A method and system for determining parameters of a BRDF model

Abstract

The application discloses a kind of BRDF model parameter determination method and system, it is related to hyperspectral satellite earth observation technical field.The specific implementation of the method includes: receiving satellite imaging request;According to imaging time, the target angle between the main plane of sun and the satellite orbit plane of imaging satellite is determined;Using the ground feature characteristics of target ground object and target angle, the relative azimuth threshold database pre-constructed is matched to generate the observation task of imaging satellite;In the case where the observation task is multi-angle observation task, the atlas data returned by imaging satellite is received, atlas data is analyzed, and the BRDF model parameter of target ground object is fitted.The implementation can obtain the reflectivity that is greatly different under different observation angles when observing ground object, so as to efficiently and accurately fit to obtain a well-performing BRDF model, various ground elements can be efficiently identified and observed, and ground classification accuracy is improved and target detection false alarm rate is reduced.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of hyperspectral satellite earth observation, and particularly relates to a BRDF model parameter determination method and system. BACKGROUND

[0002] In existing earth observation, under different illumination conditions, environmental conditions and underlying surface conditions, the spectral curves of ground objects are different, resulting in the occurrence of spectral variation phenomenon. At present, the recognition problem of ground object elements under spectral variation is mainly solved by obtaining the BRDF (bidirectional reflectance distribution function) characteristics of the ground objects. The main methods for obtaining BRDF characteristics include ground measurement (including spectrometer, multi-angle observation platform, etc.), airborne measurement (including airborne camera, etc.), and BRDF characteristic measurement based on space-based hyperspectral satellites.

[0003] However, due to the limitations of test conditions, geographical conditions and other factors, the ground measurement and airborne measurement obtain very limited BRDF characteristic data of ground objects. Due to the existence of the spectral variation characteristics of ground objects, when collecting spectral information of ground objects, the measurement based on hyperspectral satellites obtains multiple reflectivities that are very close, which leads to the failure to fit accurate BRDF model parameters (i.e. ill-posed fitting), so that the effectiveness of BRDF calculation is not ideal, and thus the ground objects cannot be accurately and finely classified to identify the detection target. SUMMARY

[0004] Therefore, the application provides a BRDF model parameter determination method and system, which can obtain reflectivities that are greatly different under different observation angles when observing ground objects, so as to efficiently and accurately fit to obtain a BRDF model with good performance, and efficiently recognize and observe various ground object elements, thereby improving the ground object classification accuracy and reducing the false alarm rate of target detection and identification.

[0005] The technical scheme of the application is as follows:

[0006] In order to achieve the above-mentioned purpose, according to an aspect of an embodiment of the application, a BRDF model parameter determination method is provided, comprising:

[0007] receiving a satellite imaging request; wherein the satellite imaging request comprises an imaging scene, a target ground object corresponding to the imaging scene, ground object characteristics of the target ground object, and an imaging time T0;

[0008] determining a target angle between a solar principal plane and a satellite orbit plane of the imaging satellite according to the imaging time T0

[0009] using the ground object characteristics of the target ground object and the target angle to determine BRDF model parameters of the target ground object matching a pre-constructed relative azimuth angle threshold database to generate an observation task of the imaging satellite; wherein the observation task comprises a multi-angle observation task and a regular observation task;

[0010] in a case that the observation task is the multi-angle observation task, receiving atlas data returned by the imaging satellite, analyzing the atlas data, and fitting a BRDF model parameter of the target ground object.

[0011] Optionally, the using the ground object characteristics of the target ground object and the target angle matching a pre-constructed relative azimuth angle threshold database to generate an observation task of the imaging satellite, comprises:

[0012] according to the solar zenith angle θ of the target ground object s , LAI index and characteristic wave band, querying a relative azimuth angle threshold database to obtain a relative azimuth angle threshold set of the target ground object

[0013] comparing the target angle with the relative azimuth angle threshold set to determine whether there is

[0014] if yes, generating a multi-angle observation task of the imaging satellite.

[0015] Optionally, the construction method of the relative azimuth angle threshold database comprises:

[0016] setting the solar zenith angle, observation zenith angle, characteristic wave band, LAI index and relative azimuth angle of different ground objects;

[0017] inputting the solar zenith angle, observation zenith angle, characteristic wave band, LAI index and relative azimuth angle of different ground objects into a pre-trained Prosail model;

[0018] determining the reflectivity of different ground objects according to the output of the Prosail model;

[0019] using the LAI index, relative azimuth angle and reflectivity of different ground objects to construct a relative azimuth angle threshold database of different ground objects.

