote sensor along the satellite motion direction and the relative position of the moon, the satellite and the sun; calculating a lunar phase angle and an orbit angle based on the extracted pixel length, and constructing a theoretical lunar phase graphic function for obtaining a satellite view angle;
the correction module is used for determining a conversion coefficient of a pre-acquired lunar image and a theoretical lunar phase graphic function through function fitting to obtain an oversampling coefficient; and correcting the moon shape stretching caused by the time sequence recombination of the satellite observation moon images based on the oversampling coefficient to realize oversampling correction.
The invention is further improved in that in the theoretical moon phase graphic function acquisition module, the moon phase angle is an included angle of a satellite, a moon and the sun, and the orbit angle is an included angle between a scanning speed along the satellite motion direction and a normal vector of a satellite, the moon and the sun plane.
The correction module is used for determining a conversion coefficient of a pre-acquired lunar image and a theoretical lunar phase graphic function through function fitting, and acquiring an oversampling coefficient by adopting a single-line mode and a multi-line mode when acquiring the oversampling coefficient; the fitting algorithm uses a least squares method.
Compared with the prior art, the invention has the following beneficial effects:
the method can realize the oversampling analysis of the moon image according to the relevant parameters of the satellite and the moon, thereby obtaining the oversampling coefficient containing the image repetition rate information; the problems that the conventional algorithm is small in applicable lunar phase range and seriously depends on image resolution and the like are solved; errors caused by the previous image segmentation and extraction on the moon can be inhibited to a certain extent, and the method is suitable for engineering practice; the influence of oversampling on lunar radiation correction can be accurately eliminated. Specifically, the oversampling correction method of the invention can complete the oversampling correction of the lunar image; compared with the classic SeaWIFS oversampling correction algorithm, the method can better correct the moon oversampling images of all lunar phase values, enlarges the time range of satellite observation moon, and can meet the actual requirement of remote sensing moon calibration. The dependence of the accuracy of the oversampling correction on the satellite pointing precision and the probe element pointing precision is greatly reduced, and the upper limit of the oversampling correction precision can be improved; the fitting process makes it possible to achieve good performance in low resolution images. The method can process the original images of different remote sensing satellites for observing the moon by substituting the space parameters of different satellites, and has good transportability and practicability.
In the invention, two different modes of single line and multi-line are provided to obtain the oversampling coefficient, wherein the fitting algorithm adopts a least square method, and infinitely approaches to an actual lunar image sequence, thereby having higher accuracy and application range. Although the single-line mode can accurately derive the oversampling coefficient by the length of the longest cut line in theory, in practice, since a digitized image is used, the length can only take an integer value of a pixel, and since the optical diffusion of the edge is used, the use of only the longest cut line causes a large error, which is particularly remarkable in a low-resolution lunar image. The invention adopts a method of functionalizing the lunar phase in the image coordinate system, and brings the whole secant sequence in the lunar image into the calculation of the oversampling coefficient, thereby improving the accuracy of the oversampling coefficient to the maximum extent.
In the present invention, the step of obtaining the oversampling coefficient by using the multiline mode specifically includes: the image is a digitized image, the image is composed of discrete pixels, so the length of the secant pixels can only take an integer, the original image with the size of M multiplied by N is amplified into M multiplied by (N multiplied by N) through cubic spline interpolation, N is defined as a scaling factor, and after the scaling factor is added, the resolution of the image can be improved, so that the oversampling coefficient is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art are briefly introduced below; it is obvious that the drawings in the following description are some embodiments of the invention, and that for a person skilled in the art, other drawings can be derived from them without inventive effort.
