A remote sensing satellite system

Abstract

A remote sensing satellite system comprises a first satellite (100), the first satellite (100) comprises at least four image sensors (131, 132, 133, 134), the at least four image sensors (131, 132, 133, 134) simultaneously collect images of the ground, the ground areas collected by the at least four image sensors (131, 132, 133, 134) completely overlap or partially overlap, the spatial resolution and spectral resolution of the images collected by the at least four image sensors (131, 132, 133, 134) are different from each other, and the first satellite (100) performs image fusion on at least a part of the images collected by the at least four image sensors (131, 132, 133, 134) to generate a fused remote sensing image. The remote sensing satellite system can efficiently obtain high-definition remote sensing images using limited satellite resources.

Description

Technical Field

[0001] The present invention relates to the field of remote sensing technology, and in particular to a remote sensing satellite system. Background Art

[0002] Remote sensing technology refers to a modern comprehensive technology that receives electromagnetic wave information from various land objects on the earth's surface from high altitude or outer space, and scans, photographs, transmits and processes this information to remotely control and identify various land objects and phenomena on the earth's surface.

[0003] Hyperspectral remote sensing is a new remote sensing technology developed in the 1980s. This technology uses satellite-borne or airborne imaging spectrometers to image the ground. While imaging the spatial features of the target area, the imaging spectrometer disperses each spatial pixel to form dozens or even hundreds of narrow bands for continuous spectral coverage, thereby forming remote sensing data with a spectral resolution of the order of nanometers. Due to its high spectral resolution, this data is usually called hyperspectral data or hyperspectral images. The spectral resolution of hyperspectral data is about 10 nanometers, which is dozens or even hundreds of times higher than that of multispectral images. With the continuous development of imaging spectroscopy technology, hyperspectral data has been applied to many fields. From environmental monitoring, urban planning, crop yield estimation, flood disaster investigation, and land and resources investigation, which are widely used in the civilian field, to satellite reconnaissance, target detection and recognition in the military field, etc.

[0004] The outstanding feature of hyperspectral images is that while obtaining the two-dimensional spatial scene information of the target image, it is also possible to obtain high-resolution one-dimensional spectral information representing its physical properties, that is, the integration of image and spectrum. By processing the spatial and spectral features of the target image in the hyperspectral image, the ground objects can be identified and distinguished with a high degree of credibility. This has important application significance and great potential for remote sensing image military reconnaissance, true / false target identification, and refined classification of agriculture and forestry. The two main development trends of remote sensing technology for a long time have been to develop in the direction of high spatial resolution and high spectral resolution, but the development of the two is often contradictory and mutually restrictive, which is mainly due to the limitations of the imaging optical system in design and implementation. The spectral resolution of hyperspectral images is generally high, but its spatial resolution is low, which is not conducive to target recognition algorithms.

[0005] Spatial resolution is simply a measure of the imaging system's ability to resolve image details. It is also an indicator of the subtlety of the target in the image. It indicates the level of detail of the scene information. It is one of the important indicators for evaluating sensor performance and remote sensing information, and is also an important basis for identifying the shape and size of ground objects. The spatial resolution of remote sensing images is directly related to the imaging optical system. If the resolution is low, there will be more mixed pixels in the remote sensing image, which will seriously affect the analysis and understanding of the image, which is very unfavorable for target classification, detection and recognition.

[0006] Spectral resolution refers to the degree of precision of the sensor's discrete sampling of the spectrum of ground objects within a certain wavelength range. Spectral resolution is the main indicator that characterizes the performance of sensors in obtaining spectral information of ground objects. Compared with spatial image information, as another way to characterize the characteristics of ground objects, spectral information obtained through remote detection can also realize the identification of ground objects, and spectral information is directly related to the material composition of the target. In particular, it is more suitable for target identification, fine classification of vegetation, quantitative monitoring of ocean water color, and military identification of camouflage from a spectral perspective than spatial images.

[0007] Image fusion is to process images with different spatial and spectral resolutions according to a specific algorithm, so that the generated new image has both the multispectral characteristics of the original image and high spatial resolution information. In multispectral remote sensing image fusion, typical image fusion methods include: fusion method based on IHS transform, fusion method based on the combination of IHS transform and wavelet transform, and fusion method based on the combination of HSV transform and wavelet transform.

[0008] At present, panchromatic images and multispectral images are usually fused. For example, a Chinese patent document with publication number CN108230281A discloses a remote sensing image processing method, and a specific implementation of the method includes: matching the features of the panchromatic image with the features of the multispectral image to obtain multiple feature pairs; based on the feature pairs, determining the inter-image mapping matrix; according to the inter-image mapping matrix, determining the overlapping area of ​​the panchromatic image and the multispectral image; fusing the overlapping area of ​​the panchromatic image with the overlapping area of ​​the multispectral image to obtain a fused remote sensing image. This implementation can process a wider range of remote sensing images, avoid the loss of image accuracy caused by bit depth conversion, and improve the accuracy of the fused image. However, with the rapid development of remote sensing technology today, we have higher and higher requirements for the resolution of remote sensing images, but the existing imaging equipment is far from meeting the requirements in all aspects. Therefore, it is necessary to improve the existing technology. Summary of the invention

[0009] In view of the deficiencies in the prior art, the present invention provides a remote sensing satellite system. The present invention can fuse images with different spatial resolutions and different spectral resolutions collected by the same satellite to obtain a fused remote sensing image. The present invention can combine data with multiple different characteristics, learn from each other's strengths and weaknesses, give full play to their respective advantages, make up for their respective deficiencies, and more comprehensively reflect ground targets, so as to efficiently obtain high-definition remote sensing images by utilizing the limited resources of satellites.

[0010] Advantageously, a remote sensing satellite system comprises a first satellite, the first satellite comprising at least four image sensors, the at least four image sensors simultaneously capturing images of the ground, the ground areas captured by the at least four image sensors completely overlapping or partially overlapping, the spatial resolutions and spectral resolutions of the images captured by the at least four image sensors being different from each other, and the first satellite performing image fusion on at least a portion of the images captured by the at least four image sensors to generate a fused remote sensing image.

[0011] Advantageously, the at least four image sensors include a first image sensor, a second image sensor, a third image sensor and a fourth image sensor; wherein the first image sensor has a first spatial resolution and a first spectral resolution, the second image sensor has a second spatial resolution and a second spectral resolution, the third image sensor has a third spatial resolution and a third spectral resolution, the fourth image sensor has a fourth spatial resolution and a fourth spectral resolution, the second spatial resolution is lower than the first spatial resolution, the second spectral resolution is higher than the first spectral resolution, the third spatial resolution is lower than the second spatial resolution, the third spectral resolution is higher than the second spectral resolution, the fourth spatial resolution is lower than the third spatial resolution, the fourth spectral resolution is higher than the third spectral resolution, the first image sensor can be used to capture a first image, the second image sensor can be used to capture a second image, the third image sensor can be used to capture a third image, and the fourth image sensor can be used to capture a fourth image.

[0012] Advantageously, the images simultaneously acquired by the at least four image sensors have a common overlapping area, and the first satellite fuses the common overlapping areas of every two images among the images simultaneously acquired by the at least four image sensors to form a number of first-category fused images, and then the first satellite fuses every two images among the number of first-category fused images to form a number of second-category fused images, and the first satellite uses at least one of the number of second-category fused images as a fused remote sensing image.

[0013] Advantageously, the first image is of panchromatic image type, the second image is of multispectral image type, the third image is of hyperspectral image type and the fourth image is of ultra-hyperspectral image type.

[0014] Advantageously, the first satellite comprises a landmark identification module and an error correction module, wherein the landmark identification module is configured to obtain landmark information associated with each image acquired by the at least four image sensors, and the error correction module is configured to calculate a state vector for correcting an orbit error and an attitude error of the first satellite associated with each image acquired by the at least four image sensors based on the landmark information.

[0015] Advantageously, the landmark recognition module is configured to: select at least three landmarks from each image captured by the at least four image sensors, determine the remote sensing landmark positions of the at least three landmarks in each image captured by the at least four image sensors and the actual landmark positions on the earth, calculate the difference between the corresponding remote sensing landmark positions and the actual landmark positions, and obtain landmark information based on the difference between the corresponding remote sensing landmark positions and the actual landmark positions.