[0020] Optionally, the using the LAI index, relative azimuth angle and reflectivity of different ground objects to construct a relative azimuth angle threshold database of different ground objects comprises:

[0021] for each kind of ground object, using the reflectivity to determine the reflectivity variation coefficient of the ground object;

[0022] According to a preset variation threshold, a relative azimuth angle threshold corresponding to the reflectivity variation coefficient is determined;

[0023] The relative azimuth angle threshold database is generated.

[0024] Optionally, the imaging satellite is an agile hyperspectral satellite.

[0025] Optionally, the target included angle between the solar principal plane and the satellite orbit plane of the imaging satellite is determined according to the imaging time T0 Previously, it also includes:

[0026] According to the current time T, the imaging time T0, the satellite attitude maneuvering time τ attitude of the imaging satellite, the camera imaging preparation time τ camera of the imaging satellite, it is judged whether T, T0, τ attitude satisfy T0-T>τ attitude , and whether T, T0, τ camera satisfy T0-T>τ camera .

[0027] In the case of T0-T>τ attitude and T0-T>τ camera , it is determined that the imaging satellite can perform observation tasks.

[0028] Optionally, the target included angle between the solar principal plane and the satellite orbit plane of the imaging satellite is determined according to the imaging time T0 , including:

[0029] According to the imaging time T0, the solar zenith angle θ s and the inclination of the solar principal plane compared to the equator

[0030] According to the satellite orbit inclination of the imaging satellite and the inclination , the target included angle between the solar principal plane and the satellite orbit plane is calculated

[0031] According to another aspect of the embodiment of the present application, a BRDF model parameter determination system is provided, including:

[0032] A receiving module is configured to receive a satellite imaging request; wherein the satellite imaging request includes an imaging scene, a target ground object corresponding to the imaging scene, a ground object characteristic of the target ground object, and an imaging time T0;

[0033] A data processing module is configured to determine a target included angle between a solar principal plane and a satellite orbit plane of the imaging satellite according to the imaging time T0

[0034] The matching module is used for matching the target ground object feature and the target angle The pre-constructed relative azimuth angle threshold database is matched to generate an observation task of the imaging satellite; wherein the type of the observation task includes a multi-angle observation task and a regular observation task;

[0035] The fitting module is used for receiving the atlas data returned by the imaging satellite, analyzing the atlas data, and fitting the BRDF model parameters of the target ground object when the observation task is the multi-angle observation task.

[0036] Advantages:

[0037] (1) In order to improve the calculation efficiency of the BRDF feature, the application provides a BRDF model parameter determination method and system, the corresponding relationship between the spectral variation degree and the relative azimuth angle under different ground object spatial structures, the solar zenith angle and the characteristic wave band is obtained by using the uniform ground object (the uniform ground object refers to that the ground object basically covers an area, and each shooting is the same ground object) obtained by ground test or simulation, a relative azimuth angle threshold database meeting the BRDF model parameter fitting requirement is established. The target angle is matched with the angle between the satellite orbit on the satellite and the solar principal plane, and whether the multi-angle observation of the ground object is needed is judged by comparison, so that the effective multi-angle observation is ensured to obtain the BRDF feature, and the blind and ineffective multi-angle observation of the ground object is avoided, the effectiveness of the satellite observation of the ground object and the acquisition efficiency of the BRDF feature are improved, and the economic cost and time cost are saved.

[0038] (2) The key of the application lies in establishing the relative azimuth angle threshold database of the reflectivity variation coefficient and the relative azimuth angle, effectively supporting the discrimination of whether the multi-angle observation task needs to be performed on the satellite, and then discriminating whether the variation degree of the target ground object meets the BRDF fitting requirement, so as to improve the measurement efficiency of the BRDF feature and the use efficiency of the multi-angle observation of the satellite on the ground object.

[0039] (3) The application firstly proposes that the hyperspectral satellite performs single-track multi-angle observation on the ground object, uses the agile maneuvering capability of the agile hyperspectral satellite to quickly adjust the attitude angle of the satellite to obtain multi-angle hyperspectral observation data of the ground object, and greatly reduces the economic cost and time cost required by data support to fit the BRDF feature, and reduces the user processing difficulty and data storage cost. BRIEF DESCRIPTION OF DRAWINGS

[0040] Figure 1 It is a flowchart of the BRDF model parameter determination method according to the embodiment of the application.

