Low-orbit moonlet constellation optical load global geometric cross calibration field construction method

Abstract

The invention provides a low-orbit moonlet constellation optical load global geometric cross calibration field construction method, which comprises the steps of performing constellation global coverage condition analysis on a low-orbit moonlet constellation, selecting a daily key observation area, and screening a geometric cross calibration field from a target area according to a geometric calibration field construction principle; the standard reference data of geometric cross calibration is derived from geometrically calibrated image products of satellites with the same resolution or approximate resolution, geometrically calibrated products of constellation satellites, disclosed image data and elevation data, and the geometric cross calibration field database is constructed through the method. According to the method, the global geometric cross calibration field of the optical load of the low-orbit moonlet constellation is constructed, rapid, high-frequency and low-cost geometric calibration is carried out on the single satellite of the low-orbit constellation, satellite group geometric calibration transmission is rapidly carried out, and the method has good significance in rapid geometric calibration of the low-orbit constellation.

Description

Technical Field

[0001] The invention relates to a method for constructing a global geometric cross calibration field for optical payloads of a low-orbit small satellite constellation. Background Art

[0002] The accuracy of the internal and external orientation elements of optical remote sensing satellites will be affected by various factors during their orbit. Traditional geometric calibration is performed by shooting fixed ground calibration fields, which has some disadvantages:

[0003] (1) It is necessary to shoot fixed ground calibration sites, which requires advance mission planning and is costly;

[0004] (2) Greatly affected by weather and seasons;

[0005] (3) It is impossible to perform high-frequency calibration in a short period of time.

[0006] In view of the shortcomings of traditional geometric calibration, the use of geometric cross calibration can better solve these problems. Geometric cross calibration does not use the existing ground calibration field, but uses the high-precision image data obtained by the existing payload of the same type or the calibrated satellite as a reference for calibration. It has the advantages of large data volume, real-time, high-frequency geometric calibration, no need for field measurement, reducing calibration costs and improving calibration timeliness.

[0007] With the rapid development of low-orbit small satellite platforms, the demand for multi-satellite rapid calibration of constellations has become more urgent. Traditional single-satellite calibration technology can no longer meet the rapid calibration needs of constellations. In order to meet the needs of rapid geometric calibration, cross calibration for low-orbit constellations is an effective way to solve the problem. The key to realizing this technology is to construct a global geometric cross calibration field for low-orbit small satellite constellations. Summary of the invention

[0008] The purpose of the present invention is to solve the defects of the existing geometric calibration technology, that is, the high cost and the calibration frequency cannot meet the actual needs, and to provide a method for constructing a global geometric cross-calibration field for the optical payload of a low-orbit small satellite constellation to solve the above problems.

[0009] The objective of the present invention is achieved through the following technical solutions:

[0010] A method for constructing a global geometric cross calibration field for optical payloads of a low-orbit small satellite constellation comprises the following steps:

[0011] (1) Conduct access coverage analysis on low-orbit small satellite constellations;

[0012] (2) Select daily key targets: Select key target areas as the calibration areas to be selected within the area covered by the low-orbit small satellite constellation;

[0013] (3) Screening the geometric calibration field: Screening a suitable geometric calibration field from the calibration field area to be selected. The screening principle refers to the construction principle of the geometric calibration field.

[0014] (4) Obtaining geometric cross-calibration field benchmark data, which includes image products of satellites with equal or similar resolution that have been geometrically calibrated, publicly available image data, and elevation data;

[0015] (5) Establishing a geometric cross calibration field database, which includes geometric calibration field information and benchmark data, and completing the construction of the global geometric cross calibration field of the optical payload of the low-orbit small satellite constellation.

[0016] Furthermore, the low-orbit small satellite constellation access coverage analysis is specifically as follows:

[0017] Analyze the global coverage of the constellation based on the satellite orbit conditions of the low-orbit small satellite constellation to determine the global range that the constellation orbit can cover on a daily basis.

