Large-dynamic-range satellite-borne visible light camera solar altitude correction method and system

Abstract

The invention provides a large-dynamic-range satellite-borne visible light camera solar altitude correction method and system. The method comprises the steps that 1, the entrance pupil radiance of a visible light camera is calculated; step 2, stray light suppression based on the image is carried out to acquire atmospheric layer top radiance; step 3, carrying out pixel-by-pixel geographic positioning on the visible light camera; 4, calculating the sun position and the earth position of a centroid celestial coordinate system BCRS at the corresponding imaging moment and a conversion matrix from a geocentric celestial coordinate system GCRS to an international terrestrial reference system ITRS; 5, calculating a solar zenith angle cosine value corresponding to each pixel observation print; 6, calculating a correction coefficient of each pixel; and 7, calculating the observation radiance of the large-dynamic-range satellite-borne visible light camera after solar altitude correction. The method can be widely applied to large-dynamic-range satellite-borne visible light camera solar elevation angle correction and radiation quantity correction of covering solar elevation angles close to 0 degree and smaller than 0 degree, calculation is simple, the method is reasonable, and the method is suitable for application.

Description

Technical Field

[0001] The present application belongs to the field of radiation correction, and specifically, relates to a method and system for correcting the solar altitude angle of a satellite-borne visible light camera with a large dynamic range. Background Art

[0002] For spaceborne cameras in the visible spectrum, the solar radiation received by the ground or clouds will be different when the sun is in different positions. This will cause the image to have radiation differences that vary with seasons and geographical locations on a large time scale; and radiation differences at different locations in the image on a small time scale.

[0003] The purpose of solar altitude correction is to correct the radiation incident at different solar inclination angles to the radiation when the sun is vertically incident, eliminating the radiation differences caused by different solar altitude angles in different seasons, times and locations.

[0004] In terms of obtaining the solar altitude angle, a Chinese patent document with publication number CN106289156A discloses a method for obtaining the solar altitude angle of a photographing point when a satellite is imaging in an arbitrary posture, the steps of which are as follows: (1) calculating the equivalent roll angle and pitch angle of the camera body coordinate system relative to the orbital coordinate system based on the attitude matrix of the satellite during maneuvering imaging and the camera installation matrix on the satellite body system; (2) calculating the geocentric angle corresponding to the large arc segment from the ground photographing point to the subsatellite point based on the result of step (1) and the satellite orbit altitude; (3) calculating the spherical angle between the large arc segment from the photographing point to the subsatellite point and the forward direction of the subsatellite line based on the result of step (1); (4) calculating the right ascension and declination of the ground photographing point based on the results of the first two steps in combination with the real-time orbital elements; (5) calculating the solar altitude angle of the photographing point based on the result of step (4) in combination with the solar direction vector available to the attitude and orbit control subsystem itself. This prior art provides a method for calculating the solar altitude angle of a photographing point using the solar direction vector and satellite attitude, camera installation matrix, and real-time orbital element information.

[0005] In terms of the solar altitude correction method, the traditional method is the cosine correction method, which is suitable for situations where the solar altitude angle is large. The Chinese patent document with publication number CN105183989A discloses a method for inverting the surface reflectivity of Landsat8 satellite data, in which the correction for the influence of different solar altitude angles adopts the zenith angle cosine correction method. The Chinese patent document with publication number CN104951656A discloses a method for inverting the surface reflectivity of wide-field satellite images, specifically including a method for inverting the surface reflectivity under a large field of view angle, in which the correction for the influence of different solar altitude angles also adopts the zenith angle cosine correction method. The Chinese patent document with publication number CN108009392A discloses a remote sensing reflectivity model construction and calibration application method for densely vegetated surfaces, specifically including a remote sensing reflectivity calibration method for densely vegetated surfaces, in which the correction for the influence of different solar altitude angles also adopts the zenith angle cosine correction method. A Chinese patent document with publication number CN106600586A discloses a TAVI adjustment factor algorithm based on the solar altitude angle, specifically including a method for calculating a terrain adjusted vegetation index (TAVI) adjustment factor based on the solar altitude angle, and calculating the adjustment factor based on the apparent reflectance (where s is a sensor parameter and α is the solar altitude angle).