[0016] Advantageously, the landmark recognition module is configured to: before selecting at least three landmarks from each image acquired by the at least four image sensors, identify the number of landmarks in each image acquired by the at least four image sensors; when the number of identifiable landmarks in each image acquired by the at least four image sensors is greater than or equal to three, select at least three landmarks from each image acquired by the at least four image sensors, determine the remote sensing landmark positions of the at least three landmarks in each image acquired by the at least four image sensors and the actual landmark positions on the earth, calculate the difference between the corresponding remote sensing landmark positions and the actual landmark positions, and obtain the corresponding remote sensing landmark positions based on the difference between the actual landmark positions. Landmark information; when the number of identifiable landmarks in each image acquired by at least four image sensors is less than three, select a landmark with directional directivity from each image acquired by at least four image sensors, determine the remote sensing landmark position and orientation of the landmark with directional directivity in each image acquired by at least four image sensors and the actual landmark position and orientation on the earth, calculate the difference between the corresponding remote sensing landmark position and orientation and the actual landmark position and orientation, and obtain landmark information based on the difference between the corresponding remote sensing landmark position and orientation and the actual landmark position and orientation, wherein the landmark with directional directivity is at least one of a river, an airport runway, a road and a coastline.

[0017] Advantageously, the first satellite further comprises: a resampling module, the resampling module being configured to resample the pixel position of each image acquired by the at least four image sensors based on the calculated state vector; and only after the pixel position of each image acquired by the at least four image sensors is resampled based on the calculated state vector, image fusion is performed on at least a portion of the images acquired by the at least four image sensors to generate a fused remote sensing image.

[0018] Advantageously, a method for generating a remote sensing image comprises: acquiring images of the ground simultaneously collected from at least four image sensors of a first satellite, performing image fusion on at least a portion of the images collected by the at least four image sensors to generate a fused remote sensing image, wherein the ground areas collected by the at least four image sensors completely overlap or partially overlap, and the spatial resolutions and spectral resolutions of the images collected by the at least four image sensors are different from each other.

[0019] Advantageously, the at least four image sensors include a first image sensor, a second image sensor, a third image sensor and a fourth image sensor; wherein the first image sensor has a first spatial resolution and a first spectral resolution, the second image sensor has a second spatial resolution and a second spectral resolution, the third image sensor has a third spatial resolution and a third spectral resolution, the fourth image sensor has a fourth spatial resolution and a fourth spectral resolution, the second spatial resolution is lower than the first spatial resolution, the second spectral resolution is higher than the first spectral resolution, the third spatial resolution is lower than the second spatial resolution, the third spectral resolution is higher than the second spectral resolution, the fourth spatial resolution is lower than the third spatial resolution, the fourth spectral resolution is higher than the third spectral resolution, the first image sensor can be used to capture a first image, the second image sensor can be used to capture a second image, the third image sensor can be used to capture a third image, the fourth image sensor can be used to capture a fourth image, and wherein the first image is a panchromatic image type, the second image is a multispectral image type, the third image is a hyperspectral image type, and the fourth image is an ultra-hyperspectral image type.

[0020] In view of the deficiencies of the prior art, the present invention provides a distributed remote sensing satellite system, which collects high-definition remote sensing images through low-orbit remote sensing satellites, and can transmit remote sensing data including remote sensing images to ground stations with the help of synchronous orbit satellites, thereby greatly improving and ensuring the transmission efficiency of remote sensing data.

[0021] Advantageously, the images simultaneously acquired by the at least four image sensors have a common overlapping area, and the first satellite fuses the common overlapping areas of every two images among the images simultaneously acquired by the at least four image sensors to form a number of first-category fused images, and then the first satellite fuses every two images among the number of first-category fused images to form a number of second-category fused images, and the first satellite uses at least one of the number of second-category fused images as a fused remote sensing image.

[0022] The present invention also provides a distributed remote sensing satellite system, which includes a plurality of first satellites and a plurality of second satellites. The remote sensing data collected by the first satellites can be directly transmitted to a ground station or indirectly transmitted to the ground station through corresponding second satellites.

[0023] The first satellite includes at least four image sensors that simultaneously capture ground images, and the ground areas captured by the at least four image sensors completely overlap or partially overlap. The first satellite fuses the common overlapping areas of every two images among the images simultaneously captured by the at least four image sensors to form a number of first-category fused images, and then the first satellite fuses every two images among the several first-category fused images to form a number of second-category fused images. The first satellite uses at least one of the several second-category fused images as a fused remote sensing image.

[0024] Advantageously, several first satellites are low-orbit remote sensing satellites and are distributed on at least two orbital planes, there are at least three first satellites on each of the at least two orbital planes, and the second satellites are geosynchronous orbit satellites. The remote sensing data collected by the first satellites can be transmitted directly to the ground station or indirectly to the ground station through the corresponding second satellites.

[0025] Advantageously, each first satellite includes at least one first acquisition and aiming tracker and at least one second acquisition and aiming tracker, and each second satellite includes at least two third acquisition and aiming trackers, the first acquisition and aiming tracker is configured to emit a laser toward the direction of the earth so as to establish laser communication between the first satellite and the ground station, the second acquisition and aiming tracker is configured to emit a laser in a direction away from the earth so as to establish laser communication between the first satellite and the second satellite together with the third acquisition and aiming tracker, and the third acquisition and aiming tracker is configured to emit a laser toward the direction of the earth so that the second satellite can establish laser communication with the first satellite and / or the ground station, and when the corresponding first satellite needs to transmit the collected remote sensing data Before data is transmitted to the ground station, the corresponding first satellite sends a transmission time comparison request to the corresponding second satellite; in response to the transmission time comparison request, the corresponding second satellite determines the estimated time of the first transmission path and the second transmission path for the corresponding first satellite at least based on meteorological conditions, and the first satellite selects a transmission path from the first transmission path and the second transmission path according to the estimated time to transmit the remote sensing data, wherein the first transmission path is a laser communication link established between the corresponding first satellite directly and the ground station receiving the remote sensing data, and the second transmission path is a laser communication link established between the corresponding first satellite and the ground station receiving the remote sensing data indirectly through the corresponding second satellite.

[0026] Advantageously, after the corresponding first satellite sends a transmission time comparison request to the corresponding second satellite, the corresponding second satellite determines the estimated time of the first transmission path and the second transmission path based at least on the position information of the corresponding first satellite, the data transceiving capability of the corresponding first satellite, the position information of the ground station receiving remote sensing data, the data transceiving capability of the ground station receiving remote sensing data, the position information of the second satellite, the data transceiving capability of the second satellite and the meteorological conditions.

[0027] Advantageously, when the corresponding second satellite determines the estimated time of the first transmission path and the second transmission path, the meteorological GIS platform of the corresponding second satellite periodically acquires meteorological data to simulate meteorological conditions based on the meteorological data. When the meteorological GIS platform of the corresponding second satellite simulates meteorological conditions, the meteorological GIS platform of the corresponding second satellite simulates meteorological elements that change with the first transmission path and the second transmission path. The corresponding second satellite determines the simulated positions of the corresponding first satellite, the ground station receiving remote sensing data and the second satellite within the meteorological GIS platform based on the position information of the corresponding first satellite, the position information of the ground station receiving remote sensing data and the position information of the second satellite. The meteorological GIS platform of the corresponding second satellite also dynamically simulates the movement of the corresponding first satellite according to time changes, so that the corresponding second satellite determines the estimated time of the first transmission path and the second transmission path when transmitting remote sensing data based on the meteorological condition simulation and the movement of the corresponding first satellite and sends it to the corresponding first satellite. The corresponding first satellite selects one of the transmission paths to transmit remote sensing data based on at least the estimated time of the first transmission path and the second transmission path.

[0028] Advantageously, the processing by which the corresponding second satellite determines the estimated time consumption for transmitting remote sensing data along the first transmission path and the second transmission path based on the meteorological condition simulation and the movement of the corresponding first satellite includes: the corresponding second satellite draws a first virtual laser beam representing the laser beam for establishing laser communication between the first satellite and the ground station between the corresponding first satellite simulated in its meteorological GIS platform and the ground station receiving the remote sensing data; the corresponding second satellite draws a second virtual laser beam representing the laser beam for establishing laser communication between the second satellite and the ground station between the corresponding second satellite simulated in its meteorological GIS platform and the ground station receiving the remote sensing data; determines a first blocking time and a first effective transmission time used by the first virtual laser beam to complete data transmission during the simulation process according to the changing meteorological elements and the first virtual laser beam with a changing angle; determines a second blocking time and a second effective transmission time used by the second virtual laser beam to complete data transmission during the simulation process according to the changing meteorological elements and the second virtual laser beam with a fixed angle; calculates the sum of the first blocking time and the first effective transmission time to obtain the estimated time consumption required for transmitting remote sensing data via the first transmission path; and calculates the sum of the second blocking time and the second effective transmission time to obtain the estimated time consumption required for transmitting remote sensing data via the second transmission path.

