A terminal device and system for low-orbit satellite positioning and image recognition

Abstract

The invention relates to the field of satellite positioning, and discloses a terminal device and system for low-orbit satellite positioning and image recognition, which are used to solve the problem that when a constellation is used for information collection, the number of satellites in the constellation cannot be adjusted in time due to environmental interference. The method comprises: capturing a cloud image under the satellite, calculating and analyzing an occlusion influence coefficient according to the cloud image, collecting the speed of the satellite in the constellation, and calculating and analyzing the speed influence coefficient according to the speed of the satellite, collecting performance index data of the satellite in the constellation, and calculating and analyzing an aging influence coefficient according to the performance index data, comprehensively evaluating the occlusion influence coefficient, the speed influence coefficient and the aging influence coefficient to obtain an accuracy index, dynamically adjusting the number of satellites in the constellation according to the accuracy index, returning to an image recognition module, and continuously updating the image, thereby effectively improving the accuracy of satellite information collection.

Description

Technical Field

[0001] The present invention relates to the field of satellite positioning, and more specifically to a terminal device and system for low-orbit satellite positioning and image recognition. Background Art

[0002] Low-orbit satellite positioning is a technology that provides precise geographic location services through satellites operating in low-Earth orbit (usually 200 to 2,000 kilometers). These satellites are close to the Earth and can quickly and accurately determine the location of targets. They are often used in navigation, remote sensing, communications and other fields, with wide coverage, low signal delay and high accuracy.

[0003] Low-orbit satellite positioning and image recognition is a system that combines the precise positioning services provided by low-orbit satellites with high-resolution image analysis technology. It can obtain the geographic location of the target in real time and perform image recognition and analysis. This technology is widely used in disaster emergency response, environmental monitoring, military reconnaissance and other fields, providing accurate spatial information and intelligent image processing results to support decision-making and action.

[0004] Existing low-orbit satellite positioning and image recognition usually receive the positioning data of ground targets through low-orbit satellites and calculate their precise positions through satellite signals; then collect high-resolution images of the target area; then use image recognition algorithms to analyze the collected images to identify specific targets or extract useful information; finally, transmit the positioning and recognition results to the ground control center or user terminal to support real-time decision-making and operations. And because a satellite can be visible for a short time in a certain area, multiple satellites are required to form a constellation to ensure continuous coverage in the area. However, this positioning and recognition method will result in the inability to adjust the number of satellites in the constellation in time when there is environmental interference, resulting in inaccurate positioning and recognition results.

[0005] In view of the above problems, the present invention proposes a solution. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, the present invention provides a terminal device and system for low-orbit satellite positioning and image recognition to solve the problems existing in the above-mentioned background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A terminal system for low-orbit satellite positioning and image recognition, the system comprising:

[0009] An image recognition module is used to capture high-resolution cloud images under the satellite through a high-definition camera and transmit the cloud images to the occlusion analysis module;

[0010] The occlusion situation analysis module is used to calculate and analyze the occlusion influence coefficient based on the cloud image collected by the image recognition module, and transmit the occlusion influence coefficient to the data evaluation module;

[0011] The speed impact analysis module is used to collect the speeds of the satellites in the constellation, calculate the speed impact coefficient based on the speed analysis of the satellites, and transmit the speed impact coefficient to the data evaluation module;

[0012] A functional aging analysis module is used to collect performance index data of satellites in the constellation, wherein the performance index data includes signal strength and data transmission rate, and calculate and obtain an aging influence coefficient based on the performance index data, and transmit the aging influence coefficient to the data evaluation module;

[0013] A data evaluation module is used to receive the occlusion influence coefficient, speed influence coefficient and aging influence coefficient transmitted by the occlusion situation analysis module, the speed influence analysis module and the function aging analysis module, and to comprehensively evaluate the occlusion influence coefficient, the speed influence coefficient and the aging influence coefficient to obtain an accuracy index, and transmit the accuracy index to the satellite quantity adjustment module;

[0014] The satellite quantity adjustment module is used to dynamically adjust the number of satellites in the constellation according to the accuracy index and return it to the image recognition module to continuously update the image.