[0006] Papers and References: Zhou Jingfei, Research on Radiometric Calibration and Solar Zenith Angle Correction of MODIS L1B Data, Journal of Southwest Forestry University, Vol. 30, No. 3, 2010, published the solar zenith angle correction of MODIS L1B data using the cosine correction method. Papers and References: Feng Shuangshuang, Application Research on Normalization Model of Terrain Correction of Remote Sensing Images with Low Solar Altitude Angle in Mountainous Areas - Taking Gaoligong Mountain as an Example, Journal of Hebei Agricultural University, Vol. 39, No. 4, 2016, published the terrain correction experiment of Landsat images with low solar altitude angle in the southern section of Gaoligong Mountain using three normalization models: second-order correction model, terrain balance model, and slope matching model. Papers and Documents: Li Hao, Research on the chlorophyll remote sensing detection capability of ocean color satellites under large solar zenith angles, Acta Oceanologica Sinica, Vol. 40, No. 11, 2018, using the vector radiation transfer model PCOART-SA that takes into account the curvature of the earth, simulated the satellite remote sensing detection limit capability of chlorophyll concentration under a solar zenith angle of 80° (i.e., a solar altitude angle of 10°). For a large dynamic range satellite-borne visible light camera, its imaging dynamic range can cover the situation where the solar altitude angle is close to 0° and less than 0°. At this time, due to the effects of airglow and moonlight, the large dynamic range satellite-borne visible light camera can still image, but the energy is weak, and the traditional cosine correction method is no longer applicable. Papers and Documents: Yu Yue, NPP / VIIRS low-light image fusion algorithm for urban lights in central and eastern China, Progress in Laser and Optoelectronics, Vol. 55, 2018, in the process of generating urban light products in central and eastern China, the influence of factors such as clouds, moonlight, solar radiation, lightning, and fires is comprehensively considered, and scenes affected by moonlight and solar radiation are screened out. Paper reference: Li Lin, The influence of solar altitude angle on image energy and its correction, Journal of Atmospheric and Ambient Light, Vol. 8, No. 1, 2013. For solar altitude angles greater than 18°, a cosine correction method is used, and for solar altitude angles between 1° and 18°, a cubic polynomial is used to construct a correction model.

[0007] Currently, there is no patent for a correction method for large dynamic range satellite-borne visible light cameras, especially when the solar altitude angle is close to 0° and less than 0°.

[0008] Therefore, it is necessary to design a method that can perform altitude angle correction on the radiation imaging data within the full dynamic range of a large dynamic range satellite-borne visible light camera, realize the normalization of the radiation data, and facilitate subsequent application processing. Summary of the invention

[0009] In view of the defects in the prior art, the purpose of the present application is to provide a method and system for correcting the solar altitude angle of a satellite-borne visible light camera with a large dynamic range.

[0010] A method for correcting the solar altitude angle of a large dynamic range spaceborne visible light camera provided by the present application includes:

[0011] Step 1: Calculate the entrance pupil radiance of the visible light camera obtained by direct radiation correction according to the radiation calibration formula and coefficient of the satellite-borne visible light camera;

[0012] Step 2: performing image-based stray light suppression on the spaceborne visible light camera according to the entrance pupil radiance of the visible light camera, and calculating the top of atmosphere radiance acquired by the spaceborne visible light camera;

[0013] Step 3: Perform pixel-by-pixel geolocation of the onboard visible light camera to obtain the three-dimensional position of each pixel observation footprint in the international ground reference coordinate system;

[0014] Step 4: According to the Earth orientation parameters, planetary ephemeris and the exposure time of the onboard visible light camera, calculate the conversion matrix from the barycentric celestial coordinate system BCRS sun position, the Earth position and the geocentric celestial coordinate system GCRS to the international terrestrial reference coordinate system ITRS at the corresponding imaging time;

[0015] Step 5: Calculate the cosine value of the solar zenith angle corresponding to the observation footprint of each pixel of the onboard visible light camera;

[0016] Step 6: Calculate the correction coefficient of each pixel according to the cosine value of the solar zenith angle;

[0017] Step 7: Calculate the observed radiance of the large dynamic range spaceborne visible light camera after solar altitude angle correction based on the correction coefficient.

[0018] Preferably, the top of atmosphere radiance calculation formula in step 2 is:

[0019] L=L1-L stray (Formula 1)

[0020] Where L is the top-of-atmosphere radiance obtained by the onboard visible light camera, L1 is the entrance pupil radiance of the visible light camera obtained by the direct radiation correction described in step 1, and L stray is the stray light component obtained based on the image analysis of the satellite-borne visible light camera.