[0018] Furthermore, the small satellite constellation adopts a low-orbit, low-inclination orbit, so that the orbital plane precesses in the opposite direction to the Earth's revolution.

[0019] Furthermore, the selection of daily key targets is specifically as follows:

[0020] According to the coverage analysis results, daily observation point targets within the coverage area are extracted and included in the candidate geometric cross calibration field as selection factors based on the importance of the target, the target revisit cycle, and the number of imaging times.

[0021] Furthermore, when screening the geometric calibration site, the geometric cross calibration site is selected from the candidate sites in step 2, and the factors considered in the selection include: cloud cover, visibility, terrain, and landform type of the site.

[0022] Furthermore, the cloud cover at the site is low cloud cover all year round, that is, the annual traffic volume is lower than its required threshold;

[0023] The visibility and terrain are flat areas with high visibility, that is, the visibility is higher than the required threshold and the address flatness meets the requirements;

[0024] There are a large number of features that can be used as control points in the site, including intersections, pedestrian crossings, playgrounds, and flower beds. The changes in the features types within a certain period of time are small and do not meet the required threshold.

[0025] Furthermore, when obtaining the geometric cross-calibration field benchmark data, the geometrically calibrated image products of satellites with equal or similar resolution are selected. The public image data and elevation data refer to the online public image data set, the global network digital map, and the public elevation data set. At the same time, the metadata information of these benchmark data is obtained, including satellite, payload, imaging angle, longitude and latitude range, imaging resolution, and orbital altitude.

[0026] Furthermore, a geometric cross calibration field database is established, including:

[0027] The database contains geometric calibration field information and benchmark data, and the geometric cross calibration field information and benchmark data are established corresponding to their respective database tables;

[0028] All kinds of data are stored in the database;

[0029] Construct data filtering and retrieval rules.

[0030] Furthermore, data screening and retrieval are achieved using NLTK text description information search.

[0031] The beneficial effects of the present invention compared with the prior art are:

[0032] The global geometric cross-calibration field construction of the optical payload of the low-orbit small satellite constellation proposed in the present invention is to conduct coverage analysis on the low-orbit small satellite constellation, select daily key observation targets within the coverage range according to the coverage analysis results, screen suitable geometric calibration fields from the target areas to be selected with reference to the geometric calibration field construction principles, and select basic data according to the calibration fields. The data comes from image products with the same resolution, products of constellation satellite geometric calibration, and public image data. The database contains various information of the geometric calibration field, various information of the reference data, and various operations related to the data such as addition, deletion, modification, and query. By constructing a global geometric cross-calibration network for the constellation for geometric calibration, compared with traditional geometric calibration, it is fast and efficient, solves costs, increases the calibration frequency of satellites to better improve the satellite calibration accuracy, and quickly transmits the star cluster, effectively improving the geometric positioning accuracy of the constellation. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0034] The specific embodiments of the present invention are further described in detail below with reference to the accompanying drawings.

[0035] The present invention proposes a method for constructing a global geometric cross calibration field for optical payloads of low-orbit small satellite constellations. By performing coverage analysis on the low-orbit constellation, the key focus areas, normalized target areas, high-frequency revisit areas with good weather conditions, terrain conditions conducive to calibration, and rich ground features are selected as cross calibration fields; the geometric cross benchmark data selects image products with the same resolution that have been geometrically calibrated, the impact of constellation satellites that have been calibrated, public impact data and elevation data, etc., and constructs a calibration field database specifically for low-orbit small satellite optical payloads to perform data storage, retrieval, screening and other operations. By constructing a global geometric cross calibration field for optical payloads of low-orbit small satellites, the advantages of the low-orbit constellation are fully utilized to meet the calibration requirements of fast, efficient, real-time and low-cost.