FIG. 1 is a schematic diagram of a mode of observing the moon by a satellite according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of lunar phases observed by a satellite in an embodiment of the present invention; wherein (a) in fig. 2 is a three-dimensional lunar phase diagram, and (b) in fig. 2 is a lunar orthographic diagram observed along the Y-axis;
FIG. 3 is a schematic representation of monthly overcrowding and insufficient moon in an embodiment of the present invention; wherein (a) in fig. 3 is a monthly excess diagram, and (b) in fig. 3 is a monthly deficit diagram;
FIG. 4 is a schematic diagram of an image coordinate system and a real coordinate system and a schematic diagram of an over-sampling analysis based on the image coordinate system according to an embodiment of the present invention; wherein (a) in fig. 4 is a schematic diagram of a lunar phase between a camera coordinate system and an actual coordinate system, (b) in fig. 4 is a schematic diagram of a lunar phase in an image coordinate system after oversampling is considered, (c) in fig. 4 is a schematic diagram of a digital image of a lunar phase without oversampling, and (d) in fig. 4 is a schematic diagram of a digital image of a lunar phase with oversampling;
FIG. 5 is a schematic diagram of three cases of cutting the thread in the embodiment of the present invention; wherein the content of the first and second substances,
and with
Is a left intersection point and a right intersection point of a straight line perpendicular to the mu axis and the excircleDot vector, a 1 Is the intersection point vector of a straight line vertical to the mu axis and the moon morning and evening line;
FIG. 6 is a diagram illustrating the fitting results of a lunar phase function according to an embodiment of the present invention; wherein the scaling ratio of the cubic spline is n-3, the points are the actual lunar length in the data, and the lines are the fitting result of the lunar phase function; the time of (a) in FIG. 6 is 2000-07-22T19:20, the time of (b) in FIG. 6 is 2001-01-15T06:40, the time of (c) in FIG. 6 is 2002-01-03T16:10, and the time of (d) in FIG. 6 is 2003-01-22T17: 35;
FIG. 7 is a schematic diagram illustrating a comparison of the result of inverting the normalized scaling factor with the result of the SeaWIFS oversampling algorithm in accordance with an embodiment of the present invention; wherein Orgin is the result which is not subjected to sampling correction, Paper is the result which is subjected to SeaWIFS oversampling algorithm correction, and New-opti1000 is the result of the inversion scaling coefficient of the invention;
FIG. 8 is a schematic diagram illustrating a comparison between an inversion normalization scaling factor result and an MODIS over-sampling algorithm result according to an embodiment of the present invention;
FIG. 9 is a schematic diagram illustrating comparison between an inversion normalization scaling coefficient result and an MODIS oversampling algorithm result after Kalman smoothing according to an embodiment of the present invention; the verification data is analyzed by using the actual lunar data of the 8 th wave band in the time from MODIS 2000-07-23T11:38:01 to 2006-08-15T00:23: 46.
Detailed Description
In order to make the purpose, technical effect and technical solution of the embodiments of the present invention clearer, the following clearly and completely describes the technical solution of the embodiments of the present invention with reference to the drawings in the embodiments of the present invention; it is to be understood that the described embodiments are part of the present invention. Other embodiments, which can be derived by one of ordinary skill in the art from the disclosed embodiments without inventive faculty, are intended to be within the scope of the invention.
The oversampling correction method based on functional fitting of the lunar phase graph comprises the following steps:
referring to fig. 1, first, a lunar image acquired from a remote sensing satellite cold space observation window is processed: and extracting the pixel length of the moon in the satellite scanning direction and the satellite motion direction according to the contrast or the edge identification.
And then, calculating the satellite parameters and the moon parameters at the current moment according to the ephemeris, and constructing a moon phase graph function of the satellite view angle at the current moment.
And finally, determining a conversion coefficient of an actual shooting moon and a theoretical lunar phase function through function fitting so as to obtain an oversampling coefficient, and further correcting moon shape stretching caused by time sequence recombination of satellite observation moon images.