[0021] Preferably, as described in step 4, the earth orientation parameters use the EOP data released by the International Earth Rotation and Reference System Service IERS, and the planetary ephemeris uses DE440 released by the Jet Propulsion Laboratory JPL.

[0022] Preferably, the calculation of the cosine value of the solar zenith angle in step 5 comprises the following steps:

[0023] Step 5.1: Calculate the sun's position in the geocentric celestial coordinate system

[0024]

[0025] in, is the sun position in BCRS, is the position of the Earth in BCRS;

[0026] Step 5.2: Calculate the ITRS coordinate system sun vector

[0027]

[0028] Among them, T GCRS2ITRS is the transformation matrix from the geocentric celestial coordinate system to the international terrestrial reference coordinate system corresponding to the imaging moment, is the sun vector in the geocentric celestial coordinate system;

[0029]

[0030] Step 5.3: Calculate the local normal vector of the observed footprint of the onboard visible light camera

[0031]

[0032] Among them, θ, is the geographic latitude and longitude of the three-dimensional position of the ground footprint obtained in step 3;

[0033] Step 5.4: Calculate the cosine of the solar zenith angle α,

[0034]

[0035] Here, · represents the inner product operation of two vectors.

[0036] Preferably, the correction coefficient of each pixel in step 6 adopts a semi-empirical model method.

[0037] Preferably, the correction coefficient f(α) is in piecewise function form:

[0038]

[0039] And satisfy

[0040]

[0041] Among them, c, b0, b1, and b2 are all parameters in the semi-empirical model, c is the low-end constant term; b0 is the constant term of the mid-end exponential term, b1 is the linear term of the mid-end exponential term, and b2 is the mid-end constant term. α1 and α2 are the inflection points of the piecewise function, α1<α2, f(α)>0.

[0042] Preferably, the method for obtaining the semi-empirical model parameters and calculating the final correction coefficient in step 6 comprises the following steps:

[0043] Step 6.1: According to the observation sensitivity, dynamic range and actual radiation response of the large dynamic range spaceborne visible light camera, determine the inflection point positions α1 and α2 of the piecewise function, where α1 corresponds to the observation sensitivity limit and α2 is the inflection point of the linear relationship between the radiation response of the large dynamic range spaceborne visible light camera and the cosine value of the solar zenith angle;

[0044] Step 6.2: Determine the constant correction coefficient c according to the radiation response value of α<α1;

[0045] Step 6.3: According to formula 8, use the numerical solution method of the transcendental equations to calculate b0, b1, and b2;

[0046] Step 6.4: For the cosine value of the solar zenith angle corresponding to the observation trace of each pixel described in step 5, calculate the correction coefficient f(α) according to the parameter values ​​in formula 7.

[0047] Preferably, in step 7, the method for calculating the radiance observed by the large dynamic range spaceborne visible light camera is pixel-by-pixel calculation.

[0048] Preferably, the pixel-by-pixel calculation formula is:

[0049]

[0050] Among them, L corrected Observing radiance for a large dynamic range spaceborne visible light camera.

[0051] A solar altitude angle correction system for a large dynamic range spaceborne visible light camera, comprising:

[0052] Module 1: used to calculate the entrance pupil radiance of the visible light camera obtained by direct radiation correction according to the radiation calibration formula and coefficient of the spaceborne visible light camera;

[0053] Module 2: used to perform image-based stray light suppression on the spaceborne visible light camera according to the entrance pupil radiance of the visible light camera, and calculate the top of atmosphere radiance acquired by the spaceborne visible light camera;

[0054] Module 3: Used to geolocate the visible light camera pixel by pixel, and obtain the three-dimensional position of each pixel observation footprint in the international ground reference coordinate system;

[0055] Module 4: used to calculate the conversion matrix from the barycentric celestial coordinate system BCRS sun position, the earth position and the geocentric celestial coordinate system GCRS to the international terrestrial reference coordinate system ITRS at the corresponding imaging time according to the earth orientation parameters, planetary ephemeris and visible light camera exposure time;

[0056] Module 5: used to calculate the cosine value of the solar zenith angle corresponding to the observation footprint of each pixel of the visible light camera;

[0057] Module 6: used to calculate the correction coefficient of each pixel according to the cosine value of the solar zenith angle;

[0058] Module 7: used to calculate the observed radiance of a large dynamic range spaceborne visible light camera after correction for the solar altitude angle based on the correction coefficient.