[0041] Figure 2 Flowchart of the method for determining the target angle according to the embodiment of the present application.

[0042] Figure 3 Flowchart of the method for constructing the relative azimuth angle threshold database according to the embodiment of the present application.

[0043] Figure 4 Flowchart of the method for determining the type of the observation task of the imaging satellite according to the embodiment of the present application.

[0044] Figure 5 Flowchart of the method for performing the multi-angle observation task of the imaging satellite according to the embodiment of the present application.

[0045] Figure 6 Module diagram of the system for determining the BRDF model parameters according to the embodiment of the present application. DETAILED DESCRIPTION

[0046] BRDF characteristics, as a branch of spectral variation phenomena, can be quantitatively analyzed by using BRDF model. BRDF characteristics are mainly related to factors such as target characteristics, seasons, and time. The spectral variation characteristics of the target can be obtained by multi-angle observation of the target, calculation of the BRDF model parameters related to the sun illumination direction and the camera observation direction, and coping with the influence of the same object with different spectra and the same spectrum with different objects on the application of hyperspectral remote sensing. Since BRDF characteristics are constrained by factors such as the sun zenith angle, the observation zenith angle of the satellite, the observation azimuth angle of the satellite, and the satellite orbit angle (the satellite orbit angle affects the observation zenith angle and the azimuth angle of the satellite), the reflectivity obtained by multi-angle observation of the same target needs to have sufficient variation to meet the inversion of the BRDF model parameters.

[0047] On the one hand, the traditional hyperspectral satellite usually blindly observes the ground object from multiple angles, and does not effectively distinguish the observation angle of the ground object, so that the efficiency and accuracy of BRDF characteristic acquisition are low, and the amount of collected data is too large, which consumes high economic cost and time cost. On the other hand, the traditional hyperspectral satellite has poor maneuverability, and the multi-angle observation of the ground object mainly adopts the way of revisiting (the satellite visits the same ground object multiple times) and side swing (different swing angles of the satellite correspond to different side swing angles), which makes the observation efficiency and the efficiency of BRDF characteristic acquisition low. For example, the currently popular MODIS data is mainly derived from the data collection of a wide imaging spectrometer, and the ground object is imaged only once in a single orbit (one revolution of the satellite), and multiple revisits are required to realize multi-angle observation of the ground object, which is low in observation efficiency. Moreover, the data is accumulated for several years by multiple satellites, and the data support cost is high. For another example, "BRDF prototype inversion algorithm verification of airborne WIDAS ground observation" (Remote Sensing, 2019, 23(4)) describes a method of using an airborne multispectral camera and a thermal infrared camera to observe the ground object from multiple angles to obtain BRDF characteristics, and the reflectivity data obtained by inversion is verified by MODIS data. However, due to the flight area limitation of the aircraft, it is difficult to effectively observe the ground object in the global environment and complex environment, and the data accumulation time is long, and the storage and processing cost is high.

[0048] With the improvement of imaging resolution and agile maneuverability of the new generation of optical remote sensing satellites, and the trend of satellite use from general survey to detailed survey, the observation task plan of the satellite has increased from the original 20 per day to 100, and the daily reconnaissance data volume has increased at least 12.5 times compared with the traditional satellite under the same imaging time, thereby generating a large amount of imaging reconnaissance data. On the one hand, it brings processing difficulties and increases data storage cost to users; on the other hand, the improvement of resolution increases the amount of multi-angle observation data, and the data processing efficiency is low, and the observation efficiency of the ground object target is low. For example, in order to obtain the BRDF characteristics of the ground object, at least 5 observation angles of the ground object spectrum need to be obtained, and under the same shooting time, the number of single orbit observation tasks that can be executed by the satellite will theoretically decrease by 5 times, which seriously affects the use of users and the efficiency of comprehensive coverage reconnaissance of satellites.

[0049] The method for determining BRDF model parameters provided by the application determines the relative azimuth angle threshold database of ground objects, determines the target angle between the sun main plane and the satellite orbit plane of the imaging satellite according to the imaging time when planning the observation task of the satellite, compares the target angle with the relative azimuth angle threshold database, indicates that the reflectivity variation of the target ground object is obvious and the BRDF characteristics can be effectively calculated in the case that the relative azimuth angle threshold database has a relative azimuth angle threshold greater than or equal to the target angle, generates a multi-angle observation task, makes the imaging satellite execute the multi-angle observation task, and fits the BRDF model parameters according to the data returned by the imaging satellite.