[0029] Advantageously, each first satellite has at least four image collectors, which can simultaneously collect images of the same area on the ground, and the spatial resolution and spectral resolution of the images collected by the at least four image collectors are different from each other. The first satellite performs image fusion on the images collected by the at least four image collectors to generate a fused remote sensing image.

[0030] Advantageously, the at least four image collectors include a first image collector, a second image collector, a third image collector and a fourth image collector, the first image collector has a first spatial resolution and a first spectral resolution, the second image collector has a second spatial resolution and a second spectral resolution, the third image collector has a third spatial resolution and a third spectral resolution, the fourth image collector has a fourth spatial resolution and a fourth spectral resolution, the second spatial resolution is lower than the first spatial resolution, the second spectral resolution is higher than the first spectral resolution, the third spatial resolution is lower than the second spatial resolution, the third spectral resolution is higher than the second spectral resolution, the fourth spatial resolution is lower than the third spatial resolution, and the fourth spectral resolution is higher than the third spectral resolution.

[0031] Advantageously, the first image collector can be used to collect the first image, the second image collector can be used to collect the second image, the third image collector can be used to collect the third image, and the fourth image collector can be used to collect the fourth image. The first satellite fuses every two images of the images of the same area on the ground simultaneously collected by the at least four image collectors to form a number of first-category fused images, and then the first satellite fuses every two images of the several first-category fused images to form a number of second-category fused images. The first satellite uses at least one of the several second-category fused images as a fused remote sensing image.

[0032] Advantageously, the first image is of panchromatic image type, the second image is of multispectral image type, the third image is of hyperspectral image type and the fourth image is of ultra-hyperspectral image type.

[0033] Advantageously, the first satellite evaluates the image clarity of a plurality of second-category fused images, and selects at least one image with a top image clarity from the plurality of second-category fused images as a fused remote sensing image, wherein the processing of the first satellite evaluating the image clarity of a plurality of second-category fused images includes: performing image segmentation on the corresponding second-category fused images by introducing high and low thresholds and removing pseudo-edges to obtain image flat areas and image edge areas; using a point sharpness method to calculate the image flat area clarity of the image flat area; using a normalized square gradient method to calculate the image edge area clarity of the image edge area; performing a weighted summation of the flat area clarity and the image edge area clarity to obtain the image clarity of the corresponding second-category fused images; and sorting the image clarity of the corresponding second-category fused images. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 is a schematic diagram of a preferred embodiment of a first satellite;

[0035] Figure 2 is a simplified schematic diagram of a preferred embodiment of the present invention;

[0036] Figure 3 It is a partial schematic diagram of a preferred embodiment of the present invention;

[0037] Figure 4 is a schematic diagram of another preferred embodiment of the first satellite;

[0038] Figure 5 is a schematic diagram of a module of a preferred embodiment of a first satellite; and

[0039] Figure 6 It is a schematic diagram of modules of a preferred embodiment of the second satellite.

[0040] Reference numerals list

[0041] 100: first satellite; 110: first ATP device; 120: second ATP device; 131: first image sensor; 132: second image sensor; 133: third image sensor; 134: fourth image sensor; 140: landmark recognition module; 150: error correction module; 160: resampling module; 200: second satellite; 210: third ATP device; 220: meteorological GIS platform; 300: ground station. DETAILED DESCRIPTION

[0042] The following is combined with Figure 1 , 2 , 3, 4, 5 and 6 are explained in detail.

[0043] The term "module" used herein describes any hardware, software, or combination of hardware and software that can perform the functions associated with the "module." Each module in the present invention can be one or more of a server, a dedicated integrated chip, and a server group.

[0044] According to a feasible mode, the present invention discloses a remote sensing satellite system, or a remote sensing system, or a remote sensing system based on a low-orbit remote sensing satellite, or a distributed remote sensing system, or a distributed remote sensing satellite system. The system is suitable for executing the various method steps recorded in the present invention to achieve the expected technical effect.

[0045] Advantageously, the system may include a first satellite 100 and / or a second satellite. The first satellite 100 may include at least four image sensors. The at least four image sensors may simultaneously capture images of the ground. The ground areas captured by the at least four image sensors may completely overlap or partially overlap. The spatial resolution and spectral resolution of the images captured by the at least four image sensors may be different from each other. The first satellite 100 may perform image fusion on at least a portion of the images captured by the at least four image sensors to generate a fused remote sensing image. The present invention can achieve at least the following beneficial technical effects by adopting this method: first, the present invention can fuse images with different spatial resolutions and different spectral resolutions collected by the same satellite to obtain a fused remote sensing image, can combine data of multiple different characteristics, learn from each other's strengths and weaknesses, give full play to their respective advantages, make up for their respective shortcomings, and more comprehensively reflect ground targets, so as to efficiently obtain high-definition images by utilizing the limited resources of satellites; second, the images to be fused are images collected by the same satellite at the same time and the same altitude above the ground. Compared with images collected by different satellites at different times and altitudes above the ground, the fusion difficulty is not only smaller, the efficiency is higher, but also the image distortion is smaller; third, the integration between spatial information is more natural; fourth, through automatic multi-level spatial and spectral resolution fusion processing, multi-level spatial and spectral information from at least four image sensors can be effectively combined to create a high-spatial resolution, large-coverage hyperspectral image.

[0046] Advantageously, the at least four image sensors may include a first image sensor 131, a second image sensor 132, a third image sensor 133, and a fourth image sensor 134. The first image sensor 131 may have a first spatial resolution and a first spectral resolution. The second image sensor 132 may have a second spatial resolution and a second spectral resolution. The third image sensor 133 may have a third spatial resolution and a third spectral resolution. The fourth image sensor 134 may have a fourth spatial resolution and a fourth spectral resolution. The second spatial resolution may be lower than the first spatial resolution. The second spectral resolution may be higher than the first spectral resolution. The third spatial resolution may be lower than the second spatial resolution. The third spectral resolution may be higher than the second spectral resolution. The fourth spatial resolution may be lower than the third spatial resolution. The fourth spectral resolution may be higher than the third spectral resolution. The first image sensor 131 may be used to capture a first image. The second image sensor 132 may be used to capture a second image. The third image sensor 133 may be used to capture a third image. The fourth image sensor 134 may be used to capture a fourth image.

[0047] Advantageously, the images simultaneously acquired by the at least four image sensors may have a common overlapping area. The first satellite 100 fuses the common overlapping areas of every two images among the images simultaneously acquired by the at least four image sensors to form a plurality of first-type fused images. The first satellite 100 may fuse every two images among the plurality of first-type fused images to form a plurality of second-type fused images. The first satellite 100 may use at least one of the plurality of second-type fused images as a fused remote sensing image.

[0048] Advantageously, the first image may be of panchromatic image type. The second image may be of multispectral image type. The third image may be of hyperspectral image type. The fourth image may be of ultra-hyperspectral image type.

[0049] Advantageously, the first satellite 100 may include a landmark recognition module 140 and an error correction module 150. The landmark recognition module 140 may be configured to obtain landmark information associated with each image acquired by the at least four image sensors. The error correction module 150 may be configured to calculate a state vector for correcting an orbit error and an attitude error of the first satellite 100 associated with each image acquired by the at least four image sensors based on the landmark information.

[0050] Advantageously, the landmark identification module 140 may be configured to: select at least three landmarks from each image captured by at least four image sensors; determine the remote sensing landmark positions of the at least three landmarks in each image captured by at least four image sensors and the actual landmark positions on the earth; calculate the difference between the corresponding remote sensing landmark positions and the actual landmark positions; and / or obtain landmark information based on the difference between the corresponding remote sensing landmark positions and the actual landmark positions.