The embodiment of the invention specifically comprises the following steps: determining the shape of the moon phase according to the moon phase angle and the moon outline and the moon light and dark boundary; rotating the orbit angle into a remote sensor observation coordinate system according to the orbit angle to obtain a lunar phase function of a lunar phase shape observed by a remote sensor; the abscissa axis is along the scanning direction of the satellite scanning mirror, and the ordinate axis is along the movement direction of the satellite; the value of the lunar phase function f (x) is the secant length of the lunar phase cut by a straight line x which is the x, the maximum value of the lunar phase function f (x) is taken, thereby obtaining the oversampling coefficient of the single line mode, and the oversampling coefficient is combined with the actual lunar image secant sequence, and the least square fitting standard is that
Magnifying an original image with the size of M multiplied by N into M multiplied by (N multiplied by N) by cubic spline interpolation, defining N as a scaling factor, adding the scaling factor, and correcting an oversampling correction coefficient into
In summary, the embodiment of the invention discloses an oversampling correction method based on functional fitting of a lunar phase graph, which can complete the function of oversampling correction of a lunar image. Firstly, compared with the classical SeaWIFS oversampling correction algorithm, the algorithm can better correct the moon oversampling images of all lunar phase values, enlarges the time range of satellite observation of the moon, and can be suitable for the actual requirement of remote sensing moon calibration; secondly, the dependence of the over-sampling correction accuracy on the satellite pointing accuracy and the probe element pointing accuracy is greatly reduced, the upper limit of the over-sampling correction accuracy is improved, and the fitting process enables the over-sampling correction accuracy to still achieve good performance in low-resolution images. And finally, by substituting the space parameters of different satellites, the algorithm can process original images of different remote sensing satellites for observing the moon, and has good transportability and practicability.
In the embodiment of the invention, two different modes of single line and multi-line are provided to obtain the oversampling coefficient, wherein the fitting algorithm adopts a least square method, and infinitely approaches to an actual lunar image sequence, so that the method has higher accuracy and application range.
Referring to fig. 2, the single-line mode in the embodiment of the present invention includes:
and establishing a coordinate system to enable the Y axis to point to the satellite position, wherein the direct solar radiation direction is positioned in the XY plane, the included angle between the X axis and the direct solar radiation direction is an acute angle, and the Z axis is vertical to the XY plane. In the evening of the moon, only the solid line is observed at the satellite position. A space rectangular coordinate system is established by taking the center of the moon as an origin, the included angle between sunlight and an X axis is theta, and the moon phase angle is
The expression for the morning and evening line and moon boundary is:
x 2 +z 2 ＝R 2 (2)
the linear equation expression of the satellite motion direction is as follows,
from the above equation, one can obtain:
in the formula: and x' are respectively the abscissa of the intersection point of the lunar outer boundary and the straight line where the morning and evening line and the satellite motion direction are located.
Order to
The longest secant is then as follows:
to obtain the longest secant, | x-x' | should be maximized, i.e.
Since the moon is an opaque sphere, only one edge of the morning and evening line ellipse is in the satellite field of view. The theoretical calculation is too complicated, and an approximate solving method of assignment comparison is adopted when solving the maximum value, namely, a b value is taken at every other tiny interval, the b value corresponding to the maximum F is selected through a program, and the b value is approximately considered to be the corresponding longitudinal intercept when the maximum value is taken by the F, and the F is the projection of the maximum tangent length on the x axis. Therefore, the first and second electrodes are formed on the substrate,
in the formula:
in the image, R is equal to
Thus, the lunar secant length and the lunar diameter Y in the image observed in any lunar phase are established Moon (α, γ) relationship, the longest secant length to be obtained from the image
Calculated Y Moon Formula of correction factor (7)
The oversampling factor of the single-line mode can be obtained.
The multi-line mode of the embodiment of the invention comprises the following steps:
although the single-line mode can accurately derive the oversampling coefficient by the length of the longest cut line in theory, in practice, since a digitized image is used, the length can only take an integer value of a pixel, and since the optical diffusion of the edge is used, the use of only the longest cut line causes a large error, which is particularly remarkable in a low-resolution lunar image. The method for functionalizing the lunar phase in the image coordinate system is adopted, and the whole secant sequence in the lunar image is incorporated into the calculation of the oversampling coefficient, so that the accuracy of the oversampling coefficient is improved to the maximum extent.