[0050] The application will be described in detail below with reference to the drawings and examples.

[0051] The application provides a method for determining BRDF model parameters, which will be described in detail below with reference to the examples and drawings.

[0052] In the embodiment of the application, as shown in Figure 1 The method for determining BRDF model parameters provided by the application comprises the following steps:

[0053] Step 11, receiving a satellite imaging request; wherein the satellite imaging request comprises an imaging scene, a target ground object corresponding to the imaging scene, ground object characteristics of the target ground object and an imaging time T0.

[0054] In the embodiment of the application, the method for generating an observation task of a satellite provided by the application is applied to a satellite on-board computer and is executed by the satellite on-board computer. The satellite on-board computer receives an imaging request and issues an observation task to an imaging satellite. The ground object characteristics comprise the ground object type (for example, the ground object type is vegetation), the LAI index of the ground object and the characteristic wave band of the ground object.

[0055] Further, the satellite imaging request can be sent by a user on the ground through a terminal, and the satellite on-board computer receives the imaging request sent by the terminal.

[0056] In the embodiment of the application, because the agile hyperspectral satellite has agile maneuvering capability, it can quickly acquire multi-angle data of ground objects, thereby rapidly accumulating BRDF characteristic data. Therefore, the imaging satellite is an agile hyperspectral satellite. The imaging scene can be a scene for observing the topography and geomorphology, such as sandy soil, grassland, dense single vegetation, forest and the like.

[0057] Step 12, determining a target angle between the sun main plane and the satellite orbit plane of the imaging satellite according to the imaging time T0

[0058] In the embodiment of the present application, the satellite mission computer determines the target angle between the main plane of the sun and the orbital plane of the satellite according to the imaging time T0 In the subsequent task planning process, the target angle can be used The target angle is compared with the relative azimuth angle threshold value in the database to determine whether the reflectivity variation coefficient with obvious variation can be obtained to generate different types of satellite observation tasks.

[0059] In the embodiment of the present application, before step 12, the satellite mission computer determines whether the imaging satellite can perform the observation task according to the current time T, the imaging time T0, the satellite attitude maneuver time τ attitude of the imaging satellite, and the camera imaging preparation time τ camera of the imaging satellite.

[0060] Further, the satellite mission computer determines whether T, T0, τ attitude satisfy T0-T>τ attitude , and whether T, T0, τ camera satisfy T0-T>τ camera . In the case of T0-T>τ attitude and T0-T>τ camera , step 12 is performed; otherwise, the satellite imaging request is rejected.

[0061] In the embodiment of the present application, as shown in Figure 2 , the method for determining the target angle of the present application includes the following steps:

[0062] Step 21, determining the solar zenith angle θ s and the inclination of the main plane of the sun compared to the equator according to the imaging time T0

[0063] In the embodiment of the present application, the satellite mission computer determines the solar zenith angle θ s according to the imaging time T0. The cosine value of the solar zenith angle θ s is shown in the following formula:

[0064] cosθ s =sin(sinB0sinδ+cosB0cosδcosφ),

[0065] In the above formula:

[0066] B0 is the latitude of the target ground object;

[0067] δ is the solar declination corresponding to the imaging time T0;

[0068] φ is the solar hour angle corresponding to the imaging time T0.​

[0069] Further, the satellite star computer calculates the solar zenith angle θ according to cosθ s . s .

[0070] In the embodiment of the present application, the satellite star computer determines the target angle between the solar principal plane and the satellite orbit plane of the imaging satellite according to the imaging time T0 and the satellite orbit inclination of the imaging satellite

[0071] In step 22, the target angle between the solar principal plane and the satellite orbit plane of the imaging satellite is calculated according to the satellite orbit inclination of the imaging satellite and the target angle .

[0072] In the embodiment of the present application, the satellite star computer calculates the target angle between the solar principal plane and the satellite orbit plane of the imaging satellite according to the satellite orbit inclination of the imaging satellite and the target angle between the solar principal plane and the equator . The target angle is calculated as follows:

[0073]

[0074] In the embodiment of the present application, the target angle between the solar principal plane and the satellite orbit plane of the imaging satellite can be determined according to the imaging time by the method for determining the target angle of the present application, so that the target angle is used as a matching reference for subsequent planning of the satellite observation task determination, thereby effectively calculating the BRDF characteristics.