[0051] Advantageously, the landmark identification module 140 may be configured to: before selecting at least three landmarks from each image acquired by the at least four image sensors, identify the number of landmarks in each image acquired by the at least four image sensors. When the number of identifiable landmarks in each image acquired by the at least four image sensors is greater than or equal to three, at least three landmarks may be selected from each image acquired by the at least four image sensors, remote sensing landmark positions of the at least three landmarks in each image acquired by the at least four image sensors and actual landmark positions on the earth may be determined, a difference between the corresponding remote sensing landmark positions and the actual landmark positions may be calculated, and landmark information may be acquired based on the difference between the corresponding remote sensing landmark positions and the actual landmark positions. When the number of identifiable landmarks in each image captured by at least four image sensors is less than three, a landmark with directional directivity can be selected from each image captured by at least four image sensors, the remote sensing landmark position and orientation of the landmark with directional directivity in each image captured by at least four image sensors and the actual landmark position and orientation on the earth are determined, the difference between the corresponding remote sensing landmark position and orientation and the actual landmark position and orientation is calculated, and landmark information is obtained based on the difference between the corresponding remote sensing landmark position and orientation and the actual landmark position and orientation. The landmark with directional directivity can be at least one of a river, an airstrip, a road, and a coastline.

[0052] Advantageously, the first satellite 100 may include a resampling module 160. The resampling module 160 is configured to resample the pixel position of each image acquired by the at least four image sensors based on the calculated state vector. After the pixel position of each image acquired by the at least four image sensors is resampled based on the calculated state vector, image fusion is performed on at least a portion of the images acquired by the at least four image sensors to generate a fused remote sensing image.

[0053] According to one feasible manner, the system may include a plurality of first satellites 100 and a plurality of second satellites 200. The plurality of first satellites 100 may be low-orbit remote sensing satellites and distributed on at least two orbital planes different from each other. There may be at least three first satellites 100 on each of the at least two orbital planes. The second satellite 200 may be a geosynchronous orbit satellite. The remote sensing images and / or remote sensing data collected by the first satellite 100 may be directly transmitted to the ground station 300 or indirectly transmitted to the ground station 300 through the corresponding second satellite 200. Preferably, the system may include at least three second satellites 200. Particularly preferably, the system may include at least nine second satellites 200. Preferably, the remote sensing data may refer to data and / or data packets containing remote sensing images. Preferably, the ground station 300 may include a microwave station and / or an optical station. Preferably, the first satellite 100 and / or the second satellite 200 may perform microwave communication with the ground station 300. The present invention can achieve at least the following beneficial technical effects by adopting this method: the present invention collects high-definition remote sensing images through low-orbit remote sensing satellites, and can transmit remote sensing data to ground stations with the help of synchronous orbit satellites, greatly improving and ensuring the transmission efficiency of remote sensing data.

[0054] According to a preferred embodiment, each first satellite 100 may include at least one first ATP device 110 and at least one second ATP device 120. Each second satellite 200 may include at least two third ATP devices 210. The first ATP device 110 may be configured to emit a laser toward the direction of the earth so as to establish laser communication between the first satellite 100 and the ground station 300. The first ATP device 110 may be configured to emit a laser toward the direction of the earth so as to establish laser communication between the first satellite 100 and the ground station 300. The second ATP device 120 may be configured to emit a laser in a direction away from the earth so as to establish laser communication between the first satellite 100 and the second satellite 200 together with the third ATP device 210. The third ATP device 210 may be configured to emit a laser toward the direction of the earth so that the second satellite 200 can establish laser communication with the first satellite 100 and / or the ground station 300. Before the corresponding first satellite 100 needs to transmit the collected remote sensing data to the ground station 300, the corresponding first satellite 100 may send a transmission time comparison request to the corresponding second satellite 200. In response to the transmission time comparison request, the corresponding second satellite 200 may determine the estimated time of the first transmission path and the second transmission path for the corresponding first satellite 100 based at least on the meteorological conditions. The first satellite 100 may select a transmission path from the first transmission path and the second transmission path to transmit the remote sensing data according to the estimated time. The first transmission path may be a laser communication link established between the corresponding first satellite 100 directly and the ground station 300 receiving the remote sensing data. The second transmission path may be a laser communication link established between the corresponding first satellite 100 and the ground station 300 receiving the remote sensing data indirectly through the corresponding second satellite 200.

[0055] Preferably, the laser communication link established by the corresponding first satellite 100 indirectly through the second satellite 200 and the ground station 300 receiving remote sensing data may include two modes. The first mode may be that the corresponding first satellite 100 indirectly through the second satellite 200 and the ground station 300 receiving remote sensing data establishes a real-time laser communication link, that is, the corresponding first satellite 100 and the corresponding second satellite 200 and the corresponding second satellite 200 and the ground station 300 receiving remote sensing data establish laser communication links at the same time. The second mode may be that the corresponding first satellite 100 first transmits the remote sensing data to the corresponding second satellite 200 through the laser communication link established by the two, and then the corresponding second satellite 200 selects an opportunity to establish a laser communication link with the ground station 300 and transmit the remote sensing data. The present invention can achieve at least the following beneficial technical effects by adopting this method: first, by determining the transmission path through analysis of the second satellite, the efficiency of data transmission can be better guaranteed; second, with the help of laser communication, the transmission efficiency can be further improved; third, the security of remote sensing data transmission can be increased; fourth, by transmitting data to the second satellite, the second satellite can send remote sensing data to the ground station under meteorological conditions suitable for laser communication, without having to wait for the first satellite to rotate once and return to a position visible to the ground station before continuing the transmission.

[0056] Preferably, the first satellite 100 can be configured to allow the first ATP device 110 to periodically establish a laser communication link with the ground station 300. Preferably, ATP can refer to Acquisition, Tracking and Pointing, that is, capture, tracking and pointing. Preferably, the ATP device can also be called an APT device, a capture and pointing tracker, a capture, tracking and pointing system, a pointing, capture and tracking device and / or a capture, tracking and pointing device. For example, taking the ground station 300 and the satellite as an example, in order to achieve reliable communication between satellites or between satellites and other communication equipment, it is first required that a satellite can capture a light beam sent by another satellite or ground station 300, which is called a beacon light, and converge the light beam to the center of the detector or antenna. This process is called capture or capture body. After the capture is completed, the satellite as the receiving party also needs to emit a light beam, which is required to accurately point to the other satellite or ground station 300 that emits the beacon light. This process is called pointing or pointing. After the satellite that emits the beacon light receives this light beam, it must also complete the capture process accordingly, so that the two satellites or the satellite and the ground station 300 can finally reach a communication connection state. To ensure that the two satellites or the satellite and the ground station 300 are always in communication, this precise connection state must be maintained, and this process is called tracking or tracking. Preferably, there are multiple mathematical expressions for determining the attitude and position of the object, such as at least one of Euler angles, Euler-Rodrigues parameters, Rodrigues-Gipps vectors, quaternions, and dual quaternions.

[0057] Advantageously, after the corresponding first satellite 100 sends a transmission time comparison request to the corresponding second satellite 200, the corresponding second satellite 200 can determine the estimated time consumption of the first transmission path and the second transmission path based at least on the position information of the corresponding first satellite 100, the data transceiving capability of the corresponding first satellite 100, the position information of the ground station 300 receiving remote sensing data, the data transceiving capability of the ground station 300 receiving remote sensing data, the position information of the second satellite 200, the data transceiving capability of the second satellite 200 and the meteorological conditions.

[0058] According to a preferred embodiment, when the corresponding second satellite 200 determines the estimated time of the first transmission path and the second transmission path, the meteorological GIS platform 220 of the corresponding second satellite 200 can periodically obtain meteorological data to simulate meteorological conditions according to the meteorological data. When the meteorological GIS platform 220 of the corresponding second satellite 200 simulates meteorological conditions, the meteorological GIS platform 220 of the corresponding second satellite 200 can simulate meteorological elements that change with the first transmission path and the second transmission path. The corresponding second satellite 200 can determine the simulated positions of the corresponding first satellite 100, the ground station 300 receiving remote sensing data, and the second satellite 200 in the meteorological GIS platform 220 based on the position information of the corresponding first satellite 100, the position information of the ground station 300 receiving remote sensing data, and the position information of the second satellite 200. The meteorological GIS platform 220 of the corresponding second satellite 200 can dynamically simulate the movement of the corresponding first satellite 100 according to the time change, so that the corresponding second satellite 200 determines the estimated time of the first transmission path and the second transmission path when transmitting the remote sensing data based on the meteorological condition simulation and the movement of the corresponding first satellite 100, and sends it to the corresponding first satellite 100. The corresponding first satellite 100 can select one of the transmission paths to transmit the remote sensing data based on at least the estimated time of the first transmission path and the second transmission path.