[0059] Compared with the prior art, this application has the following beneficial effects:

[0060] 1. This application provides a method for correcting the solar altitude angle of a satellite-borne visible light camera with a large dynamic range, and constructs a semi-empirical model covering a large dynamic range. In addition to being applicable to the case where the solar altitude angle is greater than 0°, it is also applicable to the case where the solar altitude angle is close to 0° and less than 0°;

[0061] 2. This application provides a method for solving model parameters, and the calculation method is simple;

[0062] 3. The present application provides a method for correcting the solar altitude angle of a satellite-borne visible light camera with a large dynamic range. The method is reasonable, easy to implement, and can be widely used in correcting the solar altitude angle of a satellite-borne visible light camera with a large dynamic range. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Other features, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting embodiments with reference to the following drawings:

[0064] Figure 1 This is a flow chart of the solar altitude angle correction method for a large dynamic range spaceborne visible light camera;

[0065] Figure 2 This is the corresponding relationship diagram between the radiance observed by the onboard visible light camera and the cosine value of the solar zenith angle;

[0066] Figure 3 A curve diagram showing the variation of the correction coefficient constructed by the method of the present application with the cosine value of the solar zenith angle;

[0067] Figure 4 This is the relationship between the radiation of a frame of image from a satellite-borne visible light camera and the solar altitude angle corresponding to each pixel;

[0068] Figure 5 To use the correction factor Figure 4 The radiation map of the image radiation after the solar altitude angle correction. DETAILED DESCRIPTION

[0069] The present application is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present application, but do not limit the present application in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present application. These all belong to the protection scope of the present application.

[0070] The technical problem to be solved by this application is how to achieve radiation correction covering the cases where the solar altitude angle is close to 0° and less than 0°. In order to solve the above technical problem, this application adopts the following technical solution: a method for solar altitude correction of a large dynamic range satellite-borne visible light camera, comprising the following steps:

[0071] Step 1: According to the radiation calibration formula and coefficients of the spaceborne visible light camera, calculate the visible light camera entrance pupil radiance L1 obtained by direct radiation correction.

[0072] The calibration formula for each satellite-borne visible light camera includes: linear polynomial, quadratic polynomial, cubic polynomial, and the coefficients in the formula are different depending on the camera.

[0073] Step 2: Perform image-based stray light suppression on the spaceborne visible light camera and calculate the top-of-atmosphere radiance L obtained by the spaceborne visible light camera. The image-based stray light suppression method is:

[0074] L=L1-L stray (Formula 1)

[0075] Where L is the top-of-atmosphere radiance obtained by the onboard visible light camera, L stray is the stray light component obtained based on the image analysis of the satellite-borne visible light camera.

[0076] Different cameras, different satellite platforms, and different orbits may have different forms of stray light effects. It is necessary to analyze the stray light performance in the actual image and deduct it from the total entrance pupil radiance.

[0077] Step 3: Perform pixel-by-pixel geolocation of the visible light camera to obtain the three-dimensional position of each pixel observation footprint in the international ground reference coordinate system.

[0078] Step 4: Calculate the sun position and earth position in the barycentric celestial coordinate system (BCRS) at the corresponding imaging time according to the earth orientation parameters, planetary ephemeris and visible light camera exposure time, and the conversion matrix from the geocentric celestial coordinate system (GCRS) to the international terrestrial reference coordinate system (ITRS).

[0079] Specifically, the Earth orientation parameters use the EOP data released by the International Earth Rotation and Reference System Service (IERS), and the planetary ephemeris uses DE440 released by the Jet Propulsion Laboratory JPL.

[0080] Step 5: Calculate the cosine value of the solar zenith angle corresponding to the observation trace of each pixel of the visible light camera. The calculation of the cosine value of the solar zenith angle includes the following steps:

[0081] Step 5.1: Calculate the sun's position in the geocentric celestial coordinate system

[0082]

[0083] in is the sun position in BCRS, is the position of the Earth in BCRS,

[0084] Step 5.2: Calculate the ITRS coordinate system sun vector

[0085]

[0086] Among them, T GCRS2ITRS is the transformation matrix from the geocentric celestial coordinate system to the international terrestrial reference coordinate system corresponding to the imaging time, is the sun vector in the geocentric celestial coordinate system,

[0087]

[0088] Step 5.3: Calculate the local normal vector of the visible light camera observation footprint

[0089]

[0090] Among them, θ, is the geographic latitude and longitude of the three-dimensional position of the ground footprint obtained in step 3.