[0075] In step 13, the pre-constructed relative azimuth angle threshold database is matched by using the ground object characteristics of the target ground object and the target angle to generate the observation task of the imaging satellite; wherein the type of the observation task includes a multi-angle observation task and a regular observation task.

[0076] In the embodiment of the present application, the relative azimuth angle threshold set corresponding to the ground object characteristics is determined according to the ground object characteristics of the target ground object, and the target angle is compared with the relative azimuth angle threshold set to generate the observation task of the corresponding type; wherein the type of the observation task includes a multi-angle observation task and a regular observation task.

[0077] In the embodiment of the present application, the relative azimuth angle threshold database of different ground objects can be constructed in a simulation or experimental manner; wherein the ground object of the present application is a vegetation type ground object. Taking the simulation manner as an example, as shown in the following table, the construction method of the relative azimuth angle threshold database of the present application includes the following steps: Figure 3

[0078] ​Step 31, set the solar zenith angle, observation zenith angle, characteristic wave band, LAI index and relative azimuth angle of different ground objects.

[0079] In the embodiment of the present application, the simulation input parameters of different ground objects are set, including the solar zenith angle, observation zenith angle, characteristic wave band, LAI index and relative azimuth angle. The solar zenith angle (i.e. Solar Zenith Angle) θ s refers to the angle between the light incidence direction and the zenith direction; the observation zenith angle θ v refers to the angle between the observation direction and the horizontal normal direction; the LAI index (i.e. leaf area index / leaf area coefficient) refers to the multiple of the total area of plant leaves on the unit land area to the land area. For example, the value range of the solar zenith angle θ s is set to 10°-40°, with a step of 10°; the value range of the observation zenith angle θ v is set to -50°-50°, with a step of 10°; the characteristic wave bands λ of solar radiation are respectively set to 450nm, 550nm, 650nm, 850nm, 1250nm, 1650nm and 2150nm; the value range of the LAI index is set to 1-4.

[0080] In the embodiment of the present application, the relative azimuth angle is the absolute value of the difference between the solar azimuth angle and the observation azimuth angle; wherein the solar azimuth angle (i.e. Solar Azimuth Angle) refers to the angle measured along the horizon clockwise from north of the sun, and the observation azimuth angle refers to the angle between the projection of the observation direction on the horizontal plane and the projection of the sun direction on the horizontal plane. For example, the value range of the relative azimuth angle is set to 0°-90°, with a step of 10°.

[0081] Step 32, input the solar zenith angle, observation zenith angle, characteristic wave band, LAI index and relative azimuth angle of different ground objects into the pre-trained Prosail model.

[0082] In the embodiment of the present application, Prosail radiation transfer model is used for simulation, and the Prosail model can be used to inverse the spectral reflectance of vegetation type ground objects, such as the spectral reflectance of leaves.

[0083] Step 33, determine the reflectance of different ground objects according to the output of the Prosail model.

[0084] In the embodiment of the present application, the reflectance of vegetation type ground objects is determined according to the output of the Prosail model.

[0085] Step 34, use the LAI index, relative azimuth angle and reflectance of different ground objects to construct a relative azimuth angle threshold database of different ground objects.

[0086] In the embodiments of the present application, according to the corresponding relationship of the LAI index, the relative azimuth angle and the reflectivity of different ground objects, the ground object data satisfying the requirement of the relative azimuth angle is screened, and a relative azimuth angle threshold database is constructed.

[0087] In step 341, the reflectivity variation coefficient of each type of ground object is determined by using the reflectivity.

[0088] In the embodiments of the present application, for the vegetation type ground object, the reflectivity variation coefficient of the ground object under different wave bands is calculated according to the reflectivity determined in step 33. Reflectivity variation coefficient For the set LAI index and solar zenith angle θ s , the reflectivity standard deviation of the ground object under different observation zenith angles θ v is divided by the reflectivity average value, as shown in the following formula:

[0089]

[0090] In the above formula:

[0091] represents the reflectivity variation coefficient of the ground object under the mth characteristic wave band;

[0092] For the set different solar zenith angles θ s , characteristic wave bands λ and LAI index, the reflectivity average value of the corresponding ground object under different observation zenith angles θ v

[0093] σ ρ is the standard deviation of the reflectivity. Wherein:

[0094] The calculation method of is shown in the following formula:

[0095]

[0096] In the above formula:

[0097] (θ v ) i is the different observation zenith angle, i=1, 2, …, n; wherein, n=1 represents the observation zenith angle θ v -50°, n=2 represents the observation zenith angle θ v -40°, …, n=11 represents the observation zenith angle θ v 50°;

[0098] is the ground reflectivity under the ith observation zenith angle θ v , the mth characteristic wave band λ;​

[0099] σ ρ The calculation method is as follows:

[0100]

[0101] Step 342, according to the preset variation threshold, the corresponding relative azimuth angle threshold is determined corresponding to the reflectivity variation coefficient.