[0059] Preferably, the second satellite 200 can obtain satellite data from the ground station 300 and / or a meteorological satellite. Meteorological elements may include at least clouds. Meteorological elements may include at least one of clouds, rain, snow, fog and wind. The present invention can achieve at least the following beneficial technical effects by adopting this method: the meteorological GIS platform is carried on the second satellite 200 for analysis, which can avoid poor communication caused by interference from atmospheric environmental factors and thus delay in analysis, and meteorological data can be directly obtained through the second satellite 200 for analysis quickly and efficiently.

[0060] According to a preferred embodiment, the processing in which the corresponding second satellite 200 can determine the estimated time consumption of the first transmission path and the second transmission path when transmitting remote sensing data based on the meteorological condition simulation and the movement of the corresponding first satellite 100 may include: the corresponding second satellite 200 draws a first virtual laser beam representing the laser communication established between the first satellite 100 and the ground station 300 between the corresponding first satellite 100 simulated in its meteorological GIS platform 220 and the ground station 300 receiving the remote sensing data. The processing in which the corresponding second satellite 200 can determine the estimated time consumption of the first transmission path and the second transmission path when transmitting remote sensing data based on the meteorological condition simulation and the movement of the corresponding first satellite 100 may include: the corresponding second satellite 200 draws a second virtual laser beam representing the laser beam establishing laser communication between the second satellite 200 and the ground station 300 between the corresponding second satellite 200 simulated in its meteorological GIS platform 220 and the ground station 300 receiving the remote sensing data. The processing that the corresponding second satellite 200 can determine the estimated time consumption of the first transmission path and the second transmission path when transmitting remote sensing data based on the meteorological condition simulation and the movement of the corresponding first satellite 100 may include: determining the first blocking time and the first effective transmission time used by the first virtual laser beam to complete data transmission during the simulation process according to the changing meteorological elements and the first virtual laser beam with a changing angle. The processing that the corresponding second satellite 200 can determine the estimated time consumption of the first transmission path and the second transmission path when transmitting remote sensing data based on the meteorological condition simulation and the movement of the corresponding first satellite 100 may include: determining the second blocking time and the second effective transmission time used by the second virtual laser beam to complete data transmission during the simulation process according to the changing meteorological elements and the second virtual laser beam with a fixed angle. The processing that the corresponding second satellite 200 can determine the estimated time consumption of the first transmission path and the second transmission path when transmitting remote sensing data based on the meteorological condition simulation and the movement of the corresponding first satellite 100 may include: calculating the sum of the first blocking time and the first effective transmission time to obtain the estimated time consumption required for transmitting remote sensing data through the first transmission path. The corresponding second satellite 200 can determine the estimated time consumption of the first transmission path and the second transmission path when transmitting remote sensing data based on the meteorological condition simulation and the movement of the corresponding first satellite 100, and the processing may include: calculating the sum of the second blocking time and the second effective transmission time to obtain the estimated time consumption required for transmitting remote sensing data through the second transmission path. The first blocking time may refer to the time when the first virtual laser beam is affected by meteorological factors and cannot communicate during the simulation process. The first blocking time may include the time when the first virtual laser beam is blocked and the link establishment time required for the first virtual laser beam to re-establish the laser communication link each time it changes from being blocked to being unblocked. The second blocking time may refer to the time when the second virtual laser beam is affected by meteorological factors and cannot communicate during the simulation process.The second blocking time may include the time when the second virtual laser beam is blocked and the link establishment time of re-establishing the laser communication link each time the second virtual laser beam changes from being blocked to being unblocked. The link establishment time may be the average time or estimated time required for two capture and aiming trackers to build a laser communication link with each other. Preferably, the second satellite 200 draws the first virtual laser beam may be a line segment drawn between the corresponding first satellite 100 simulated in the meteorological GIS platform and the ground station 300 receiving remote sensing data. Since the position of the simulated ground station is stationary, and the corresponding simulated first satellite 100 is moving, the angle of the first virtual laser beam will change. Preferably, the second satellite 200 draws the second virtual laser beam may be a line segment drawn between the corresponding second satellite 200 simulated in the meteorological GIS platform and the ground station 300 receiving remote sensing data. Since the position of the simulated ground station is stationary, and the position of the corresponding simulated second satellite 200 is also not moving, the angle of the second virtual laser beam is fixed. Preferably, the meteorological element may include at least one of cloud, rain, snow, fog and wind. Preferably, the second satellite 200 stores the set blocking coefficient of the corresponding meteorological element. For example, the second satellite 200 can set the cloud blocking coefficient to 0-1 according to the thickness of the cloud layer. The second satellite 200 can set the rain blocking coefficient to 0-1 according to the amount of precipitation. The second satellite 200 can set the snow blocking coefficient to 0-1 according to the amount of precipitation. The second satellite 200 can set the fog blocking coefficient to 0-1 according to the size of the droplet diameter. The size and direction of the wind can determine the movement of the cloud. The blocking threshold can be set to 1. The second satellite 200 can determine that the first virtual laser beam is blocked when the sum of the blocking coefficients of all meteorological elements that the first virtual laser beam is to penetrate at the corresponding moment is greater than or equal to the blocking threshold. The second virtual laser beam is blocked when the sum of the blocking coefficients of all meteorological elements that the second virtual laser beam is to penetrate at the corresponding moment is greater than or equal to the blocking threshold. For example, when the sum of the blocking coefficients of all meteorological elements that the first virtual laser beam or the second laser beam is to penetrate at the corresponding moment is 1 or 1.5, it is determined to be blocked. When the sum of the blocking coefficients of all meteorological elements that the first virtual laser beam is to penetrate at the corresponding moment is less than the blocking threshold, the first virtual laser beam is deemed to be unblocked. When the sum of the blocking coefficients of all meteorological elements that the second virtual laser beam is to penetrate at the corresponding moment is less than the blocking threshold, the second virtual laser beam is deemed to be unblocked. For example, when the sum of the blocking coefficients of all meteorological elements that the first virtual laser beam or the second laser beam is to penetrate at the corresponding moment is 0.2 or 0.5, it is deemed to be unblocked. Particularly preferably, the blocking coefficients of all clouds, rain, snow and fog in the second satellite 200 can be set to 1. The blocking threshold can be set to 1. That is, as long as the first virtual laser beam or the second virtual laser beam needs to penetrate clouds, rain, snow and fog at the corresponding moment during the simulation, it is considered to be blocked.The present invention adopts this method to achieve at least the following beneficial technical effects: first, the present invention adopts the first virtual laser beam or the second virtual laser beam to quickly determine the meteorological elements that the corresponding laser communication link needs to experience or penetrate during the simulation process, so as to shorten the simulation time; second, because the establishment speed of laser communication links is not as fast as that of microwave communication links, the present invention takes into account the link establishment time required for rebuilding the laser and re-establishing the laser communication link after the first virtual laser beam or the second virtual laser beam is blocked each time, so as to make the calculation of the estimated time more accurate, and make the present invention more reliable.

[0061] Advantageously, each first satellite 100 may include at least four image sensors. At least four image sensors may simultaneously capture images of the same area on the ground. The spatial resolution and spectral resolution of the images captured by the at least four image sensors may be different from each other. The first satellite 100 may perform image fusion on the images captured by the at least four image sensors to generate a fused remote sensing image. Preferably, the image fusion method may, for example, adopt at least one of a band algebraic cloud algorithm, an IHS transform fusion method, a wavelet transform fusion algorithm, a spectral sharpening fusion method, and a principal component transform fusion method. Particularly preferably, the present invention adopts a spectral sharpening fusion method for image fusion. The invention can achieve at least the following beneficial technical effects by adopting this method: First, the present invention can fuse images with different spatial resolutions and different spectral resolutions collected by the same satellite to obtain a fused remote sensing image, can combine data of multiple different characteristics, learn from each other's strengths and weaknesses, give full play to their respective advantages, make up for their respective shortcomings, and more comprehensively reflect ground targets, so as to efficiently obtain high-definition images by utilizing the limited resources of satellites; second, the images to be fused are images collected by the same satellite at the same time and the same altitude above the ground. Compared with images collected by different satellites at different times and altitudes above the ground, the fusion difficulty is not only smaller, the efficiency is higher, but also the image distortion is smaller; third, the integration between spatial information is more natural; fourth, through automatic multi-level spatial and spectral resolution fusion processing, multi-level spatial and spectral information from at least four image collectors can be effectively combined to create high-spatial resolution, large-coverage hyperspectral images.