[0036] like Figure 1 As shown, the present invention proposes a method for constructing a global geometric cross calibration field for optical payloads of a low-orbit small satellite constellation, and the main steps are as follows:

[0037] (1) Conducting coverage analysis of low-orbit small satellite constellation access

[0038] The global coverage of the constellation is analyzed based on the satellite orbit of the low-orbit small satellite constellation to determine the global range that the constellation orbit can cover on a daily basis. In order to achieve the shortest possible revisit period of key areas, the constellation should adopt a low-orbit, low-inclination orbit so that the orbital plane precession is opposite to the Earth's revolution.

[0039] The present invention takes an orbital altitude of 500km, an orbital inclination of 35°, and a constellation configuration composed of 10 orbital planes as an example. Since the precession of the orbital plane of the low-inclination orbit is opposite to the direction of the earth's revolution, the westward precession of the orbital plane and the eastward rotation of the sun are superimposed, which can achieve complete coverage of a specific latitude band relatively quickly, and the time of visiting the same area is constantly shifted forward, with faster revisit and time traversal characteristics. The constellation is covered and the revisit characteristics of the orbit to the target area are analyzed. The distribution of the number of visits per day to 19 point targets uniformly distributed in latitude on the 26° longitude line is simulated, and the optical payload revisit period is shown in Table 1.

[0040] Table 1 Optical payload revisit cycle simulation

[0041]

[0042]

[0043] According to the above revisit analysis, it can be found that the constellation revisits the most targets between 30°-35° north and south latitudes, with an average revisit interval of less than 8-9 minutes and a maximum revisit time of about 20 minutes. For targets between 25°-30° north and south latitudes, the constellation revisits 120 times a day, with an average revisit interval of less than 13 minutes, and the revisit time interval meets the requirements.

[0044] (2) Selection of daily key targets

[0045] Within the area that the low-orbit constellation can cover, select key target areas as candidate calibration areas and include them in the candidate geometric cross calibration areas. Factors to be considered include: the constellation's revisit period and observation frequency of the target point, whether the target point is a normalized key observation area, etc. For the low-orbit constellation simulated in step (1), with a 15-minute revisit requirement as a condition, all target points meet the minimum revisit period of <15 minutes, 53% of the average revisit period meets the requirement, and the number of daily optical imaging times meets >60 times / day;

[0046] (3) Screening of geometric cross calibration fields

[0047] The geometric cross calibration field is selected from the key target library in step (2). The selection principle of the geometric cross calibration field refers to the construction principle of the geometric cross calibration field: it is necessary to select a flat area with low cloud cover and high visibility all year round. There are a large number of objects that can be used as control points in the field, such as intersections, crosswalks, playgrounds, flower beds, etc., and the objects change little in a certain period of time. Combined with the low-orbit inclination constellation access coverage analysis results for typical target areas, the typical area location of the global geometric cross calibration field is determined.

[0048] (4) Acquisition of Geometric Cross-calibration Field Benchmark Data

[0049] The benchmark data is the geometrically calibrated image products of satellites with the same or similar resolutions, and public data such as online public image datasets, global network digital maps, and public elevation datasets. The data should meet the following conditions: the cross-coverage range with the calibration field is large enough, the cloud cover does not cover the key areas, and the shooting angle is small. At the same time, the metadata information of these benchmark data should be obtained, including satellite, payload, resolution, orbit altitude, orbit inclination, imaging side swing angle, imaging pitch angle, satellite zenith angle, satellite azimuth angle, geometric positioning accuracy, etc.

[0050] When geometrically calibrating a certain scene image, it is necessary to select benchmark data from the basic database. The following conditions should be met when selecting benchmark data:

[0051] ① Cloud cover: meets the cloud-free condition, or the cloud-covered area is not in the calibration area;

[0052] ②Angle: Smaller side swing angle (-10°~10°);

[0053] ③ Coverage: Completely cover the calibration field, or there are enough control points in the coverage area;

[0054] ④ Time: within 1 month from the time of shooting the target scene to be determined;

[0055] ⑤ Lighting conditions: avoid selecting image control points that are overexposed or too dark;