[0091] Step 5.4: Calculate the cosine of the solar zenith angle α,

[0092]

[0093] Here, · represents the inner product operation of two vectors.

[0094] Step 6: Calculate the correction coefficient of each pixel based on the cosine value of the solar zenith angle.

[0095] The correction coefficient of each pixel adopts the semi-empirical model method, and the piecewise function form of the correction coefficient f(α) is:

[0096]

[0097] And satisfy

[0098]

[0099] Among them, c, b0, b1, and b2 are all parameters in the semi-empirical model, c is the low-end constant term; b0 is the constant term of the mid-end exponential term, b1 is the linear term of the mid-end exponential term, and b2 is the mid-end constant term. α1 and α2 are the inflection points of the piecewise function, α1<α2, f(α)>0.

[0100] The semi-empirical model parameter acquisition and the final correction coefficient calculation method include the following steps:

[0101] Step 6.1: According to the observation sensitivity, dynamic range and actual radiation response of the large dynamic range spaceborne visible light camera, determine the inflection point positions α1 and α2 of the piecewise function, where α1 corresponds to the observation sensitivity limit and α2 is the inflection point of the linear relationship between the radiation response of the large dynamic range spaceborne visible light camera and the cosine value of the solar zenith angle.

[0102] Step 6.2: Determine the constant correction coefficient c based on the radiation response value of α<α1.

[0103] Step 6.3: According to Formula 8, use the numerical solution method of the transcendental equations to calculate b0, b1, and b2.

[0104] Step 6.4: For the cosine value of the solar zenith angle corresponding to each pixel observation footprint obtained in step 5, calculate the correction coefficient f(α) according to the parameter values ​​in formula 7.

[0105] Step 7: Calculate the observed radiance L of the large dynamic range spaceborne visible light camera after solar altitude angle correction based on the correction coefficient f(α). corrected .

[0106]

[0107] The present application also provides a solar altitude angle correction system for a large dynamic range satellite-borne visible light camera. The solar altitude angle correction system for a large dynamic range satellite-borne visible light camera can be implemented by executing the process steps of the solar altitude angle correction method for a large dynamic range satellite-borne visible light camera, that is, those skilled in the art can understand the solar altitude angle correction method for a large dynamic range satellite-borne visible light camera as a preferred implementation of the solar altitude angle correction system for a large dynamic range satellite-borne visible light camera.

[0108] A solar altitude angle correction system for a large dynamic range spaceborne visible light camera, comprising:

[0109] Module 1: used to calculate the entrance pupil radiance of the visible light camera obtained by direct radiation correction according to the radiation calibration formula and coefficient of the spaceborne visible light camera;

[0110] Module 2: used to perform image-based stray light suppression on the spaceborne visible light camera according to the entrance pupil radiance of the visible light camera, and calculate the top of atmosphere radiance acquired by the spaceborne visible light camera;

[0111] Module 3: Used to geolocate the visible light camera pixel by pixel, and obtain the three-dimensional position of each pixel observation footprint in the international ground reference coordinate system;

[0112] Module 4: used to calculate the conversion matrix from the barycentric celestial coordinate system BCRS sun position, the earth position and the geocentric celestial coordinate system GCRS to the international terrestrial reference coordinate system ITRS at the corresponding imaging time according to the earth orientation parameters, planetary ephemeris and visible light camera exposure time;

[0113] Module 5: used to calculate the cosine value of the solar zenith angle corresponding to the observation footprint of each pixel of the visible light camera;

[0114] Module 6: used to calculate the correction coefficient of each pixel according to the cosine value of the solar zenith angle;

[0115] Module 7: used to calculate the observed radiance of a large dynamic range spaceborne visible light camera after correction for the solar altitude angle based on the correction coefficient.

[0116] The preferred embodiments of the present application are given below in conjunction with the accompanying drawings to illustrate the technical solution of the present application in detail.

[0117] The traditional cosine correction method is applicable to the case of large solar altitude angles, when the observed radiation mainly comes from the reflected light under the sunlight. For satellite-borne visible light cameras with a large dynamic range, weak radiation can also be observed at low solar altitude angles and negative solar altitude angles. At this time, the observed radiation comes from multiple factors, including: airglow, moonlight, and stray radiation on the satellite.

[0118] This application provides a method for correcting the solar altitude angle of a large dynamic range spaceborne visible light camera. Figure 1 As shown:

[0119] First, this embodiment provides a medium-resolution spectral imager for the Fengyun-3E satellite. According to the satellite-borne visible light camera, the radiation calibration formula and coefficient, the entrance pupil radiance of the visible light camera can be calculated.