[0062] Advantageously, the number of image sensors may vary according to the design of the image sensors, the materials used, and / or the computing performance of the device used for image fusion. For example, 5, 6, 7, 8, 10, 16 or more image sensors may be used.

[0063] Advantageously, at least four image sensors may have the same FOV and / or the same ground strip. At least four image sensors may have a common overlapping area to capture images of the same area. Preferably, when performing image fusion on the images captured by the at least four image sensors, the image data used for fusion may include all or part of the image data in the common overlapping area. Preferably, the fused image data may include all spectral bands of the third image and / or the fourth image that define the spectral resolution of the overlapping area. Preferably, all spectral bands of the third image and the fourth image define the spectral resolution of the common overlapping area.

[0064] Advantageously, the at least four image sensors may include a first image sensor 131, a second image sensor 132, a third image sensor 133, and a fourth image sensor 134. The first image sensor 131 has a first spatial resolution and a first spectral resolution, the second image sensor 132 has a second spatial resolution and a second spectral resolution, the third image sensor 133 has a third spatial resolution and a third spectral resolution, the fourth image sensor 134 has a fourth spatial resolution and a fourth spectral resolution, the second spatial resolution is lower than the first spatial resolution, the second spectral resolution is higher than the first spectral resolution, the third spatial resolution is lower than the second spatial resolution, the third spectral resolution is higher than the second spectral resolution, the fourth spatial resolution is lower than the third spatial resolution, and the fourth spectral resolution is higher than the third spectral resolution. Preferably, the first image sensor 131 acquires a first image, the second image sensor 132 acquires a second image, the third image sensor 133 acquires a third image, and the fourth image sensor 134 acquires a fourth image. Preferably, the first image, the second image, the third image, or the fourth image may be at least one of a panchromatic image type, a multispectral image type, a hyperspectral image type, and an ultra-hyperspectral image type. Particularly preferably, the first image may be a panchromatic image type. The second image may be a multispectral image type. The third image may be a hyperspectral image type. The fourth image may be an ultra-hyperspectral image type. Thus, the image fusion method of the present invention can significantly improve the imaging quality of remote sensing images. Preferably, panchromatic may refer to all visible light bands of 0.38 to 0.76um, and the panchromatic image is a mixed image of this band range, generally a black and white image. Preferably, the multispectral image type may refer to an image collected using multispectral imaging technology, generally having 10 to 20 spectral channels and a spectral resolution of λ / Δλ≈10. Preferably, the hyperspectral image type may refer to an image collected using hyperspectral imaging technology. Generally, it has a detection capability of 100 to 400 spectral channels, and the general spectral resolution can reach λ / Δλ≈100. Preferably, the ultra-hyperspectral image type may refer to an image collected using ultra-hyperspectral imaging. Generally, the number of spectral channels is around 1000, and the spectral resolution is generally λ / Δλ≥1000.

[0065] Advantageously, the first satellite 100 may fuse every two images of the same area on the ground collected simultaneously by at least four image sensors to form a plurality of first-category fused images. The first satellite 100 may fuse every two images of the plurality of first-category fused images to form a plurality of second-category fused images. The first satellite 100 may use at least one of the plurality of second-category fused images as a fused remote sensing image. Preferably, for example, the first satellite 100 may fuse every two images of the first image, the second image, the third image, and the fourth image to form six first-category fused images. The first satellite 100 may fuse every two images of the six first-category fused images to form fifteen second-category fused images. The present invention adopts this method to achieve at least the following beneficial technical effects: since the images collected from high altitudes on satellites will be affected by various factors, such as satellite vibration, radiation or imaging angle differences, the images collected by different image sensors have different effects on the fused image. If a fixed form of image fusion method is used, the quality of image fusion may fluctuate greatly. The present invention adopts this method to select at least one fused image from multiple fused second-category fused images as the fused remote sensing image, so as to ensure or improve the quality of the fused image.

[0066] Advantageously, the first satellite 100 can evaluate the image clarity of the plurality of second-type fused images. The first satellite 100 can select at least one image with a higher image clarity from the plurality of second-type fused images as the fused remote sensing image.

[0067] Advantageously, the first satellite 100 may evaluate the image clarity of several second-category fused images by: segmenting the corresponding second-category fused images by introducing high and low thresholds and removing pseudo-edges to obtain image flat areas and image edge areas; using the point sharpness method to calculate the image flat area clarity of the image flat area; using the normalized square gradient method to calculate the image edge area clarity of the image edge area; performing weighted summation of the flat area clarity and the image edge area clarity to obtain the image clarity of the corresponding second-category fused images; and / or sorting the image clarity of the corresponding second-category fused images. The present invention can achieve at least the following beneficial technical effects by adopting this method: first, the preferred embodiment utilizes the advantages of the point sharpness method and the square gradient method in terms of good noise resistance, strong unimodality, high sensitivity, and good unbiasedness, and can accurately and stably evaluate image clarity; second, it is suitable for evaluating image clarity without a reference image.

[0068] Advantageously, the first satellite 100 may include a landmark recognition module 140 and / or an error correction module 150. The landmark recognition module 140 and the error correction module 150 may be one or more of a server, a dedicated integrated chip, and a server group.

[0069] The landmark recognition module 140 may be configured to obtain landmark information associated with each image acquired by the at least four image sensors. The error correction module 150 may be configured to calculate a state vector for correcting at least one of an orbit error and an attitude error of the first satellite 100 associated with each image acquired by the at least four image sensors based on the landmark information. Preferably, the first satellite 100 may be a low-orbit remote sensing satellite.

[0070] Advantageously, the first satellite 100 may be configured to control the first ATP device 110 to establish a laser communication link with the ground station 300. The error correction module 150 may correct the orbit, position and attitude of the first satellite 100 based at least on the laser communication link established between the first ATP device 110 and the ground station 300. Advantageously, the calculation of the state vector may include calculating the state vector using a Kalman filter algorithm.

[0071] Advantageously, the landmark identification module 140 may be configured to: select at least three landmarks from each image captured by at least four image sensors; determine the remote sensing landmark positions of the at least three landmarks in each image captured by at least four image sensors and the actual landmark positions on the earth; calculate the difference between the corresponding remote sensing landmark positions and the actual landmark positions; and / or obtain landmark information based on the difference between the corresponding remote sensing landmark positions and the actual landmark positions.

[0072] Advantageously, the first satellite 100 may include a landmark recognition module 140 and an error correction module 150. The landmark recognition module 140 may be configured to obtain landmark information associated with each image acquired by at least four image sensors. The error correction module 150 may be configured to correct a state vector of at least one of an orbit error, an attitude error, and a payload misalignment error for each image acquired by at least four image sensors based on the landmark information. Preferably, the first satellite 100 may be a low-orbit remote sensing satellite. The present invention may achieve at least the following beneficial technical effects by adopting this method: when satellite-level distributed spacecraft acquire remote sensing images, image distortion may occur, and therefore it is necessary to correct geometric distortion in remote sensing images to provide accurate observation information. The system of FIG. 1 uses landmarks and stars as reference points for geometric correction. Landmarks are sensitive to both the orbit and attitude of the satellite, and therefore can be used to correct the orbit and attitude. In contrast, stars are only sensitive to the attitude of the satellite, and therefore may be useful for correcting the attitude. However, due to the large number of stars, there are more than 5,000 stars in the celestial sphere with apparent magnitude brighter than 6. Unlike the sun, moon, and earth, there is only one reference celestial body. Star identification must be performed, and it must be close to real-time identification. This is the technical difficulty of star sensors. In addition, star sensors have low-frequency errors. The low-frequency error of star sensors is mainly due to the movement of the optical axis pointing of the star sensor under the change of the solar illumination angle, resulting in periodic errors. This has been found in the transmission data of multiple satellites such as the Advanced Earth Observation Satellite and Tianhui No. 1. The Sentinel-2 satellite models the low-frequency error of the star sensor as a first-order Gauss-Markov process, and filters the low-frequency error of the star sensor through covariance adjustment, but the model fails to fully reflect the changing trend of the low-frequency error, and the correction effect is limited. The present invention adopts this method to make good use of landmarks for correction. In addition to orbit error and attitude error, the payload misalignment error is also considered, so that the correction effect is better.

[0073] Preferably, the landmark may also be called a landmark, and may refer to a land feature with significant structural features, such as an island, a lake, a river, a coastline, a road, and a building. Advantageously, the calculation of the state vector may include calculating the state vector using a Kalman filter algorithm.