Referring to FIG. 4, a remote sensor scanning coordinate system is established, where μ axis is along the scanning direction of the scanning mirror, υ axis is along the satellite motion direction, and XZ axis is defined as described above in the present invention.
f(u 1 ) Is u 1 The length of a line segment which is intersected by the moon of a straight line (namely, a straight line in the same direction with the motion direction of the satellite) which is vertical to the mu axis, and gamma is an included angle between the upsilon axis and the Z axis, namely an orbital angle.
The image coordinate system u' has coordinate axes oriented in the same direction as u, but due to oversampling, there is some stretching in the ν axis.
Definition of
And
is the bit vector of the left intersection point and the right intersection point of the straight line vertical to the mu axis and the excircle,
is a line perpendicular to the mu axis andthe bit vector of the intersection point of the moon morning and evening lines,
x 1 respectively, the projection of the three intersection points on the x-axis.
The secant is divided into three cases, wherein the content of all the intersections can be solved in the XZ coordinate system to obtain:
(1) two intersection points are all on the outer circle
Conditions are as follows:
and
exists with abscissa greater than 0, a 1 There is no, i.e. (4) the first equation has no real solution and the second equation has a real solution.
(2) Two intersections, one at the morning and evening line and one at the outer circle
Conditions are as follows:
and
the presence of the one or more of the one,
the abscissa is greater than 0 and
abscissa is less than 0, a 1 Are present. That is, (4) both equations have real solutions, an
Is less than 0, and is less than 0,
greater than 0.
(3) The two intersection points are both located at the morning and evening line
Conditions are as follows: a is 1 Both solutions in the equation are real solutions.
(4) Not intersecting with the moon
Conditions are as follows:
and
is absent, a 1 There is no, i.e. (4) neither equation has a real solution.
f(x)＝0 (11)
The time for shooting the moon by the satellite is extremely short, and the scanning speed and the scanning direction of the moon by the satellite along the motion direction can be considered to be unchanged when shooting the moon. The coordinate system u' in the image should be linear with the actual coordinates u, i.e. it should be linear
u' ═ Bu + C (B is constant) (12)
The time for shooting the moon by the satellite is extremely short, and it can be considered that when shooting the moon, the satellite keeps unchanged along the motion direction to the scanning speed direction of the moon, that is, the oversampling coefficients of the whole image formed by the single image elements are the same (the stretching degrees are the same), and the pixel secant and the actual secant in the same image should be in a direct proportion relation.
f′(u′)＝Af(u′)＝Af(Bu+C) (13)
Since the image is a digitized image, the actual data points are not continuous functions, but rather are a set of discrete values, and due to edge light diffusion, etc., the actual data points have some error. Therefore, the total variance of all data points is ensured to be minimum through multi-parameter least square fitting, and the most accurate length corresponding relation is fitted
In the formula: v' iobs Is different u 'extracted from actual image' i And (4) solving the A which is the proportion of the image shooting length in the Y-axis direction to the actual length of the secant length of the position.
The oversampling correction factor is defined as:
in the formula: θ is the viewing angle resolution, i.e., the angular width of one pixel; r sm The distance of the moon from the satellite.
Since the images taken by the satellites are digitized images, which are made up of discrete pixels, the secant pixel lengths can only be taken as integers. In order to improve the resolution of the image, the original image of M × N size is enlarged to M × (N × N) by cubic spline interpolation, and after adding a scaling factor N, the oversampling correction coefficient is changed to
Finally, oversampling correction is carried out on the original moon image by using an oversampling coefficient to obtain a moon image without oversampling influence; the comparison of the results in FIG. 6 shows that the lunar phase function of the present invention can accurately reflect the shape of the lunar phase, and better realize the analysis and correction of the oversampling, and the comparison of the results in FIG. 7 shows that compared with the SeaWIFS classic algorithm, the curve is smoother, the curve trend is more obvious, and the change trend of the calibration coefficient along with the time can be reflected better. The comparison of the results of fig. 8 and 9 shows that even when the low-resolution image which is not good for processing by the algorithm is processed, the accuracy of the low-resolution image can still be kept the same as that of the classic MODIS algorithm, after the uncertainty caused by other factors is removed by kalman smoothing, the results of the low-resolution image and the classic MODIS algorithm are highly consistent, the heights are close to the true values, and the accurate parameters of the satellite required by the algorithm are far less than those of the MODIS algorithm. The invention improves the accuracy and the application range of the oversampling algorithm, thereby providing favorable conditions for putting the algorithm into practical engineering application.