[0015] Preferably, the step of calculating and analyzing the occlusion influence coefficient based on the cloud layer image collected by the image recognition module is:

[0016] Perform geometric and radiometric corrections on cloud images captured by satellites, and use image processing techniques to classify different cloud types in cloud images;

[0017] The cloud layer is distinguished from the cloud-free area using the threshold segmentation technique, and the cloud coverage of the cloud image is calculated;

[0018] Estimate cloud thickness and height through multispectral image analysis;

[0019] The optical thickness of the cloud layer is calculated using satellite reflectivity data, and the calculation formula is: Where τ represents the optical thickness of the cloud layer, R observed Expressed as the cloud reflectivity observed by satellite, R clear It is expressed as the surface reflectivity when there is no cloud;

[0020] The occlusion influence coefficient is calculated based on the cloud coverage, cloud thickness, cloud height and optical thickness, and its calculation formula is: Where OL is the occlusion influence coefficient, C f Expressed as cloud cover ratio, C hExpressed as cloud thickness, C t It is expressed as the cloud type coefficient. Different types of clouds have different effects on the signal. o Expressed as cloud optical thickness, H s Expressed as the altitude of the satellite.

[0021] Preferably, the step of using a threshold segmentation technique to distinguish clouds from cloud-free areas and calculating the cloud coverage of the cloud image is:

[0022] The cloud image is converted into a grayscale image and the threshold is determined using Otsu's adaptive thresholding algorithm;

[0023] Apply the threshold to segment the grayscale image, mark the area with pixel values ​​higher than the threshold as cloud layer, and mark the area with pixel values ​​lower than the threshold as cloudless area. The specific operation expression is: Among them I binary (x, y) is the pixel value of the known image, I gray (x, y) represents the pixel value of the grayscale image, and T is the determined threshold;

[0024] Count the number of cloud pixels in the image and calculate the total number of pixels in the image. Calculate the cloud coverage rate based on the ratio of the number of cloud pixels to the total number of pixels. The calculation formula is: Where Or represents the cloud coverage rate, N cloud It is expressed as the number of pixels marked as clouds in the binary image, N tota Expressed as the total number of pixels in the image.

[0025] Preferably, the step of estimating the thickness and height of the cloud layer by multispectral image analysis is:

[0026] Select the spectral band and calculate the cloud reflectivity of each band. The calculation formula is: Where R(λ) represents the cloud reflectivity at wavelength λ, I cloud (λ) represents the radiation value of the cloud image, I ref (λ) represents the radiation value of the reference image;

[0027] Calculate the temperature difference between the top and bottom of the cloud layer, and calculate the cloud thickness based on the temperature difference. The calculation formula is: Where T cloud Expressed as cloud thickness, T top Expressed as the cloud top temperature, T bottom It is represented as the temperature at the bottom of the cloud layer, and L is represented as the vertical gradient of the atmospheric temperature;

[0028] The cloud height is calculated based on the cloud top temperature and ground temperature. The calculation formula is: Among them, H cloudExpressed as cloud height, T surface Expressed as ground temperature.

[0029] Preferably, the step of calculating and obtaining the speed influence coefficient based on the satellite speed analysis is:

[0030] Collect the orbital velocity data of each satellite in the constellation, and obtain the satellite's operating speed that has the least impact on the accuracy of the collected data, which is recorded as the optimal operating speed;

[0031] For each satellite, the speed impact degree is obtained by calculating the ratio of its actual speed to the optimal operating speed;

[0032] The speed influence coefficient is calculated by weighting the speed influence degree of each satellite and taking the average value. The calculation formula is: Where SD is the speed influence coefficient, n is the number of satellites in the current constellation, S i It is expressed as the velocity influence of the i-th satellite.

[0033] Preferably, the step of analyzing and calculating the aging influence coefficient based on the performance indicator data is:

[0034] Obtain the initial signal strength and initial data transmission rate of each satellite at the time of launch, and obtain the current signal strength and current data transmission rate of each satellite;

[0035] Calculate the aging degree of each satellite signal strength, and the calculation formula is: Where As represents the aging degree of signal strength, S initial Expressed as the initial signal strength, S current Indicates the current signal strength;

[0036] Calculate the degree of decrease in data transmission rate of each satellite, and the calculation formula is: Ar is the degree of data transmission rate reduction, R initial Denoted as the initial data transmission rate, R current Indicates the current data transmission rate;

[0037] The aging degree is calculated based on the aging degree of the satellite signal strength and the degree of data transmission rate decline. The calculation formula is: Where Lh represents the degree of aging;

[0038] The aging degree of each satellite in the constellation is summed up and the average is calculated to obtain the aging influence coefficient, which is calculated as follows: Where AG is the aging influence coefficient, Lh i It is represented as the aging degree of the i-th satellite.

[0039] Preferably, the step of comprehensively evaluating the occlusion influence coefficient, the speed influence coefficient and the aging influence coefficient to obtain the accuracy index is:

[0040] The accuracy index is calculated by weighted summing the occlusion influence coefficient, speed influence coefficient and aging influence coefficient. The calculation formula is: Wherein, PC represents the accuracy index, OL represents the occlusion influence coefficient, SD represents the speed influence coefficient, AG represents the aging influence coefficient, and a1, a2, and a3 represent the weight coefficients of the occlusion influence coefficient, the speed influence coefficient, and the aging influence coefficient.

[0041] Preferably, the step of dynamically adjusting the number of satellites in the constellation according to the accuracy index is:

[0042] The accuracy index is compared with a preset threshold. If the accuracy index is greater than the preset threshold, it is determined that the current constellation has high accuracy and the number of satellites in the constellation is not adjusted.

[0043] If the accuracy index is less than a preset threshold, the accuracy of the current constellation is determined to be low, and the number of satellites in the constellation is adjusted;

[0044] The steps for adjusting the number of satellites in the constellation are:

[0045] The adjustment coefficient is calculated by comparing the preset threshold with the accuracy index. The calculation formula is: Where Ad represents the adjustment coefficient, PC represents the accuracy index, and PC0 represents the preset threshold;

[0046] The original number of satellites in the constellation is specified, and the adjustment coefficient is multiplied by the original number of satellites to obtain the actual number of satellites required. The calculation formula is NUM 实际 =Ad×NUM0, where NUM 实际 It represents the number of satellites actually needed, Ad represents the adjustment factor, and NUM0 represents the original number of satellites;

[0047] The number of satellites in the constellation is adjusted according to the calculated number of satellites actually required.

[0048] Preferably, a terminal device for low-orbit satellite positioning and image recognition comprises:

[0049] GNSS receiver, used to receive and decode positioning signals from low-orbit satellites;

[0050] High-definition camera, used to capture high-resolution images or videos, and supports multi-spectral imaging to obtain more environmental information;

[0051] Image recognition processors for rapid image processing and analysis;

[0052] A communication device used to communicate with low-orbit satellites to transmit positioning and image data.

[0053] Technical effects and advantages of the present invention:

[0054] Capture the cloud image under the satellite, calculate and analyze the occlusion impact coefficient based on the cloud image, collect the speed of the satellites in the constellation, and calculate the speed impact coefficient based on the satellite speed analysis, collect the performance indicator data of the satellites in the constellation, and calculate the aging impact coefficient based on the performance indicator data analysis, conduct a comprehensive evaluation of the occlusion impact coefficient, speed impact coefficient and aging impact coefficient to obtain the accuracy index, dynamically adjust the number of satellites in the constellation based on the accuracy index, and return it to the image recognition module to continuously update the image, effectively improving the accuracy of satellite information collection. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is the overall flow chart of the present invention. DETAILED DESCRIPTION

[0056] The technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the present invention. In addition, the forms of the various structures recorded in the following embodiments are merely illustrative. The terminal device and system for low-orbit satellite positioning and image recognition involved in the present invention are not limited to the various structures recorded in the following embodiments. All other implementations obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0057] The present invention provides a terminal system for low-orbit satellite positioning and image recognition, the system comprising:

[0058] An image recognition module is used to capture high-resolution cloud images under the satellite through a high-definition camera and transmit the cloud images to the occlusion analysis module;

[0059] In this embodiment, it should be specifically explained that the steps of capturing a high-resolution cloud image below the satellite by a high-definition camera are:

[0060] According to the needs of the target area, adjust the satellite's orbital parameters (such as altitude and inclination) to ensure coverage of the target area. Choose the appropriate transit time to capture the best state of the cloud layer. Before capturing images, calibrate the high-definition camera on the satellite to ensure that the focus, exposure and other imaging parameters are suitable for the current lighting conditions and target altitude;

[0061] Using the satellite’s attitude control function, the camera is directed to the target area below the Earth’s surface, ensuring that the satellite’s attitude adjustment and stabilization systems will ensure that the camera is always directed to the target area to avoid image blur; the high-definition camera is used to continuously capture cloud images to avoid motion blur caused by the high-speed movement of the satellite, and the cloud image data is temporarily stored in the satellite’s storage device;

[0062] When the satellite passes through a ground station or relay satellite, a communication link is established to transmit the cloud image to the ground receiving station through the satellite communication link. In order to effectively utilize the bandwidth, the image data is further compressed before transmission and encrypted as required to ensure that the data is not tampered with or stolen during transmission.

[0063] After receiving the image data from the satellite, the ground station decodes and decompresses it, and then stores the cloud images in the ground data center.

[0064] The occlusion situation analysis module is used to calculate and analyze the occlusion influence coefficient based on the cloud image collected by the image recognition module, and transmit the occlusion influence coefficient to the data evaluation module;

[0065] In this embodiment, it should be specifically explained that the step of calculating and analyzing the occlusion influence coefficient based on the cloud layer image collected by the image recognition module is as follows:

[0066] Perform geometric and radiometric corrections on cloud images captured by satellites to ensure that the image data matches the actual geographic location and optical characteristics. Use image processing technology to classify different cloud types in cloud images (such as cumulus, stratus, cirrus, etc.) to determine the structure and distribution of clouds.

[0067] The geometric correction and radiation correction are two key steps to accurately process cloud images captured by satellites. Geometric correction aims to correct the geometric distortion caused by the satellite position, attitude and curvature of the earth in the image, so that the features of the objects in the image are consistent with the actual geographical location. Radiation correction adjusts the brightness and color information of the image, eliminates the radiation error caused by atmospheric scattering, sensor noise or illumination changes, and ensures that the spectral characteristics of the image can truly reflect the reflection characteristics of ground objects.

[0068] The cloud layer is distinguished from the cloud-free area by using the threshold segmentation technique, and the cloud coverage of the cloud image is calculated. The threshold segmentation technique is a basic and widely used image processing method, which is used to classify the pixels in the image into two or more categories. This technique sets one or more thresholds and compares the pixel values ​​of the image with these thresholds to distinguish the target area from the background;

[0069] The thickness and height of clouds are estimated through multispectral image analysis, which is a technique for extracting information by capturing and analyzing image data from different spectral bands. This method uses image data acquired in multiple spectral bands (such as visible light, near infrared, and mid-infrared) to enhance the recognition and analysis capabilities of target object features. In cloud analysis, multispectral image analysis can use the spectral characteristics of different bands to distinguish different types, thicknesses, and heights of clouds, because clouds have different reflection and transmission characteristics in different bands;

[0070] The optical thickness of the cloud layer is estimated and calculated using satellite reflectivity data. The calculation formula is: Where τ represents the optical thickness of the cloud layer, R observed Expressed as the cloud reflectivity observed by satellite, R clear It is expressed as the surface reflectivity when there is no cloud, usually referring to the reflectivity when it is sunny or there are no clouds;

[0071] The occlusion influence coefficient is calculated based on the cloud coverage, cloud thickness, cloud height and optical thickness, and its calculation formula is: Where OL is the occlusion influence coefficient, C f Expressed as cloud cover ratio, C h Expressed as cloud thickness, C t It is expressed as the cloud type coefficient. Different types of clouds have different effects on the signal. o Expressed as cloud optical thickness, H s Expressed as the height of the satellite, this formula shows that the occlusion influence coefficient increases with the increase of cloud amount, cloud thickness and optical thickness. At the same time, the higher the satellite altitude, the smaller the occlusion influence coefficient.

[0072] The reason why the higher the satellite altitude, the smaller the occlusion coefficient is that as the satellite altitude increases, its viewing angle of the earth's surface becomes larger. High-orbit satellites can cover a wider area, so the impact of cloud occlusion will be diluted in the overall image. Low-orbit satellites are more sensitive to local occlusion due to their smaller viewing angles, and the relative proportion of clouds in the satellite's field of view is smaller in higher orbits. Even if clouds have an impact on the signals of high-orbit satellites, this impact accounts for a relatively low proportion of the overall data, so the occlusion coefficient is smaller; in high orbits, the geometric impact of cloud occlusion is more dispersed. The projected area of ​​clouds from a high-orbit perspective is relatively small, so the actual occlusion effect relative to the earth's surface is reduced.

[0073] In this embodiment, it should be specifically explained that the steps of classifying different cloud types in the cloud image using image processing technology are as follows:

[0074] Calculate the histogram of each color channel to describe the color distribution of the cloud layer, and convert the image from RGB to other color spaces (such as HSV) to obtain characteristics such as color saturation and brightness;

[0075] Use edge detection algorithms (such as Canny, Sobel) to extract edge features of clouds, extract cloud contours and calculate their shape features, such as perimeter, area and shape factor;

[0076] Use global or adaptive threshold technology to perform preliminary segmentation on the image, separate the cloud layer from the background in the image, perform region growing on the preliminary segmented area, and extract the specific area of ​​the cloud layer;

[0077] Collect pictures of known cloud types and use the extracted features (color, texture, shape) to classify cloud images. Common classification algorithms include K-nearest neighbor (K-NN), support vector machine (SVM), decision tree, etc.

[0078] In this embodiment, it should be specifically explained that the step of using the threshold segmentation technology to distinguish the cloud layer from the cloud-free area and calculating the cloud coverage rate of the cloud layer image is as follows:

[0079] The cloud images were converted to grayscale images to simplify processing, and the threshold was determined using Otsu's adaptive thresholding algorithm;

[0080] Apply the threshold to segment the grayscale image, mark the area with pixel values ​​higher than the threshold as cloud layer, and mark the area with pixel values ​​lower than the threshold as cloudless area. The specific operation expression is: Among them I binary (x, y) is the pixel value of the known image, I gray (x, y) represents the pixel value of the grayscale image, and T is the determined threshold;

[0081] The number of cloud pixels in the binary image is counted, and the total number of pixels in the image is calculated. The cloud coverage is calculated based on the ratio of the number of cloud pixels to the total number of pixels. The calculation formula is: Where Or represents the cloud coverage rate, N cloud It is expressed as the number of pixels marked as clouds in the binary image, N total Expressed as the total number of pixels in the image.

[0082] In this embodiment, it should be specifically explained that the steps of estimating the thickness and height of the cloud layer by multispectral image analysis are as follows:

[0083] Select the spectral bands suitable for cloud analysis and calculate the cloud reflectance of each band. The calculation formula is: Where R(λ) represents the cloud reflectivity at wavelength λ, I cloud(λ) represents the radiation value of the cloud image, I ref (λ) represents the radiation value of the reference image;

[0084] The temperature difference between the top and bottom of the cloud layer is calculated by using the radiation values ​​of different bands obtained from multispectral images and the atmospheric model. The cloud thickness is calculated based on the temperature difference. The calculation formula is: Where T cloud Expressed as cloud thickness, T top Expressed as the cloud top temperature, T bottom It is expressed as the cloud bottom temperature, and L is the vertical gradient of atmospheric temperature, which is usually provided by the atmospheric model;

[0085] The cloud height is calculated based on the cloud top temperature and ground temperature. The calculation formula is: Among them, H cloud Expressed as cloud height, T surface Expressed as ground temperature.

[0086] The speed impact analysis module is used to collect the speeds of the satellites in the constellation, calculate the speed impact coefficient based on the speed analysis of the satellites, and transmit the speed impact coefficient to the data evaluation module;

[0087] The speeds of the satellites in the constellation are usually not exactly the same, and the satellites have different orbital altitudes and different operating speeds. Satellites in lower orbits are faster, and satellites in higher orbits are slower. This is because the lower the orbital altitude, the stronger the effect of the Earth's gravity on the satellite, requiring the satellite to maintain the orbit at a higher speed; if the satellites in the constellation are operating on different orbital shapes, such as circular orbits and elliptical orbits, the speed of the satellite will change at different locations in the orbit. In an elliptical orbit, the satellite runs faster at the perigee (the closest point to the Earth) and slower at the apogee (the farthest point from the Earth); in order to maintain the stability of the constellation or perform a mission, the satellite may perform fine-tuning operations on the orbit (such as raising or lowering the orbital altitude), which will cause speed changes in a short period of time; because the Earth is not completely spherical and has uneven mass distribution, the speed of the satellite in different orbits will be affected by slight differences in the Earth's gravitational field.

[0088] In this embodiment, it should be specifically explained that the step of calculating and obtaining the speed influence coefficient according to the satellite speed analysis is as follows:

[0089] Collect orbital velocity data of each satellite in the constellation, usually in kilometers per second (km / s), and obtain the satellite's operating speed that has the least impact on the accuracy of the collected data, which is recorded as the best operating speed;

[0090] For each satellite, the speed impact degree is obtained by calculating the ratio of its actual speed to the optimal operating speed;

[0091] The speed influence coefficient is calculated by weighting the speed influence degree of each satellite and taking the average value. The calculation formula is: Where SD is the speed influence coefficient, n is the number of satellites in the current constellation, S i It is expressed as the velocity influence of the i-th satellite.

[0092] A functional aging analysis module is used to collect performance index data of satellites in the constellation, wherein the performance index data includes signal strength and data transmission rate, and calculate and obtain an aging influence coefficient based on the performance index data, and transmit the aging influence coefficient to the data evaluation module;

[0093] The lifespan of a low-orbit satellite is relatively short, usually between 5 and 7 years. Over time, satellites may gradually fail or degrade in function; there are strong earth radiation belts in the orbital environment, and satellites are exposed to high-energy particles and radiation for a long time, which can damage the satellite's electronic components and gradually cause their performance to degrade. Satellites rely on solar panels for energy. Over time, the efficiency of solar panels will gradually decrease. At the same time, the performance of the battery energy storage system will also degrade, causing the satellite to be unable to maintain normal operation. Over time, fuel consumption, reaction wheel wear or sensor failure may cause attitude control failure. Various sensors on satellites (such as cameras, imaging equipment, measuring instruments) and communication equipment will gradually age with the increase in years of use, resulting in a decrease in data collection quality or communication interruption; due to sensor aging or unstable attitude control, the positioning accuracy of the satellite may gradually decrease, affecting the quality of positioning services.

[0094] In this embodiment, it should be specifically explained that the step of calculating and analyzing the aging influence coefficient according to the performance indicator data is as follows:

[0095] Obtain the initial signal strength and initial data transmission rate of each satellite at the time of launch, and obtain the current signal strength and current data transmission rate of each satellite;

[0096] Calculate the aging degree of each satellite signal strength, and the calculation formula is: Where As represents the aging degree of signal strength, S initial Expressed as the initial signal strength, S current Indicates the current signal strength;

[0097] Calculate the degree of decrease in data transmission rate of each satellite, and the calculation formula is: Ar is the degree of data transmission rate reduction, R initial Denoted as the initial data transmission rate, R current Indicates the current data transmission rate;

[0098] The aging degree is calculated based on the aging degree of the satellite signal strength and the degree of data transmission rate decline. The calculation formula is: Where Lh represents the degree of aging;

[0099] The aging degree of each satellite in the constellation is summed up and the average is calculated to obtain the aging influence coefficient, which is calculated as follows: Where AG is the aging influence coefficient, Lh i It is represented as the aging degree of the i-th satellite.

[0100] A data evaluation module is used to receive the occlusion influence coefficient, speed influence coefficient and aging influence coefficient transmitted by the occlusion situation analysis module, the speed influence analysis module and the function aging analysis module, and to comprehensively evaluate the occlusion influence coefficient, the speed influence coefficient and the aging influence coefficient to obtain an accuracy index, and transmit the accuracy index to the satellite quantity adjustment module;

[0101] In this embodiment, it should be specifically explained that the step of comprehensively evaluating the occlusion influence coefficient, the speed influence coefficient and the aging influence coefficient to obtain the accuracy index is:

[0102] The accuracy index is calculated by weighted summing the occlusion influence coefficient, speed influence coefficient and aging influence coefficient. The calculation formula is: PC is the accuracy index, OL is the occlusion influence coefficient. As the signal occlusion factors in the environment increase (such as the density and height of buildings, mountains, vegetation, etc.), the positioning and image acquisition accuracy of the satellite will decrease accordingly. A high occlusion influence coefficient means that the path of the signal through obstacles is complex, the signal quality is reduced, and the multipath effect is enhanced, which leads to an increase in the deviation or loss of the collected data. As a result, the accuracy index of satellite acquisition decreases, that is, the accuracy, reliability and information level of positioning and image recognition decrease. SD is the speed influence coefficient. As the satellite speed changes, especially when the speed changes significantly under orbit adjustment or different orbit shapes (such as elliptical orbits), the positioning and image acquisition accuracy of the satellite will decrease. Factors such as the Doppler effect, orbit instability, and time synchronization error caused by speed changes will make it impossible for the satellite to accurately measure the position of the target or obtain clear images, thereby reducing the accuracy index of the collected data. AG is the aging influence coefficient. As the satellite's on-orbit time increases and its equipment gradually ages, the satellite's acquisition accuracy and data reliability will gradually decrease. A high aging influence coefficient means that the key components of the satellite (such as sensors, antennas, energy systems, attitude control systems, etc.) may experience performance degradation, increased failure rates, data transmission delays or errors, and other problems, thereby affecting the satellite's positioning accuracy and image acquisition quality. This inverse relationship shows that the older the satellite, the lower the acquisition accuracy index, the greater the error and uncertainty of the data, a1, a2, a3 are expressed as the weight coefficients of the occlusion influence coefficient, the speed influence coefficient, and the aging influence coefficient. The specific values ​​of a1, a2, a3 are determined by professionals based on actual conditions, and a1+a2+a3=1. For example, a1, a2, a3 can be 0.4, 0.4, and 0.2.

[0103] The satellite quantity adjustment module is used to dynamically adjust the number of satellites in the constellation according to the accuracy index and return it to the image recognition module to continuously update the image.

[0104] In this embodiment, it should be specifically explained that the step of dynamically adjusting the number of satellites in the constellation according to the accuracy index is:

[0105] The accuracy index is compared with a preset threshold. If the accuracy index is greater than the preset threshold, it is determined that the current constellation has a high accuracy and the number of satellites in the constellation is not adjusted.

[0106] If the accuracy index is less than the preset threshold, it is determined that the accuracy of the current constellation is low, and the number of satellites in the constellation is adjusted to ensure the accuracy of the satellite collected data;

[0107] The steps for adjusting the number of satellites in the constellation are:

[0108] The adjustment coefficient is calculated by comparing the preset threshold with the accuracy index. The calculation formula is: Where Ad represents the adjustment coefficient, PC represents the accuracy index, and PC0 represents the preset threshold;

[0109] The original number of satellites in the constellation is specified, and the adjustment coefficient is multiplied by the original number of satellites to obtain the actual number of satellites required. The calculation formula is NUM 实际 =Ad×NUM0, where NUM 实际 It represents the number of satellites actually needed, Ad represents the adjustment factor, and NUM0 represents the original number of satellites;

[0110] The number of satellites in the constellation is adjusted according to the calculated number of satellites actually required.

[0111] In this embodiment, it is necessary to specifically explain a terminal device for low-orbit satellite positioning and image recognition, the device comprising:

[0112] GNSS receiver, used to receive and decode positioning signals from low-orbit satellites;

[0113] High-definition camera, used to capture high-resolution images or videos, and supports multi-spectral imaging to obtain more environmental information;

[0114] Image recognition processors for rapid image processing and analysis;

[0115] Communication device, used to communicate with low-orbit satellites and transmit positioning and image data.

[0116] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

[0117] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A terminal system for low-orbit satellite positioning and image recognition, characterized in that: The system comprises: An image recognition module is used to capture high-resolution cloud images under the satellite through a high-definition camera and transmit the cloud images to the occlusion analysis module; The occlusion situation analysis module is used to calculate and analyze the occlusion influence coefficient based on the cloud image collected by the image recognition module, and transmit the occlusion influence coefficient to the data evaluation module; The speed impact analysis module is used to collect the speeds of the satellites in the constellation, calculate the speed impact coefficient based on the speed analysis of the satellites, and transmit the speed impact coefficient to the data evaluation module; A functional aging analysis module is used to collect performance index data of satellites in the constellation, wherein the performance index data includes signal strength and data transmission rate, and calculate and obtain an aging influence coefficient based on the performance index data, and transmit the aging influence coefficient to the data evaluation module; A data evaluation module is used to receive the occlusion influence coefficient, speed influence coefficient and aging influence coefficient transmitted by the occlusion situation analysis module, the speed influence analysis module and the function aging analysis module, and to comprehensively evaluate the occlusion influence coefficient, the speed influence coefficient and the aging influence coefficient to obtain an accuracy index, and transmit the accuracy index to the satellite quantity adjustment module; The satellite quantity adjustment module is used to dynamically adjust the number of satellites in the constellation according to the accuracy index and return to the image recognition module to continuously update the image; The steps of calculating and analyzing the occlusion influence coefficient based on the cloud layer image collected by the image recognition module are as follows: Perform geometric and radiometric corrections on cloud images captured by satellites, and use image processing techniques to classify different cloud types in cloud images; The cloud layer is distinguished from the cloud-free area using the threshold segmentation technique, and the cloud coverage of the cloud image is calculated; Estimate cloud thickness and height through multispectral image analysis; The optical thickness of the cloud layer is calculated using satellite reflectivity data, and the calculation formula is: Where τ represents the optical thickness of the cloud layer, R observed Expressed as the cloud reflectivity observed by satellite, R clear It is expressed as the surface reflectivity when there is no cloud; The occlusion influence coefficient is calculated based on the cloud coverage, cloud thickness, cloud height and optical thickness, and its calculation formula is: Where OL is the occlusion influence coefficient, C f Expressed as cloud cover ratio, C h Expressed as cloud thickness, C t It is expressed as the cloud type coefficient. Different types of clouds have different effects on the signal. o Expressed as cloud optical thickness, H s Expressed as the altitude of the satellite.

2. A low-orbit satellite positioning and image recognition terminal system according to claim 1, characterized in that: The threshold segmentation technique is used to distinguish the cloud layer from the cloud-free area, and the steps of calculating the cloud coverage rate of the cloud layer image are as follows: The cloud image is converted into a grayscale image and the threshold is determined using Otsu's adaptive thresholding algorithm; Apply the threshold to segment the grayscale image, mark the area with pixel values ​​higher than the threshold as cloud layer, and mark the area with pixel values ​​lower than the threshold as cloudless area. The specific operation expression is: Among them I binary (x, y) is the pixel value of the known image, I gray (x, y) represents the pixel value of the grayscale image, and T is the determined threshold; Count the number of cloud pixels in the image and calculate the total number of pixels in the image. Calculate the cloud coverage rate based on the ratio of the number of cloud pixels to the total number of pixels. The calculation formula is: Where Or represents the cloud coverage rate, N cloud It is expressed as the number of pixels marked as clouds in the binary image, N total Expressed as the total number of pixels in the image.

3. A low-orbit satellite positioning and image recognition terminal system according to claim 2, characterized in that: The steps of estimating the thickness and height of the cloud layer by multispectral image analysis are as follows: Select the spectral band and calculate the cloud reflectivity of each band. The calculation formula is: Where R(λ) represents the cloud reflectivity at wavelength λ, I cloud (λ) represents the radiation value of the cloud image, I ref (λ) represents the radiation value of the reference image; Calculate the temperature difference between the top and bottom of the cloud layer, and calculate the cloud thickness based on the temperature difference. The calculation formula is: Where T cloud Expressed as cloud thickness, T top Expressed as the cloud top temperature, T bottom It is represented as the temperature at the bottom of the cloud layer, and L is represented as the vertical gradient of the atmospheric temperature; The cloud height is calculated based on the cloud top temperature and ground temperature. The calculation formula is: Among them, H cloud Expressed as cloud height, T surface Expressed as ground temperature.

4. The terminal system for low-orbit satellite positioning and image recognition according to claim 1, characterized in that: The steps of calculating the speed influence coefficient based on the satellite speed analysis are as follows: Collect the orbital velocity data of each satellite in the constellation, and obtain the satellite's operating speed that has the least impact on the accuracy of the collected data, which is recorded as the optimal operating speed; For each satellite, the speed impact degree is obtained by calculating the ratio of its actual speed to the optimal operating speed; The speed influence coefficient is calculated by weighting the speed influence degree of each satellite and taking the average value. The calculation formula is: Where SD is the speed influence coefficient, n is the number of satellites in the current constellation, S i It is expressed as the velocity influence of the i-th satellite.

5. The terminal system for low-orbit satellite positioning and image recognition according to claim 1, characterized in that: The steps of analyzing and calculating the aging influence coefficient according to the performance index data are as follows: Obtain the initial signal strength and initial data transmission rate of each satellite at the time of launch, and obtain the current signal strength and current data transmission rate of each satellite; Calculate the aging degree of each satellite signal strength, and the calculation formula is: Where As represents the aging degree of signal strength, S initial Expressed as the initial signal strength, S current Indicates the current signal strength; Calculate the degree of decrease in data transmission rate of each satellite, and the calculation formula is: Ar is the degree of data transmission rate reduction, R initial Denoted as the initial data transmission rate, R current Indicates the current data transmission rate; The aging degree is calculated based on the aging degree of the satellite signal strength and the degree of data transmission rate decline. The calculation formula is: Where Lh represents the degree of aging; The aging degree of each satellite in the constellation is summed up and the average is calculated to obtain the aging influence coefficient, which is calculated as follows: Where AG is the aging influence coefficient, Lh i It is represented as the aging degree of the i-th satellite.

6. The terminal system for low-orbit satellite positioning and image recognition according to claim 1, characterized in that: The steps of comprehensively evaluating the occlusion influence coefficient, the speed influence coefficient and the aging influence coefficient to obtain the accuracy index are as follows: The accuracy index is calculated by weighted summing the occlusion influence coefficient, speed influence coefficient and aging influence coefficient. The calculation formula is: Wherein, PC represents the accuracy index, OL represents the occlusion influence coefficient, SD represents the speed influence coefficient, AG represents the aging influence coefficient, and a1, a2, and a3 represent the weight coefficients of the occlusion influence coefficient, the speed influence coefficient, and the aging influence coefficient.

7. The terminal system for low-orbit satellite positioning and image recognition according to claim 1, characterized in that: The step of dynamically adjusting the number of satellites in the constellation according to the accuracy index is as follows: The accuracy index is compared with a preset threshold. If the accuracy index is greater than the preset threshold, it is determined that the current constellation has high accuracy and the number of satellites in the constellation is not adjusted. If the accuracy index is less than a preset threshold, the accuracy of the current constellation is determined to be low, and the number of satellites in the constellation is adjusted; The steps for adjusting the number of satellites in the constellation are: The adjustment coefficient is calculated by comparing the preset threshold with the accuracy index. The calculation formula is: Where Ad represents the adjustment coefficient, PC represents the accuracy index, and PC0 represents the preset threshold; The original number of satellites in the constellation is specified, and the adjustment coefficient is multiplied by the original number of satellites to obtain the actual number of satellites required. The calculation formula is NUM 实际 =Ad×NUM0, where NUM 实际 It represents the number of satellites actually needed, Ad represents the adjustment factor, and NUM0 represents the original number of satellites; The number of satellites in the constellation is adjusted according to the calculated number of satellites actually required.

8. A terminal device for low-orbit satellite positioning and image recognition, used to implement the terminal system for low-orbit satellite positioning and image recognition as claimed in any one of claims 1 to 7, characterized in that: The device comprises: GNSS receiver, used to receive and decode positioning signals from low-orbit satellites; High-definition camera, used to capture high-resolution images or videos, and supports multi-spectral imaging to obtain more environmental information; Image recognition processors for rapid image processing and analysis; A communication device used to communicate with low-orbit satellites to transmit positioning and image data.

Images