[0056] ⑥ Resolution: as close as possible to the calibration scene;

[0057] (5) Construction of global geometric cross-calibration field database

[0058] a. Geometric cross calibration field database design

[0059] The geometric cross calibration field database table must include the following elements:

[0060] Table 3 Geometric cross calibration field database table

[0061] Serial number Table fields Support Chinese type Fill out the requirements Remark 1 FieldName Venue Name string Required —— 2 Longitude longitude double Required Spend 3 Latitude latitude double Required Spend 4 Height Elevation double Required rice 5 RevisitPeriod Revisit cycle double Required minute 6 LandCoverType Feature Type string Required —— 7 UpdateTime Update time string Required Time string 8 Weather Daily weather conditions string Required ——

[0062] b. Basic database design

[0063] The collected imaging data from satellites with the same resolution or calibrated satellites must include information such as imaging satellite, payload, resolution, orbital altitude, orbital inclination, imaging side swing angle, imaging pitch angle, satellite zenith angle, satellite azimuth angle, geometric positioning accuracy, image data and corresponding auxiliary data, and be included in the benchmark database.

[0064] Table 3 Basic reference database table

[0065]

[0066]

[0067] c. Data retrieval and screening

[0068] With long-term accumulation, a large amount of data from different star sources, different times, and different imaging conditions will be stored in the global geometric cross-calibration field. In order to realize the rapid inter-satellite transmission of geometric calibration, the generated data set is effectively managed, the call request analysis is realized, the file content is parsed as needed, the corresponding data items in the call file are extracted and the call is returned, and the optimal data set is obtained from the existing data set. It supports screening queries from multiple dimensions such as time range, spatial range, imaging conditions, spectral range, etc., to improve retrieval efficiency.

[0069] NLTK text description information search: Use NLTK to analyze the text description information containing search intent entered by the user, extract the keywords and the position relationship between the keywords. Use T to represent the keyword information obtained by NLP: T τ →Y τ ={y1,y2,...,y m},y i ∈{0,1} m If T τ Contains tag c i ,y τ (i)=1otherwise y τ (i) = 0. Taking the upper and lower relationship of an image as an example, this algorithm constructs its keyword upper position association matrix The sub-element Indicates the relationship that the text description contains keyword i above keyword j:

[0070]

[0071] Similarly, we get the relationship matrix below the keywords And the relationship matrix next to it

[0072]

[0073]

[0074] Analyze text query information through NLP technology to obtain keyword set T τ Semantic relationship matrix between keywords These query conditions are matched in the "seed" image set to obtain similar images that meet the conditions as "seed" images. The similarity calculation model used here is the Euclidean distance in the space vector model. The text keyword set T τ and the “seed” image I k Similarity of tag sets:

[0075]

[0076] Text Semantic Correlation Matrix and the “seed” image semantic association matrix The similarity is defined as follows:

[0077]

[0078] Text Semantic Correlation Matrix and the “seed” image semantic association matrix The similarity is defined as follows:

[0079]

[0080] Text Semantic Correlation Matrix and the “seed” image semantic association matrix The similarity is defined as follows:

[0081]

[0082] In summary, the match between the query text and the “seed” image is defined as:

[0083] sim(τ,k)=a×sim1(τ,k)+(1-α)[sim2(τ,k)+sim3(τ,k)+sim4(τ,k)]

[0084] Among them, α is a parameter that determines the weight of the text keyword matching degree and the tag semantic matching degree in the comprehensive matching degree. According to later experiments, α is generally 0.5. Finally, the previous data with the largest matching degree is selected as the result of the initial search.

[0085] The technical solution of the present invention is aimed at the requirements for constructing a global geometric cross-calibration field for optical payloads of low-orbit small satellite constellations. It explains the access coverage analysis of low-orbit small satellite constellations, the selection of daily key targets, the screening of geometric calibration fields, the acquisition of geometric cross-calibration field benchmark data, and the establishment of a geometric cross-calibration field database. It provides a theoretical and practical basis for the construction of a global geometric cross-calibration field for optical payloads of low-orbit small satellite constellations.