The embodiment of the invention discloses an oversampling correction method based on functional fitting of a lunar phase graph, which comprises the following steps: firstly, processing a moon image acquired from a remote sensing satellite cold space observation window: and according to the contrast or the edge identification, extracting the pixel length of the moon in the satellite scanning direction and the satellite movement direction. And then, calculating the satellite parameters and the moon parameters at the current moment according to the ephemeris, and constructing a moon phase graph function of the satellite view angle at the current moment. Finally, through function fitting, determining the conversion coefficient of the actual shooting moon and the theoretical moon phase function, quantifying the moon shape stretching due to oversampling, and calculating the oversampling coefficient according to the method. Compared with the prior art, the method has the characteristics of wide application range, accuracy and stability, and is particularly suitable for the conditions of large lunar phase range and insufficient satellite precise parameters, so that the accuracy and the application range of the oversampling algorithm are improved, and favorable conditions are provided for putting the algorithm into practical engineering application.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and so forth) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art can make modifications and equivalents to the embodiments of the present invention without departing from the spirit and scope of the present invention, and such modifications and equivalents are within the scope of the claims of the present invention as hereinafter claimed.
Claims (2)
1. An oversampling correction method based on functional fitting of a lunar phase graph is characterized by comprising the following steps:
step 1, extracting the pixel lengths of a pre-acquired lunar image in a satellite scanning direction and a satellite movement direction based on contrast or edge identification;
step 2, calculating satellite parameters and moon parameters according to the ephemeris; the satellite parameters and the moon parameters are respectively the scanning speed of the remote sensor along the satellite motion direction and the relative position of moon, satellite and sun;
calculating to obtain a moon phase angle and an orbit angle based on the satellite parameters, the moon parameters and the pixel length extracted in the step 1, and constructing a theoretical moon phase graphic function of the satellite view angle based on the moon phase angle and the orbit angle;
step 3, determining a conversion coefficient of the pre-acquired lunar image and a theoretical lunar phase graphic function through function fitting to obtain an oversampling coefficient; correcting the moon shape stretching caused by the time sequence recombination of the satellite observation moon images based on the oversampling coefficient to realize oversampling correction;
determining a conversion coefficient of a pre-acquired lunar image and a theoretical lunar phase graphic function through function fitting in the step 3, and acquiring an oversampling coefficient by adopting a single-line mode and a multi-line mode when the oversampling coefficient is acquired;
the step of obtaining the oversampling coefficient in the single-line mode specifically includes:
establishing a coordinate system to enable the Y axis to point to the satellite position, wherein the direct solar radiation direction is positioned in an XY plane, the included angle between the X axis and the direct solar radiation direction is an acute angle, and the Z axis is vertical to the XY plane; in the evening, only the solid line can be observed at the satellite position; a space rectangular coordinate system is established by taking the center of the moon as an origin, the included angle between sunlight and an X axis is theta, and the moon phase angle is
The oversampling factor of the single-line mode is expressed as
In the formula (I), the compound is shown in the specification,
is the over-sampling coefficient, xi is the angular width corresponding to one pixel, D moon Is the diameter of the moon, Y moon Length of pixel, R, of diameter of moon in moon image along satellite motion direction sm Is the satellite to moon distance;
wherein the content of the first and second substances,
in the formula, Y obs The pixel length of the longest secant in all lunar sections along the direction of satellite motion,