[0102] In the embodiment of the application, according to the reflectivity variation coefficient obtained in step 341 and the preset variation threshold, when the reflectivity variation coefficient is equal to the preset variation threshold, the corresponding relative azimuth angle threshold is determined.

[0103] Further, the reflectivity variation coefficient equal to 10% is taken as the threshold of the effectiveness of BRDF characteristic measurement, and accordingly, the preset variation threshold is 10%, and the relative azimuth angle threshold when the reflectivity variation coefficient is equal to 10% is determined.

[0104] When the ground object is observed according to the relative azimuth angle whose reflectivity variation coefficient is greater than 10%, the ground object spectral characteristics with obvious variation can be obtained; when the ground object is observed according to the relative azimuth angle whose reflectivity variation coefficient is less than 10%, even if multi-angle observation is performed, effective spectral characteristic change cannot be obtained, and the effectiveness of BRDF calculation is not ideal. Therefore, in the subsequent observation process, the effective observation angle can be determined according to the relative angle threshold, so as to facilitate fitting and determining the model parameters of the BRDF model.

[0105] Step 343, the relative azimuth angle threshold database is generated.

[0106] In the embodiment of the application, according to the corresponding relationship among the ground object type, the solar zenith angle, the characteristic wave band, the LAI index, the reflectivity variation coefficient and the relative azimuth angle threshold, the relative azimuth angle threshold database is generated; wherein the ground object type is vegetation.

[0107] Further, according to the corresponding relationship among the ground object type, the solar zenith angle, the observation zenith angle, the characteristic wave band, the LAI index, the reflectivity variation coefficient and the relative azimuth angle threshold, the relative azimuth angle threshold database is generated.

[0108] Still further, the generated relative azimuth angle threshold database is imported into the satellite service computer, so as to assist the satellite service computer to perform observation task planning.

[0109] In this embodiment of the invention, the relative azimuth threshold database construction method of the present invention can obtain the reflectivity variation coefficient with significant variation through simulation or experimentation, thereby determining the relative azimuth threshold using a preset variation threshold and constructing a relative azimuth threshold database for ground objects, so as to facilitate subsequent satellite observation missions and effectively calculate BRDF characteristics.

[0110] In embodiments of the present invention, such as Figure 4 As shown, the method for determining the type of observation mission of the imaging satellite of the present invention includes the following steps:

[0111] Step 41, based on the solar zenith angle θ of the target object. s Using the LAI index and characteristic band λ, the relative azimuth threshold set of the target ground object is obtained by querying the relative azimuth threshold database.

[0112] Step 42, adjust the target angle. With the relative azimuth threshold set Compare and determine if it exists. If yes, proceed to step 43; if no, proceed to step 44.

[0113] Step 43: Generate the multi-angle observation task of the imaging satellite.

[0114] In this embodiment of the invention, when the relative azimuth threshold set of the target ground object... There exists an angle greater than or equal to the target angle. When the relative azimuth threshold is reached, it represents the coefficient of variation of the reflectance of the target ground object when observing it from multiple angles. A value greater than or equal to 10% indicates significant variation, allowing for effective calculation of BRDF characteristics. Therefore, based on the relative azimuth threshold... Generate multi-angle observation tasks from imaging satellites.

[0115] In this embodiment of the invention, the satellite mission computer uses a relative azimuth angle threshold. The plan is to plan a multi-angle observation mission for the imaging satellite, which includes multi-angle imaging at at least 5 attitude angles, meaning the imaging satellite needs to change its attitude angle at least 4 times for observation; among which, the attitude angles include the yaw angle and the pitch angle.

[0116] Step 44: Generate the routine observation tasks for the imaging satellite.