[0074] Advantageously, the landmark identification module 140 may be configured to: select at least three landmarks from each image captured by at least four image sensors; determine the remote sensing landmark positions of the at least three landmarks in each image captured by at least four image sensors and the actual landmark positions on the earth; calculate the difference between the corresponding remote sensing landmark positions and the actual landmark positions; and / or obtain landmark information based on the difference between the corresponding remote sensing landmark positions and the actual landmark positions.

[0075] Advantageously, the landmark recognition module 140 may be configured to: before selecting at least three landmarks from each image acquired by the at least four image sensors, identify the number of landmarks in each image acquired by the at least four image sensors; when the number of identifiable landmarks in each image acquired by the at least four image sensors is greater than or equal to three, select at least three landmarks from each image acquired by the at least four image sensors, determine the remote sensing landmark positions of the at least three landmarks in each image acquired by the at least four image sensors and the actual landmark positions on the earth, calculate the difference between the corresponding remote sensing landmark positions and the actual landmark positions, and based on the corresponding The landmark information is obtained by using the difference between the remote sensing landmark position and the actual landmark position; when the number of identifiable landmarks in each image collected by at least four image sensors is less than three, a landmark with directional directivity is selected from each image collected by at least four image sensors, the remote sensing landmark position and orientation of the landmark with directional directivity in each image collected by at least four image sensors and the actual landmark position and orientation on the earth are determined, the difference between the corresponding remote sensing landmark position and orientation and the actual landmark position and orientation is calculated, and the landmark information is obtained based on the difference between the corresponding remote sensing landmark position and orientation and the actual landmark position and orientation. Preferably, the landmark with directional directivity can be, for example, at least one of a river, an airstrip, a road and a coastline. The present invention can achieve at least the following beneficial technical effects by adopting this method: the present invention can more accurately determine the landmark information by selecting at least three landmarks when the number of identifiable landmarks is large, and when the number of identifiable landmarks is small, the accuracy of the landmark information can be improved as much as possible by using the position and orientation of the landmark with directional directivity.

[0076] Advantageously, the first satellite 100 may include a resampling module 160 configured to resample the pixel position of each image acquired by the at least four image sensors based on the calculated state vector. Preferably, after the pixel position of each image acquired by the at least four image sensors is resampled based on the calculated state vector, the images acquired by the at least four image sensors are image fused to generate a fused remote sensing image.

[0077] Advantageously, the ground station 300 may store the remote sensing image in a database, the processor communicates with the database to obtain the remote sensing image, divide the remote sensing image into a plurality of sub-images, obtain cropped sub-images by removing overlapping areas overlapping with adjacent images, generate pre-processed images each including the cropped sub-images, select a reference image and a target image therefrom, the pre-processed image determines a plurality of corresponding pairs in the overlapping area between the reference image and the target image based on a feature matching algorithm, obtains a transformation matrix by a least squares algorithm based on coordinates of the corresponding pairs, obtains calibration coordinates of each corresponding pixel of the target image by applying the transformation matrix, and stitches the target image into the wide-angle image based on the calibration coordinates of the target image.

[0078] According to a feasible manner, the method may include: acquiring images of the ground collected simultaneously from at least four image sensors of the first satellite 100; and / or performing image fusion on at least a portion of the images collected by the at least four image sensors to generate a fused remote sensing image. The ground areas collected by the at least four image sensors completely overlap or partially overlap, and the spatial resolutions and spectral resolutions of the images collected by the at least four image sensors are different from each other.

[0079] Preferably, the at least four image sensors include a first image sensor 131, a second image sensor 132, a third image sensor 133 and a fourth image sensor 134. The first image sensor 131 has a first spatial resolution and a first spectral resolution, the second image sensor 132 has a second spatial resolution and a second spectral resolution, the third image sensor 133 has a third spatial resolution and a third spectral resolution, the fourth image sensor 134 has a fourth spatial resolution and a fourth spectral resolution, the second spatial resolution is lower than the first spatial resolution, the second spectral resolution is higher than the first spectral resolution, the third spatial resolution is lower than the second spatial resolution, the third spectral resolution is higher than the second spectral resolution, the fourth spatial resolution is lower than the third spatial resolution, the fourth spectral resolution is higher than the third spectral resolution, the first image sensor 131 can be used to collect the first image, the second image sensor 132 can be used to collect the second image, the third image sensor 133 can be used to collect the third image, and the fourth image sensor 134 can be used to collect the fourth image. And wherein the first image is a panchromatic image type, the second image is a multispectral image type, the third image is a hyperspectral image type, and the fourth image is an ultra-hyperspectral image type.

[0080] Preferably, the images simultaneously captured by the at least four image sensors have a common overlapping area, and the first satellite 100 fuses the common overlapping areas of every two images among the images simultaneously captured by the at least four image sensors to form a number of first-category fused images, and then the first satellite 100 fuses every two images among the number of first-category fused images to form a number of second-category fused images, and the first satellite 100 uses at least one of the number of second-category fused images as a fused remote sensing image.

[0081] According to one feasible manner, the method for generating remote sensing images may include: using the system of the present invention to perform at least one of the acquisition, processing and transmission of remote sensing data. The method may be implemented by the system of the present invention and / or other replaceable components. For example, the method of the present invention is implemented by using various components in the system of the present invention. For example, error correction, resampling, image fusion and image stitching, etc.

[0082] Although the present invention has been described in detail, modifications within the spirit and scope of the present invention will be apparent to those skilled in the art. Such modifications are also considered to be a part of the present disclosure. In view of the foregoing discussion, the relevant knowledge in the art, and the references or information discussed above in conjunction with the background (all incorporated herein by reference), further description is considered unnecessary. In addition, it should be understood that the various aspects of the present invention and the parts of each embodiment can be combined or interchanged in whole or in part. Moreover, it will be understood by those of ordinary skill in the art that the foregoing description is merely illustrative and is not intended to limit the present invention.

Claims

1. A remote sensing satellite system, characterized in that: The invention comprises a first satellite (100), the first satellite (100) comprising at least four image sensors, the at least four image sensors simultaneously collecting images of the ground, the ground areas collected by the at least four image sensors completely overlap or partially overlap, the spatial resolution and spectral resolution of the images collected by the at least four image sensors are different from each other, the first satellite (100) performs image fusion on at least a part of the images collected by the at least four image sensors to generate a fused remote sensing image, The first satellite (100) comprises a landmark recognition module (140) and an error correction module (150), wherein the landmark recognition module (140) is configured to obtain landmark information associated with each image acquired by the at least four image sensors, and the error correction module (150) is configured to calculate a state vector for correcting an orbit error and an attitude error of the first satellite (100) associated with each image acquired by the at least four image sensors based on the landmark information.

2. The system according to claim 1, characterized in that The at least four image sensors include a first image sensor (131), a second image sensor (132), a third image sensor (133) and a fourth image sensor (134); The first image sensor (131) has a first spatial resolution and a first spectral resolution, the second image sensor (132) has a second spatial resolution and a second spectral resolution, the third image sensor (133) has a third spatial resolution and a third spectral resolution, the fourth image sensor (134) has a fourth spatial resolution and a fourth spectral resolution, the second spatial resolution is lower than the first spatial resolution, the second spectral resolution is higher than the first spectral resolution, the third spatial resolution is lower than the second spatial resolution, the third spectral resolution is higher than the second spectral resolution, the fourth spatial resolution is lower than the third spatial resolution, the fourth spectral resolution is higher than the third spectral resolution, the first image sensor (131) can be used to capture a first image, the second image sensor (132) can be used to capture a second image, the third image sensor (133) can be used to capture a third image, and the fourth image sensor (134) can be used to capture a fourth image.

3. The system according to claim 2, characterized in that The images simultaneously acquired by the at least four image sensors have a common overlapping area. The first satellite (100) fuses the common overlapping areas of every two images among the images simultaneously acquired by the at least four image sensors to form a plurality of first-type fused images. Then, the first satellite (100) fuses every two images among the plurality of first-type fused images to form a plurality of second-type fused images. The first satellite (100) uses at least one of the plurality of second-type fused images as a fused remote sensing image.

4. The system according to claim 3, characterized in that The first image is of a panchromatic image type, the second image is of a multispectral image type, the third image is of a hyperspectral image type, and the fourth image is of an ultra-hyperspectral image type.