is Y moon And Y obs The relationship of (a) to (b) is,
is a moon phase angle, tau is an included angle between the satellite motion direction and the moon rotation axis;
wherein the content of the first and second substances,
in the formula, F is the pixel length of the longest secant in all lunar sections in the X direction, and gamma is the included angle between the satellite motion direction and the Z axis;
wherein, F is obtained by a linear equation of the morning and evening line, the moon boundary and the satellite motion direction through an approximate solution method of assignment comparison,
x 2 +z 2 ＝R 2 ；
in the formula, b is the intercept of a linear equation where the satellite motion direction is located at the Z axis;
the step of obtaining the oversampling coefficients using the multiline mode specifically includes:
establishing a coordinate system to make the Y axis point to the satellite position, the direct solar radiation direction is in the XY plane, the X axis and the straight lineThe included angle of the shooting directions is an acute angle, and the Z axis is vertical to the XY plane; establishing a remote sensor scanning coordinate system, and setting a mu axis along the scanning direction of a scanning mirror and a upsilon axis along the satellite motion direction; f (u) is the length of the section of the line where u is perpendicular to the mu axis and is intersected by the lunar phase, which is the lunar phase angle
And a function of the orbit angle gamma, wherein the orbit angle gamma is an included angle between the upsilon axis and the Z axis; the coordinate axis direction of the image coordinate system u' is the same as u;
the coordinate system u' in the image is linear with the actual coordinate u, and is expressed as:
u' ═ Bu + C, B is a constant;
the pixel secant in the same image is in a proportional relation with the actual secant, and is expressed as follows:
f′(u′)＝Af(u′)＝Af(Bu+C)；
the length correspondence, obtained by multiparameter least squares fitting, is expressed as:
in the formula: upsilon' iobs For different u extracted from actual image i ' secant length of position;
the oversampling correction factor is defined as:
in the formula: f. of os V' is the length of the cut line in the actual image, theta is the resolution of the viewing angle, f (x) is the functional relationship between the pixel cut line and the actual cut line, A is the ratio of the pixel cut line to the actual cut line, R sm Is the satellite to moon distance;
the original image of M × N size is magnified to M × (N × N) by cubic spline interpolation, and after adding a scaling factor N, the oversampling correction coefficient is expressed as
2. An oversampling correction system based on functional fitting of a lunar phase image, comprising:
the pixel length acquisition module is used for extracting the pixel lengths of the pre-acquired lunar image in the satellite scanning direction and the satellite motion direction according to the contrast or edge identification;
the theoretical lunar phase graph function acquisition module is used for calculating satellite parameters and lunar parameters according to ephemeris; the satellite parameters and the moon parameters are respectively the scanning speed of the remote sensor along the satellite motion direction and the relative position of moon, satellite and sun; calculating a lunar phase angle and an orbit angle based on the extracted pixel length, and constructing a theoretical lunar phase graphic function for obtaining a satellite view angle;
the correction module is used for determining a conversion coefficient of a pre-acquired lunar image and a theoretical lunar phase graphic function through function fitting to obtain an oversampling coefficient; correcting the moon shape stretching caused by the time sequence recombination of the satellite observation moon images based on the oversampling coefficient to realize oversampling correction;
wherein the content of the first and second substances,
determining a conversion coefficient of a pre-acquired lunar image and a theoretical lunar phase graphic function through function fitting, and acquiring an oversampling coefficient by adopting a single-line mode and a multi-line mode when the oversampling coefficient is acquired;
the step of obtaining the oversampling coefficient in the single-line mode specifically includes:
establishing a coordinate system to enable the Y axis to point to the satellite position, enabling the direct solar radiation direction to be located in an XY plane, enabling the included angle between the X axis and the direct solar radiation direction to be an acute angle, and enabling the Z axis to be vertical to the XY plane; in the morning and evening of the moon, only the solid line can be observed at the satellite position; a space rectangular coordinate system is established by taking the center of the moon as an origin, the included angle between sunlight and an X axis is theta, and the moon phase angle is
The oversampling factor of the single-line mode is expressed as
In the formula (I), the compound is shown in the specification,
is the over-sampling coefficient, xi is the angular width corresponding to one pixel, D moon Is the diameter of the moon, Y moon Length of pixel, R, of diameter of moon in the moon image along the satellite movement direction sm Is the satellite to moon distance;
wherein, the first and the second end of the pipe are connected with each other,
in the formula, Y obs The pixel length of the longest secant in all sections of the moon along the direction of motion of the satellite,
is Y moon And Y obs The relationship of (a) to (b) is,
is a moon phase angle, tau is an included angle between the satellite motion direction and the moon rotation axis;
wherein, the first and the second end of the pipe are connected with each other,
in the formula, F is the pixel length of the longest secant in all lunar sections in the X direction, and gamma is the included angle between the satellite motion direction and the Z axis;
wherein, F is obtained by a linear equation of the morning and evening line, the moon boundary and the satellite motion direction through an approximate solution method of assignment comparison,
x 2 +z 2 ＝R 2 ；
in the formula, b is the intercept of a linear equation where the satellite motion direction is located at the Z axis;
the step of obtaining the oversampling coefficients using the multiline mode specifically includes:
establishing a coordinate system to enable the Y axis to point to the satellite position, wherein the direct solar radiation direction is positioned in an XY plane, the included angle between the X axis and the direct solar radiation direction is an acute angle, and the Z axis is vertical to the XY plane; establishing a remote sensor scanning coordinate system, and setting a mu axis along the scanning direction of a scanning mirror and a upsilon axis along the satellite motion direction; f (u) is the length of the section of the line where u is perpendicular to the mu axis and is intersected by the lunar phase, which is the lunar phase angle
And a function of the orbit angle gamma, wherein the orbit angle gamma is an included angle between the upsilon axis and the Z axis; the coordinate axis direction of the image coordinate system u' is the same as u;
the coordinate system u' in the image is linear with the actual coordinate u, and is expressed as:
u' ═ Bu + C, B is a constant;
the pixel secant in the same image is in a proportional relationship with the actual secant, and is expressed as:
f′(u′)＝Af(u′)＝Af(Bu+C)；
the length correspondence, which is fit by multiparameter least squares, is expressed as:
in the formula: v is a cell i ′ obs For different u extracted from actual image i ' secant length of position;
the oversampling correction factor is defined as:
in the formula: f. of os V' is the length of the cut line in the actual image, theta is the resolution of the view angle, f (x) is the functional relation between the pixel cut line and the actual cut line, and A is the pixel cut line andratio of actual secant, R sm Is the satellite to moon distance;
the original image of M × N size is magnified to M × (N × N) by cubic spline interpolation, and after adding a scaling factor N, the oversampling correction coefficient is expressed as
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011200555.2A CN112381769B (en) | 2020-10-30 | 2020-10-30 | Oversampling correction method and system based on functional fitting of lunar phase graph |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011200555.2A CN112381769B (en) | 2020-10-30 | 2020-10-30 | Oversampling correction method and system based on functional fitting of lunar phase graph |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112381769A CN112381769A (en) | 2021-02-19 |
| CN112381769B true CN112381769B (en) | 2022-09-27 |
Family
ID=74577563