[0086] The contents not described in detail in the specification of the present invention belong to the common knowledge of those skilled in the art.

Claims

1. A method for constructing a global geometric cross-calibration field for optical payloads of a low-orbit small satellite constellation, characterized in that: include: Conduct access coverage analysis of low-orbit small satellite constellations; Select daily key targets: within the area covered by the low-orbit small satellite constellation, select key target areas as the calibration field areas to be selected; Screening geometric calibration field: Screening a suitable geometric calibration field from the calibration field area to be selected. The screening principle refers to the construction principle of the geometric calibration field. Obtain geometric cross-calibration field benchmark data, which includes image products of satellites with equal or similar resolution that have been geometrically calibrated, public image data, and elevation data; A geometric cross calibration field database is established, which includes geometric calibration field information and reference data, and completes the construction of the global geometric cross calibration field of the optical payload of the low-orbit small satellite constellation.

2. The method for constructing a global geometric cross calibration field for optical payloads of a low-orbit small satellite constellation according to claim 1 is characterized in that: The low-orbit small satellite constellation access coverage analysis is specifically as follows: Analyze the global coverage of the constellation based on the satellite orbit conditions of the low-orbit small satellite constellation to determine the global range that the constellation orbit can cover on a daily basis.

3. The method for constructing a global geometric cross calibration field for optical payloads of a low-orbit small satellite constellation according to claim 2 is characterized in that: The small satellite constellation uses a low-orbit, low-inclination orbit, so that the orbital plane precesses in the opposite direction to the Earth's revolution.

4. The method for constructing a global geometric cross calibration field for optical payloads of a low-orbit small satellite constellation according to claim 1, characterized in that: The selection of daily key targets is specifically as follows: According to the coverage analysis results, daily observation point targets within the coverage area are extracted, and the importance of the targets, the revisit period of the targets, and the number of imaging times are used as selection factors to be included in the selected geometric cross calibration field.

5. The method for constructing a global geometric cross calibration field for optical payloads of a low-orbit small satellite constellation according to claim 1, characterized in that: When screening the geometric calibration site, select the geometric cross calibration site from the candidate sites in step 2. The factors to be considered during the selection include: cloud cover, visibility, terrain, and landform type of the site.

6. The method for constructing a global geometric cross calibration field for optical payloads of a low-orbit small satellite constellation according to claim 5, characterized in that: The cloud cover at the site is low cloud cover all year round, that is, the annual traffic volume is lower than its required threshold; The visibility and terrain are flat areas with high visibility, that is, the visibility is higher than the required threshold and the address flatness meets the requirements; There are a large number of features that can be used as control points in the site, including intersections, pedestrian crossings, playgrounds, and flower beds. The changes in the features types within a certain period of time are small and do not meet the required threshold.

7. The method for constructing a global geometric cross calibration field for optical payloads of a low-orbit small satellite constellation according to claim 1, characterized in that: When obtaining the geometric cross-calibration field benchmark data, select the geometrically calibrated image products of satellites with equal or similar resolution. The public image data and elevation data refer to the online public image data set, the global network digital map, and the public elevation data set. At the same time, obtain the metadata information of these benchmark data, including satellite, payload, imaging angle, longitude and latitude range, imaging resolution, and orbit altitude.

8. The method for constructing a global geometric cross calibration field for optical payloads of a low-orbit small satellite constellation according to claim 1, characterized in that: Establish a geometric cross calibration field database, including: The database contains geometric calibration field information and benchmark data, and the geometric cross calibration field information and benchmark data are established corresponding to their respective database tables; All kinds of data are stored in the database; Construct data filtering and retrieval rules.

9. The method for constructing a global geometric cross calibration field for optical payloads of a low-orbit small satellite constellation according to claim 8, characterized in that: Data screening and retrieval are achieved using NLTK text description information search.

Images