[0117] In this embodiment of the invention, when the relative azimuth threshold set of the target ground object... There is no angle greater than or equal to the target angle in the middle. a reflectivity variation coefficient of the target ground object when the target ground object is observed at multiple angles less than 10%, a conventional observation task of the imaging satellite is generated.

[0118] In the embodiment of the present application, the conventional observation task refers to that the imaging satellite only performs imaging once when passing the target ground object, and does not need to change the attitude angle multiple times for observation.

[0119] In the embodiment of the present application, by using the determination method of the observation task type of the imaging satellite of the present application, the relative azimuth angle threshold database constructed in advance can be matched by using the target included angle, the observation task type of the imaging satellite is determined, and the reflectivity variation coefficient of the target ground object with obvious variation is obtained in the case of successful matching, so that the BRDF characteristics of the target ground object can be effectively calculated.

[0120] In the embodiment of the present application, as shown in Figure 5 The execution method of the multi-angle observation task of the imaging satellite of the present application includes the following steps:

[0121] Step 51, according to the attitude angle indicated by the multi-angle observation task, the imaging satellite is controlled to perform attitude maneuvering.

[0122] In the embodiment of the present application, the shooting instruction of the multi-angle observation task can be shooting according to the imaging time and the attitude angle. After the imaging satellite receives the multi-angle observation task of the satellite service computer, attitude maneuvering is performed according to the specified attitude angle. The attitude maneuvering refers to that the imaging satellite can quickly adjust the pitch, yaw and other angles, so that the camera can quickly point to the target ground object, and the satellite can perform multi-angle observation during one transit.

[0123] Step 52, the camera of the imaging satellite is controlled to execute the shooting instruction at the specified time.

[0124] In the embodiment of the present application, the camera of the imaging satellite executes the shooting instruction at the specified time (such as imaging time T0) to perform multi-angle observation on the ground object.

[0125] In the embodiment of the present application, alternatively, after the imaging satellite receives the conventional observation task, the camera of the imaging satellite executes the shooting instruction at the specified time (such as imaging time T0) to perform conventional observation on the ground object.

[0126] Further, after the imaging satellite receives the observation task, the imaging satellite executes the imaging start-up instruction (such as the imaging circuit is powered on), and when the specified time arrives, the camera of the imaging satellite accurately executes the shooting instruction (such as the shooting is started).

[0127] In the embodiment of the present application, the imaging satellite can be controlled to perform attitude maneuver to execute the multi-angle observation task, so as to obtain the map data of the target ground object, thereby facilitating the subsequent effective calculation of the BRDF characteristics of the target ground object.

[0128] Step 14, in the case that the observation task is a multi-angle observation task, receiving the map data returned by the imaging satellite, and analyzing the map data to fit the BRDF model parameters of the target ground object.

[0129] In the embodiment of the present application, the map data (also referred to as hyperspectral data or data cube) of the target ground object returned by the imaging satellite includes reflectivity of the target ground object at different characteristic wavebands. The reflectivity at different characteristic wavebands is input into the BRDF model to fit the BRDF model parameters of the target ground object. For example, the BRDF model is a kernel-driven model, and the model function is as follows:

[0130]

[0131] In the above formula,

[0132] ρ is the reflectivity of the target ground object in the hyperspectral data, which can be determined according to the map data of the target ground object returned by the imaging satellite; θ s , θ v , are the solar zenith angle, the observation zenith angle and the relative azimuth angle, respectively; k is related to the LAI index of the vegetation and represents the ground cover type (which can be vegetation or non-vegetation), and the target ground object of the present application is vegetation; λ is the characteristic waveband; K vol , K geo are the Ross-Thick kernel and the Li-Sparse kernel, respectively; f iso (k, λ), f vol (k, λ) and f geo (k, λ) are BRDF model parameters to be determined, which respectively represent the proportions of isotropic scattering, volume scattering and geometric optical scattering.

[0133] In fitting the BRDF model parameters of the target ground object, ρ can be determined according to the map data of the target ground object returned by the imaging satellite, θ s , θ v , k and λ are known quantities in the imaging process of the imaging satellite, K vol , K geo are fixed kernel functions, and the BRDF model parameters f iso (k, λ), f vol (k, λ) and fgeo (k, λ).

[0134] Further, in the subsequent use process, the BRDF model parameters f iso (k, λ), f vol (k, λ), f geo (k, λ) of the target ground object fitted can be determined, and the reflectivity of the ground object under the corresponding vegetation type can be calculated.