5. The system according to claim 1, wherein: The landmark recognition module (140) is configured to: select at least three landmarks from each image captured by the at least four image sensors, determine the remote sensing landmark positions of the at least three landmarks in each image captured by the at least four image sensors and the actual landmark positions on the earth, calculate the difference between the corresponding remote sensing landmark positions and the actual landmark positions, and obtain landmark information based on the difference between the corresponding remote sensing landmark positions and the actual landmark positions.

6. The system according to claim 5, characterized in that The landmark identification module (140) is configured to: before selecting at least three landmarks from each image acquired by the at least four image sensors, identify the number of landmarks in each image acquired by the at least four image sensors; When the number of identifiable landmarks in each image acquired by the at least four image sensors is greater than or equal to three, selecting at least three landmarks from each image acquired by the at least four image sensors, determining remote sensing landmark positions of the at least three landmarks in each image acquired by the at least four image sensors and actual landmark positions on the earth, calculating the difference between the corresponding remote sensing landmark positions and the actual landmark positions, and acquiring landmark information based on the difference between the corresponding remote sensing landmark positions and the actual landmark positions; When the number of identifiable landmarks in each image acquired by at least four image sensors is less than three, a landmark with directional directivity is selected from each image acquired by at least four image sensors, the remote sensing landmark position and orientation of the landmark with directional directivity in each image acquired by at least four image sensors and the actual landmark position and orientation on the earth are determined, the difference between the corresponding remote sensing landmark position and orientation and the actual landmark position and orientation is calculated, and landmark information is acquired based on the difference between the corresponding remote sensing landmark position and orientation and the actual landmark position and orientation, wherein the landmark with directional directivity is at least one of a river, an airport runway, a road and a coastline.

7. The system according to claim 6, characterized in that The first satellite (100) further comprises: a resampling module (160), the resampling module (160) being configured to resample the pixel position of each image acquired by the at least four image sensors based on the calculated state vector; And only after the pixel position of each image collected by the at least four image sensors is resampled based on the calculated state vector, image fusion is performed on at least a portion of the images collected by the at least four image sensors to generate a fused remote sensing image.

8. A method for generating a remote sensing image, characterized in that: include: Acquire images of the ground collected simultaneously from at least four image sensors of a first satellite (100), performing image fusion on at least a portion of the images acquired by the at least four image sensors to generate a fused remote sensing image, The ground areas captured by the at least four image sensors completely overlap or partially overlap, and the spatial resolutions and spectral resolutions of the images captured by the at least four image sensors are different from each other.

9. The method according to claim 8, characterized in that The at least four image sensors include a first image sensor (131), a second image sensor (132), a third image sensor (133) and a fourth image sensor (134); The first image sensor (131) has a first spatial resolution and a first spectral resolution, the second image sensor (132) has a second spatial resolution and a second spectral resolution, the third image sensor (133) has a third spatial resolution and a third spectral resolution, the fourth image sensor (134) has a fourth spatial resolution and a fourth spectral resolution, the second spatial resolution is lower than the first spatial resolution, the second spectral resolution is higher than the first spectral resolution, the third spatial resolution is lower than the second spatial resolution, the third spectral resolution is higher than the second spectral resolution, the fourth spatial resolution is lower than the third spatial resolution, and the fourth spectral resolution is higher than the third spectral resolution. The first image sensor (131) can be used to capture a first image, the second image sensor (132) can be used to capture a second image, the third image sensor (133) can be used to capture a third image, and the fourth image sensor (134) can be used to capture a fourth image. And wherein the first image is of a panchromatic image type, the second image is of a multispectral image type, the third image is of a hyperspectral image type, and the fourth image is of an ultra-hyperspectral image type.

10. The method according to claim 8 or 9, characterized in that The images simultaneously acquired by the at least four image sensors have a common overlapping area. The first satellite (100) fuses the common overlapping areas of every two images among the images simultaneously acquired by the at least four image sensors to form a plurality of first-type fused images. Then, the first satellite (100) fuses every two images among the plurality of first-type fused images to form a plurality of second-type fused images. The first satellite (100) uses at least one of the plurality of second-type fused images as a fused remote sensing image.

11. A distributed remote sensing satellite system, characterized in that: The system comprises a plurality of first satellites (100) and a plurality of second satellites (200), wherein the remote sensing data collected by the first satellites (100) can be directly transmitted to a ground station (300) or indirectly transmitted to the ground station (300) via corresponding second satellites (200), wherein: The first satellite (100) comprises at least four image sensors for simultaneously acquiring ground images, the ground areas acquired by the at least four image sensors completely overlap or partially overlap, the first satellite (100) fuses common overlapping areas of every two images among the images acquired simultaneously by the at least four image sensors to form a plurality of first-type fused images, then the first satellite (100) fuses every two images among the plurality of first-type fused images to form a plurality of second-type fused images, and the first satellite (100) uses at least one of the plurality of second-type fused images as a fused remote sensing image, The first satellite (100) comprises a landmark recognition module (140) and an error correction module (150), wherein the landmark recognition module (140) is configured to obtain landmark information associated with each image acquired by the at least four image sensors, and the error correction module (150) is configured to calculate a state vector for correcting an orbit error and an attitude error of the first satellite (100) associated with each image acquired by the at least four image sensors based on the landmark information.

12. The distributed remote sensing satellite system according to claim 11, characterized in that: Each first satellite (100) includes at least one first ATP device (110) and at least one second ATP device (120), and each second satellite (200) includes at least two third ATP devices (210). The first ATP device (110) is configured to emit a laser toward the earth so as to establish laser communication between the first satellite (100) and the ground station (300), the second ATP device (120) is configured to emit a laser in a direction away from the earth so as to establish laser communication between the first satellite (100) and the second satellite (200) together with the third ATP device (210), and the third ATP device (210) is configured to emit a laser toward the earth so as to enable the second satellite (200) to establish laser communication with the first satellite (100) and / or the ground station (300). Before the corresponding first satellite (100) needs to transmit the collected remote sensing data to the ground station (300), the corresponding first satellite (100) sends a transmission time comparison request to the corresponding second satellite (200); In response to the transmission time comparison request, the corresponding second satellite (200) determines the estimated time of the first transmission path and the second transmission path for the corresponding first satellite (100) based at least on meteorological conditions, and the first satellite (100) selects a transmission path from the first transmission path and the second transmission path to transmit remote sensing data according to the estimated time. The first transmission path is a laser communication link established between the corresponding first satellite (100) directly and a ground station (300) receiving remote sensing data, and the second transmission path is a laser communication link established between the corresponding first satellite (100) and a ground station (300) receiving remote sensing data indirectly via a corresponding second satellite (200).

13. The distributed remote sensing satellite system according to claim 12, characterized in that: After the corresponding first satellite (100) sends a transmission time comparison request to the corresponding second satellite (200), the corresponding second satellite (200) determines the estimated time of the first transmission path and the second transmission path based at least on the position information of the corresponding first satellite (100), the data transceiving capability of the corresponding first satellite (100), the position information of the ground station (300) receiving remote sensing data, the data transceiving capability of the ground station (300) receiving remote sensing data, the position information of the second satellite (200), the data transceiving capability of the second satellite (200) and meteorological conditions.

14. The distributed remote sensing satellite system according to claim 12 or 13, characterized in that: When the corresponding second satellite (200) determines the estimated time of the first transmission path and the second transmission path, the meteorological GIS platform (220) of the corresponding second satellite (200) periodically acquires meteorological data to perform meteorological condition simulation according to the meteorological data. When the meteorological GIS platform (220) of the corresponding second satellite (200) performs a meteorological condition simulation, the meteorological GIS platform (220) of the corresponding second satellite (200) performs a simulation on meteorological elements that vary with the first transmission path and the second transmission path. The corresponding second satellite (200) determines the simulated positions of the corresponding first satellite (100), the ground station (300) receiving the remote sensing data and the second satellite (200) in a meteorological GIS platform (220) based on the position information of the corresponding first satellite (100), the position information of the ground station (300) receiving the remote sensing data and the position information of the second satellite (200), Furthermore, the meteorological GIS platform (220) of the corresponding second satellite (200) also dynamically simulates the movement of the corresponding first satellite (100) according to time changes, so that the corresponding second satellite (200) determines the estimated time consumption of the first transmission path and the second transmission path when transmitting remote sensing data based on the meteorological condition simulation and the movement of the corresponding first satellite (100) and sends the estimated time consumption to the corresponding first transmission path (100), and the corresponding first satellite (100) selects one of the transmission paths to transmit remote sensing data based on at least the estimated time consumption of the first transmission path and the second transmission path.

Images