[0120] For large dynamic range satellite-borne visible light cameras, stray light will affect their radiation effects, especially at the low end of the dynamic range, where the effect of stray light is more obvious. Stray light is irrelevant to the observed target and is observation noise. In order to suppress the influence of stray light, image-based stray light suppression is required, and stray light is modeled as additive noise. If the entrance pupil radiance of the visible light camera obtained by direct radiation correction is L1, the image-based stray light suppression method is

[0121] L=L1-Lstray (Formula 1)

[0122] Where L is the top-of-atmosphere radiance obtained by the onboard visible light camera, L1 is the entrance pupil radiance of the visible light camera obtained by direct radiation correction, and L stray is the stray light component obtained based on the image analysis of the satellite-borne visible light camera.

[0123] Different cameras, different satellite platforms, and different orbits may have different forms of stray light effects. It is necessary to analyze the stray light performance in the actual image and deduct it from the total entrance pupil radiance.

[0124] The necessary step of solar altitude correction is to calculate the solar altitude, which requires knowing the relative position of the sun. The relative position of the sun can be described by the sun vector, or by the altitude angle and azimuth angle. The altitude angle and the zenith angle are complementary. First, the visible light camera is geo-positioned pixel by pixel to obtain the three-dimensional position of each pixel observation footprint in the international ground reference coordinate system.

[0125] Then the sun position of the international terrestrial reference coordinate system is calculated. The calculation of the sun's high-precision position information requires the use of astronomical parameters, including earth orientation parameters and planetary ephemeris. The earth orientation parameters can use the EOP data released by the International Earth Rotation and Reference System Service (IERS), and the planetary ephemeris can use the DE440 released by the Jet Propulsion Laboratory JPL. Combined with the exposure time of the visible light camera (UTC time), the sun position of the barycentric celestial coordinate system (BCRS), the earth position, and the conversion matrix from the geocentric celestial coordinate system (GCRS) to the international terrestrial reference coordinate system (ITRS) at the corresponding imaging time are calculated.

[0126] The position of the sun in the geocentric celestial coordinate system can be calculated from the planetary ephemeris

[0127]

[0128] in is the sun position in BCRS, is the position of the Earth in BCRS,

[0129] Convert the geocentric celestial coordinate system sun position to ITRS coordinate system sun vector

[0130]

[0131] Among them, T GCRS2ITRS is the transformation matrix from the geocentric celestial coordinate system to the international terrestrial reference coordinate system corresponding to the imaging time, is the sun vector in the geocentric celestial coordinate system,

[0132]

[0133] Calculate the local normal vector of the visible light camera observation footprint

[0134]

[0135] Among them, θ, is the geographic latitude and longitude of the ground footprint obtained in step 3.

[0136] According to the local normal vector and the sun vector observed by the visible light camera, the cosine value of the solar zenith angle α can be calculated.

[0137]

[0138] Here, · represents the inner product operation of two vectors.

[0139] The solar zenith angle and solar altitude angle can be calculated from the cosine value α of the solar zenith angle. The semi-empirical model in this application constructs the corresponding relationship between the correction coefficient and the cosine value of the solar zenith angle, so the specific value of the solar zenith angle or altitude angle does not need to be calculated, such as Figure 2 As shown, there is a correspondence between the radiance observed by the satellite-borne visible light camera and the cosine value of the solar zenith angle.

[0140] The correction coefficient of each pixel adopts the semi-empirical model method, and the correction coefficient f(α) has the following piecewise function form:

[0141]

[0142] And satisfy

[0143]

[0144] Among them, c, b0, b1, and b2 are all parameters in the semi-empirical model, c is the low-end constant term; b0 is the constant term of the mid-end exponential term, b1 is the linear term of the mid-end exponential term, and b2 is the mid-end constant term. α1 and α2 are the inflection points of the piecewise function, α1<α2, f(α)>0.

[0145] According to the observation sensitivity, dynamic range and actual radiation response of the large dynamic range spaceborne visible light camera, the inflection points α1 and α2 of the piecewise function are determined, where α1 corresponds to the observation sensitivity limit and α2 is the inflection point of the linear relationship between the radiation response of the large dynamic range spaceborne visible light camera and the cosine value of the solar zenith angle. The constant correction coefficient c is determined according to the radiation response value of α<α1.

[0146] When α1, α2, and c are determined, the transcendental equations are numerically solved to calculate b0, b1, and b2 (with a unique solution) according to Formula 8. For a large dynamic range spaceborne visible light camera, its radiation response can be considered stable within a certain period of time, so the parameters in the semi-empirical model are also fixed and do not need to be calculated for each frame.

[0147] According to the parameter values ​​in formula 7, substitute the cosine of the solar zenith angle of each pixel corresponding to the imaging time to calculate the correction coefficient f(α): Figure 3 The curve of the correction coefficient changing with the cosine value of the solar zenith angle is shown.

[0148] Finally, the observed radiance L of the large dynamic range spaceborne visible light camera after solar altitude angle correction is calculated according to the correction coefficient f(α) of each pixel. corrected , the corresponding relationship is as follows Figure 2 As shown:

[0149]

[0150] The effectiveness of the method of the present application is illustrated below with reference to the measured graphic data of a large dynamic range satellite-borne visible light camera.

[0151] Figure 4 This is the relationship between the radiation of a frame of image from a satellite-borne visible light camera and the solar altitude angle corresponding to each pixel;

[0152] Figure 5 To use the correction factor Figure 4 The radiation map of the image radiation after the solar altitude angle correction. Figure 4 In the image, the pixel radiation of a large dynamic range satellite-borne visible light camera has a strong correlation with the solar altitude angle. The overall radiation of pixels with a large solar altitude angle is higher. For pixels with a small solar altitude angle (close to 0° and negative values), the texture fluctuation of the radiation is no longer discernible. Figure 5 In the figure, after correction, there is no obvious correlation between the pixel radiation and the solar altitude angle. Figure 5 The fluctuation of pixel radiation in the image can better reflect the texture information of the ground objects observed by the satellite-borne visible light camera, especially for pixels with small solar altitude angle.

[0153] Those skilled in the art know that, in addition to implementing the system and its various devices, modules, and units provided by the present application in a purely computer-readable program code, it is entirely possible to implement the same functions in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system and its various devices, modules, and units provided by the present application can be considered as a hardware component, and the devices, modules, and units included therein for realizing various functions can also be regarded as structures within the hardware component; the devices, modules, and units for realizing various functions can also be regarded as both software modules for realizing the method and structures within the hardware component.

[0154] The above describes the specific embodiments of the present application. It should be understood that the present application is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the substantive content of the present application. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be arbitrarily combined with each other.

Claims

1. A method for correcting the solar altitude angle of a large dynamic range spaceborne visible light camera, characterized in that: include: Step 1: Calculate the entrance pupil radiance of the visible light camera obtained by direct radiation correction according to the radiation calibration formula and coefficient of the satellite-borne visible light camera; Step 2: performing image-based stray light suppression on the spaceborne visible light camera according to the entrance pupil radiance of the visible light camera, and calculating the top of atmosphere radiance acquired by the spaceborne visible light camera; Step 3: Perform pixel-by-pixel geolocation of the onboard visible light camera to obtain the three-dimensional position of each pixel observation footprint in the international ground reference coordinate system; Step 4: According to the Earth orientation parameters, planetary ephemeris and the exposure time of the onboard visible light camera, calculate the conversion matrix from the barycentric celestial coordinate system BCRS sun position, the Earth position and the geocentric celestial coordinate system GCRS to the international terrestrial reference coordinate system ITRS at the corresponding imaging time; Step 5: Calculate the cosine value of the solar zenith angle corresponding to the observation footprint of each pixel of the onboard visible light camera; Step 6: Calculate the correction coefficient of each pixel according to the cosine value of the solar zenith angle; Step 7: Calculate the observed radiance of the large dynamic range spaceborne visible light camera after solar altitude angle correction based on the correction coefficient.

2. The method for correcting the solar altitude angle of a large dynamic range spaceborne visible light camera according to claim 1, characterized in that: The calculation formula for the top of atmosphere radiance in step 2 is: L=L1-L stray (Formula 1) Where L is the top-of-atmosphere radiance obtained by the onboard visible light camera, L1 is the entrance pupil radiance of the visible light camera obtained by the direct radiation correction described in step 1, and L stray is the stray light component obtained based on the image analysis of the satellite-borne visible light camera.

3. The method for correcting the solar altitude angle of a large dynamic range spaceborne visible light camera according to claim 1, characterized in that: As described in step 4, the Earth orientation parameters use the EOP data released by the International Earth Rotation and Reference System Service (IERS), and the planetary ephemeris uses DE440 released by the Jet Propulsion Laboratory (JPL).

4. The method for correcting the solar altitude angle of a large dynamic range spaceborne visible light camera according to claim 1, characterized in that: The calculation of the cosine value of the solar zenith angle in step 5 comprises the following steps: Step 5.1: Calculate the sun's position in the geocentric celestial coordinate system in, is the sun position in BCRS, is the position of the Earth in BCRS; Step 5.2: Calculate the ITRS coordinate system sun vector Among them, T GCRS2ITRS is the transformation matrix from the geocentric celestial coordinate system to the international terrestrial reference coordinate system corresponding to the imaging moment, is the sun vector in the geocentric celestial coordinate system; Step 5.3: Calculate the local normal vector of the observed footprint of the onboard visible light camera Among them, θ, is the geographic latitude and longitude of the three-dimensional position of the ground footprint obtained in step 3; Step 5.4: Calculate the cosine of the solar zenith angle α, Here, · represents the inner product operation of two vectors.

5. The method for correcting the solar altitude angle of a large dynamic range spaceborne visible light camera according to claim 1, characterized in that: The correction coefficient of each pixel described in step 6 adopts the semi-empirical model method.

6. The method for correcting the solar altitude angle of a large dynamic range spaceborne visible light camera according to claim 5, characterized in that: The piecewise function form of the correction coefficient f(α) is: And satisfy Among them, c, b0, b1, and b2 are all parameters in the semi-empirical model, c is the low-end constant term; b0 is the constant term of the mid-end exponential term, b1 is the linear term of the mid-end exponential term, and b2 is the mid-end constant term. α1 and α2 are the inflection points of the piecewise function, α1<α2, f(α)>0.

7. The method for correcting the solar altitude angle of a large dynamic range spaceborne visible light camera according to claim 6, characterized in that: The method for obtaining the semi-empirical model parameters and calculating the final correction coefficient in step 6 includes the following steps: Step 6.1: According to the observation sensitivity, dynamic range and actual radiation response of the large dynamic range spaceborne visible light camera, determine the inflection point positions α1 and α2 of the piecewise function, where α1 corresponds to the observation sensitivity limit and α2 is the inflection point of the linear relationship between the radiation response of the large dynamic range spaceborne visible light camera and the cosine value of the solar zenith angle; Step 6.2: Determine the constant correction coefficient c according to the radiation response value of α<α1; Step 6.3: According to formula 8, use the numerical solution method of the transcendental equations to calculate b0, b1, and b2; Step 6.4: For the cosine value of the solar zenith angle corresponding to the observation trace of each pixel described in step 5, calculate the correction coefficient f(α) according to the parameter values ​​in formula 7.

8. The method for correcting the solar altitude angle of a large dynamic range spaceborne visible light camera according to claim 1, characterized in that: In step 7, the calculation method for the radiance observed by the large dynamic range spaceborne visible light camera is pixel by pixel calculation.

9. The method for correcting the solar altitude angle of a large dynamic range spaceborne visible light camera according to claim 8, characterized in that: The pixel-by-pixel calculation formula is: Among them, L corrected Observing radiance for a large dynamic range spaceborne visible light camera.

10. A solar altitude angle correction system for a large dynamic range satellite-borne visible light camera, characterized in that: include: Module 1: used to calculate the entrance pupil radiance of the visible light camera obtained by direct radiation correction according to the radiation calibration formula and coefficient of the spaceborne visible light camera; Module 2: used to perform image-based stray light suppression on the spaceborne visible light camera according to the entrance pupil radiance of the visible light camera, and calculate the top of atmosphere radiance acquired by the spaceborne visible light camera; Module 3: Used to geolocate the visible light camera pixel by pixel, and obtain the three-dimensional position of each pixel observation footprint in the international ground reference coordinate system; Module 4: used to calculate the conversion matrix from the barycentric celestial coordinate system BCRS sun position, the earth position and the geocentric celestial coordinate system GCRS to the international terrestrial reference coordinate system ITRS at the corresponding imaging time according to the earth orientation parameters, planetary ephemeris and visible light camera exposure time; Module 5: used to calculate the cosine value of the solar zenith angle corresponding to the observation footprint of each pixel of the visible light camera; Module 6: used to calculate the correction coefficient of each pixel according to the cosine value of the solar zenith angle; Module 7: used to calculate the observed radiance of a large dynamic range spaceborne visible light camera after correction for the solar altitude angle based on the correction coefficient.

Images