[0135] Figure 6 Fig. 1 is a schematic diagram of the main modules of the BRDF model parameter determination system according to an embodiment of the present application, as shown in the figure, the BRDF model parameter determination system 60 of the present application comprises: Figure 6

[0136] a receiving module 61, configured to receive a satellite imaging request; wherein the satellite imaging request comprises an imaging scene, a target ground object corresponding to the imaging scene, ground object characteristics of the target ground object, and an imaging time T0.

[0137] a data processing module 62, configured to determine a target angle between a solar principal plane and a satellite orbit plane of the imaging satellite according to the imaging time T0.

[0138] a matching module 63, configured to match a pre-constructed relative azimuth angle threshold database with the ground object characteristics of the target ground object and the target angle to generate an observation task of the imaging satellite; wherein the type of the observation task comprises a multi-angle observation task and a regular observation task.

[0139] a fitting module 64, configured to, in the case that the observation task is a multi-angle observation task, receive atlas data returned by the imaging satellite, analyze the atlas data, and fit BRDF model parameters of the target ground object.

[0140] In the embodiments of the present application, through the receiving module, the data processing module, the matching module, and the fitting module, etc., the reflectivity that is greatly different under different observation angles can be obtained when the ground object is observed, so that the fitting can be efficiently and accurately performed to obtain a BRDF model with good performance, various ground object elements can be efficiently identified and observed, and the ground object classification precision is improved and the target detection and identification false alarm rate is reduced.

[0141] To sum up, the above is only a preferred embodiment of the present application, and is not used to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.​

Claims

1. A method for determining BRDF model parameters, characterized in that, include: Receive a satellite imaging request; wherein the satellite imaging request includes an imaging scene, target features corresponding to the imaging scene, features characteristics of the target features, and imaging time. ; According to the imaging time Determine the target angle between the principal plane of the sun and the orbital plane of the imaging satellite. ; Utilizing the terrain features of the target and the included angle of the target The observation tasks of the imaging satellite are generated by matching a pre-built database of relative azimuth angle thresholds; wherein the types of observation tasks include multi-angle observation tasks and regular observation tasks. In the case of a multi-angle observation mission, the imaging satellite returns map data, the map data is analyzed, and the BRDF model parameters of the target ground object are fitted. The method utilizes the terrain features of the target landform and the target angle. Matching a pre-built database of relative azimuth thresholds to generate observation tasks for the imaging satellite includes: Based on the solar zenith angle of the target location Using the LAI index and characteristic bands, the relative azimuth threshold set of the target ground object is obtained by querying the relative azimuth threshold database. ; The target angle With the relative azimuth threshold set Compare and determine if it exists. ; If so, generate a multi-angle observation task for the imaging satellite; The method for constructing the relative azimuth threshold database includes: Set the solar zenith angle, observation zenith angle, characteristic band, LAI index, and relative azimuth angle for different ground features; The solar zenith angle, observed zenith angle, characteristic band, LAI index, and relative azimuth angle of different ground features are input into the pre-trained Prosail model; Based on the output of the Prosail model, the reflectivity of different ground features is determined; A database of relative azimuth thresholds for different land features is constructed using the LAI index, relative azimuth angle, and reflectivity of the different land features. According to the imaging time Determine the target angle between the solar principal plane and the orbital plane of the imaging satellite. Previously, it also included: According to the current time The imaging time The satellite attitude maneuvering time of the imaging satellite The camera imaging preparation time of the imaging satellite is... ,judge , , Does it meet the requirements? ,and, , , Does it meet the requirements? ; exist and In this case, it is determined that the imaging satellite can perform observation tasks; According to the imaging time Determine the target angle between the solar principal plane and the orbital plane of the imaging satellite. ,include: According to the imaging time Determine the solar zenith angle Inclination of the solar principal plane relative to the equator ; Based on the satellite orbital inclination of the imaging satellite and the tilt angle Calculate the target angle between the solar principal plane and the satellite orbital plane. .

2. The method as described in claim 1, characterized in that, The step of constructing a relative azimuth threshold database for different land features using their LAI indices, relative azimuth angles, and reflectance includes: For each of the aforementioned land features, the reflectance variation coefficient of the land feature is determined using reflectance. Based on the preset variation threshold, determine the relative azimuth angle threshold corresponding to the reflectivity variation coefficient; Generate the relative azimuth threshold database.

3. The method as described in claim 1, characterized in that, The imaging satellite is an agile hyperspectral satellite.

Images