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011200555.2A Active CN112381769B (en) | 2020-10-30 | 2020-10-30 | Oversampling correction method and system based on functional fitting of lunar phase graph |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112381769B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113589343B (en) * | 2021-07-19 | 2023-07-25 | 中国科学院微小卫星创新研究院 | Moon center vector and sun direction extraction method based on moon imaging sensor |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106908813A (en) * | 2017-04-28 | 2017-06-30 | 国家卫星气象中心 | A kind of satellite remote sensor Calibration Method and device |
| WO2019062166A1 (en) * | 2017-09-30 | 2019-04-04 | 中国科学院遥感与数字地球研究所 | Method for automatic geometric correction of cross-platform moon-based earth observation imaging |
| CN111256818A (en) * | 2020-02-05 | 2020-06-09 | 国家卫星气象中心(国家空间天气监测预警中心) | Calculation method for moon irradiance observed value of linear array push-broom type remote sensing imager |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10524663B2 (en) * | 2015-06-19 | 2020-01-07 | The Board Of Trustees Of The University Of Illinois | Phase measurement, analysis, and correction methods for coherent imaging systems |
-
2020
- 2020-10-30 CN CN202011200555.2A patent/CN112381769B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106908813A (en) * | 2017-04-28 | 2017-06-30 | 国家卫星气象中心 | A kind of satellite remote sensor Calibration Method and device |
| WO2019062166A1 (en) * | 2017-09-30 | 2019-04-04 | 中国科学院遥感与数字地球研究所 | Method for automatic geometric correction of cross-platform moon-based earth observation imaging |
| CN111256818A (en) * | 2020-02-05 | 2020-06-09 | 国家卫星气象中心(国家空间天气监测预警中心) | Calculation method for moon irradiance observed value of linear array push-broom type remote sensing imager |
Non-Patent Citations (4)
| Title |
|---|
| FY-3D MERSI On-Orbit Radiometric Calibration from the Lunar View;Ronghua Wu 等;《Sensors》;20200820;第1-13页 * |
| SeaWiFS lunar calibration methodology after six years on orbit;Robert Eplee 等;《Earth Observing Systems IX》;20041026;第1-13页 * |
| 低轨高分辨率遥感卫星姿态机动对月定标方法;朱军等;《光学精密工程》;20200930;第28卷(第09期);第1913-1923页 * |
| 基于月球反射的遥感仪器在轨定标方法;牛明慧等;《遥感技术与应用》;20180431;第33卷(第02期);第337-341页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112381769A (en) | 2021-02-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8938111B2 (en) | Computed tomography imaging process and system | |
| Wei et al. | Implicit and explicit camera calibration: Theory and experiments | |
| CN109272574B (en) | Construction method and calibration method of linear array rotary scanning camera imaging model based on projection transformation | |
| US20070146391A1 (en) | Method of and system for determining inaccuracy information in an augmented reality system | |
| US20040150643A1 (en) | Method for digitally rendering an object using measured BRDF data | |
| EP3273412A1 (en) | Three-dimensional modelling method and device | |
| CN105913435B (en) | A kind of multiscale morphology image matching method and system suitable for big region | |
| JPH07120434B2 (en) | Method and apparatus for volume rendering | |
| CN103136758A (en) | Rapid edge detecting method based on orthogonal polynomial fitting | |
| CN112381769B (en) | Oversampling correction method and system based on functional fitting of lunar phase graph | |
| CN111627070B (en) | Method, device and storage medium for calibrating rotation shaft | |
| Garding | Direct estimation of shape from texture | |
| WO2020136523A1 (en) | System and method for the recognition of geometric shapes | |
| US20080131029A1 (en) | Systems and methods for visualizing and measuring real world 3-d spatial data | |
| Baltsavias et al. | Metric information extraction from SPOT images and the role of polynomial mapping functions | |
| CN112085752A (en) | Image processing method, device, equipment and medium | |
| CN111915666A (en) | Volume measurement method and device based on mobile terminal | |
| CN112097798B (en) | High-precision calibration method and device for high resolution camera of high resolution six-satellite | |
| Oglesby et al. | Development and assessment of a photogrammetry system for rock specimen surface characterization | |
| Maset et al. | Scales of oblique photographs updated | |
| CN112070810A (en) | Positioning method, mobile device and computer readable storage medium | |
| US20210042995A1 (en) | Gridding global data into a minimally distorted global raster | |
| MacKay et al. | 3D Galatea: Entry of three-dimensional moving points from multiple perspective views | |
| CN109785384A (en) | The level height scaling method of level machine vision metrology instrument in a kind of transparent vessel | |
| CN116225236B (en) | Intelligent home scene interaction method based on acousto-optic control |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |