Remote sensing image processing method and device and storage medium

Abstract

The invention discloses a remote sensing image processing method and device and a storage medium, relates to the field of remote sensing data processing, and is used for solving at least one of the problems of high requirement on satellite-ground link bandwidth, high satellite computing resource occupancy rate and poor data processing real-time performance. A remote sensing data processing task is received at a ground station, and satellites are scheduled for the remote sensing data processing task; the satellite and the ground station cooperate to complete a remote sensing data processing task; a first processing model used when the satellite processes the remote sensing data processing task is optimized on the ground station, and the optimized model parameters are at least uploaded to the satellite. According to the method, the requirement on the satellite-ground link bandwidth is reduced, the data processing real-time performance is improved, the satellite calculation efficiency is improved, and the service life is prolonged.

Description

Technical Field

[0001] The present invention relates to the field of remote sensing data processing, and in particular to a remote sensing image processing method, device and storage medium. Background Art

[0002] Remote sensing data is surface or atmospheric information collected by long-distance remote sensors. Remote sensing data reflects the various physical and chemical properties of the surface and is widely used in many fields such as environmental monitoring, disaster response, agriculture, and urban planning.

[0003] Compared with traditional manual analysis methods, the use of technologies such as artificial intelligence can greatly improve the efficiency of remote sensing data processing (such as target recognition, classification, etc.).

[0004] Currently, there are two broad approaches to processing remote sensing data using artificial intelligence: one is for satellites to collect remote sensing data and transmit it to ground stations for centralized processing; the other is for satellites to process the data onboard using their own data processing resources and then transmit the results to ground stations. The former, due to the massive daily increase in remote sensing data, places extremely high demands on bandwidth for downlink. Limited by current channel conditions, it is difficult to transmit all remote sensing data in real time, resulting in a serious data backlog, increased processing latency at ground stations, and low real-time processing performance. The latter, due to the satellite's size and heat dissipation requirements, limits the computing power of the processors onboard. Complex remote sensing data processing tasks require a long time, resulting in low processing efficiency, poor real-time performance, and limited processing accuracy. Furthermore, prolonged data processing consumes limited energy and causes rapid temperature rise, impacting the satellite's energy storage and service life. Summary of the Invention

[0005] The object of the present invention is to provide a ground station, satellite, satellite-ground collaboration method and system to address all or part of the above-mentioned problems, so as to solve at least one of the problems of high satellite-ground link bandwidth requirements, high satellite computing resource occupancy and poor real-time data processing.

[0006] The technical solution adopted in the present invention is as follows: A remote sensing image processing method, applied to a ground station, comprising: Receive remote sensing data processing tasks; Scheduling satellites to cooperate in processing remote sensing image data collected by the satellite based on the remote sensing data processing task includes: receiving first remote sensing image data transmitted by the satellite, and processing a processing result retained by the remote sensing data processing task, wherein the first remote sensing image data only includes remote sensing image data for which the processing result has not been transmitted by the satellite; and processing the first remote sensing image data according to the remote sensing data processing task; Optimizing a first processing model used by the satellite to process the remote sensing data processing task; The optimized model parameters are uploaded to at least the satellite.

[0007] In a second aspect, the present application also provides another remote sensing image processing method, which is applied to satellites, and the method includes: Receive dispatch instructions; According to the remote sensing data processing task indicated by the scheduling instruction, the first processing model is used to cooperate with the ground station to complete processing of the collected remote sensing image data, including: transmitting the first remote sensing image data and the processing results retained by the remote sensing data processing task to the ground station, wherein the first remote sensing image data only includes the remote sensing image data for which the processing results have not been transmitted; Receive model parameters after the ground station optimizes the first processing model.

[0008] In a third aspect, the present application also provides another remote sensing image processing method, which includes: Receive remote sensing data processing tasks; Controlling a ground station to dispatch a satellite to cooperate in processing remote sensing image data collected by the satellite based on the remote sensing data processing task, including: controlling the ground station to receive first remote sensing image data transmitted by the satellite, and processing a processing result retained by the remote sensing data processing task, wherein the first remote sensing image data only includes remote sensing image data for which the processing result has not been transmitted by the satellite; and controlling the ground station to process the first remote sensing image data according to the remote sensing data processing task; controlling the ground station to optimize a first processing model used by the satellite to process the remote sensing data processing task; The optimized model parameters are uploaded to at least the satellite.

[0009] According to the concept of the present application, the present application also provides a computer-readable storage medium storing computer instructions, and running the computer instructions can execute the above-mentioned remote sensing image processing method.

[0010] In addition, the present application also provides a remote sensing image processing device, which is applied to a ground station and includes: Information management module, used to receive remote sensing data processing tasks; a task scheduling module, configured to schedule satellites to cooperate in processing remote sensing image data collected by the satellites based on the remote sensing data processing task, including: instructing the scheduled satellites to downlink first remote sensing image data, and processing results retained by the remote sensing data processing task, wherein the first remote sensing image data only includes remote sensing image data for which the satellites have not downlinked processing results; a ground-side algorithm module, configured to process the first remote sensing image data according to a remote sensing data processing task, and optimize a first processing model used by the satellite to process the remote sensing data processing task; The ground communication module is used to receive the first remote sensing image data and upload the optimized model parameters to at least the satellite.

[0011] Furthermore, the present application also provides another remote sensing image processing device for use with a satellite, the device comprising: Data acquisition module, used for collecting remote sensing image data; The satellite-side algorithm module is configured to cooperate with the ground station to complete processing of the remote sensing image data using a first processing model according to the remote sensing data processing task indicated by the scheduling instruction, including: determining a processing result retained by the remote sensing data processing task and first remote sensing image data, where the first remote sensing image data only includes remote sensing image data for which the processing result is not retained; The satellite communication module is used to receive the scheduling instruction and the model parameters after the ground station optimizes the first processing model, and is also used to transmit the first remote sensing image data and the retained processing results to the ground station that sends the scheduling instruction.

[0012] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: This application dispatches satellites to collaborate (relay) with ground stations to handle remote sensing data processing tasks, eliminating the need for satellites to downlink all remote sensing image data to the ground station. This reduces the amount of data to be downlinked, lowers the bandwidth requirements for satellite-to-ground link communications, avoids data backlogs, and improves the real-time nature of data transmission and processing. This collaborative satellite-to-ground processing reduces the satellite's computational load, lowers its operating energy consumption, extends its operating hours and service life, avoids the impact of high satellite heat on its computing performance, and improves its computing efficiency. Furthermore, by optimizing the onboard primary processing model at the ground station, the satellite's computational load is reduced, while improving the utilization of the satellite's limited storage resources and the efficiency of updating model parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The present invention will now be described by way of example with reference to the accompanying drawings, in which: Figure 1 This is a flowchart of an embodiment of the remote sensing image processing method provided by the present application applied to a ground station.

[0014] Figure 2 This is a flowchart of another embodiment of the remote sensing image processing method provided by the present application applied to a ground station.

[0015] Figure 3The present application provides a flowchart of a remote sensing image processing method in an embodiment applied to a satellite.

[0016] Figure 4 This is a flowchart of another embodiment of the remote sensing image processing method provided by the present application applied to a satellite.

[0017] Figure 5 This is a flowchart of the remote sensing image processing method provided by this application in an embodiment of realizing satellite-ground coordinated control.

[0018] Figure 6 This is a data flow diagram of the remote sensing image processing method provided by this application in an embodiment for realizing satellite-ground coordinated control.

[0019] Figure 7 This is a structural diagram of a remote sensing image processing system provided by this application in one embodiment.

[0020] Figure 8 This is a structural diagram of another embodiment of the remote sensing image processing system provided by this application. DETAILED DESCRIPTION

[0021] All features disclosed in this specification, or all steps in the disclosed methods or processes, except mutually exclusive features and / or steps, can be combined in any manner.

[0022] Any feature disclosed in this specification (including any appended claims and abstract), unless otherwise stated, may be replaced by other equivalent or similar features. In other words, unless otherwise stated, each feature is only an example of a series of equivalent or similar features.

[0023] In response to the current problems of high satellite-to-ground link communication bandwidth requirements, poor real-time data processing, and high satellite computing resource utilization in remote sensing data processing, the embodiments of the present application provide a remote sensing image processing method, device, and storage medium, aiming to reduce the requirements for satellite-to-ground link communication bandwidth, improve data processing timeliness, and reduce satellite computing resource utilization.

[0024] In one embodiment applied to a ground station, Figure 1 As shown, the remote sensing image processing method provided in this application includes the following process: Receive remote sensing data processing tasks. Remote sensing data processing tasks indicate what type of remote sensing image data to process (i.e., what type of remote sensing image data needs to be collected) and what type of processing to perform on the remote sensing image data (i.e., the processing content of the remote sensing image data, such as classification, recognition, processing, etc.).

[0025] Satellites are dispatched based on this remote sensing data processing task to collaborate with ground stations to process the remote sensing image data. Since remote sensing image data is collected by satellites, the satellite typically performs the remote sensing data processing task first, followed by the ground station. This collaborative relationship can also be called a relay relationship, meaning that after the satellite completes the remote sensing data processing task, the ground station relays the task to the satellite.

[0026] Optimize a first processing model, such as a machine learning model, used by a satellite to process remote sensing data processing tasks, and upload the optimized model parameters to at least the satellite (i.e., the cooperating satellite).

[0027] By cooperating between satellites and ground stations to process remote sensing data processing tasks, the amount of data transmitted by satellites can be reduced, thereby alleviating the communication bandwidth requirements for the satellite-to-ground link while improving the real-time performance of data transmission, thereby improving the real-time performance of remote sensing data processing tasks.

[0028] The first processing model configured on the satellite is a machine learning model, such as a basic image classification model or image recognition model. The ground station updates the model parameters, reducing and improving the utilization of satellite computing resources and increasing the efficiency of first processing model updates. The initial model parameters of the first processing model can also be obtained by training the initialized first processing model using sample data (such as historical remote sensing image data) at the ground station.

[0029] Remote sensing data processing tasks can include tasks such as recognizing or classifying remote sensing image data. These tasks are typically initiated by users of ground stations, who (via appropriate terminal devices) send remote sensing data processing tasks to the ground station. Accordingly, the results of these tasks are ultimately fed back to the users by the ground station.

[0030] As an optional implementation, the first processing model deployed on the satellite is a lightweight model. Compared to a full machine learning model, a lightweight model significantly reduces the amount of parameter data, resulting in a corresponding performance tradeoff. This allows the satellite to rapidly perform preliminary processing of remote sensing image data using limited computing resources, screening out the first remote sensing image data that requires relay processing by the ground station, thereby improving overall mission processing efficiency and real-time performance. It also conserves satellite energy and prevents problems caused by excessive heating.

[0031] In one optional implementation, satellite scheduling is based on the current status information of each satellite in the constellation to which the ground station is connected, including attributes (such as orbit number, altitude, hardware configuration) and operating status information (such as load, operating temperature, and energy consumption). Based on the requirements of the remote sensing data processing task, at least one satellite in the constellation is selected for scheduling. By selecting satellites for task scheduling, tasks can be executed most efficiently, ensuring real-time processing.

[0032] In some possible implementations, satellite status information may include attributes such as orbital position, altitude, status, available resources (computing power, energy conditions, etc.), and the requirements of remote sensing data processing tasks for satellites may be requirements for some or all attributes of satellite status information.

[0033] For example, methods for scheduling satellites include: 1) Filtering. Based on the attributes required for the remote sensing data processing task, satellites in the constellation that do not meet the necessary attributes are filtered out. For example, the remote sensing data processing task may require satellites in a specific orbit or altitude; require satellites to have a certain external device; require satellites to be within a certain geographic location range; or other conditions are required by the remote sensing data processing task. Satellites that do not meet all the necessary attribute requirements are filtered out based on the task's requirements.

[0034] 2) Best Selection. Among the filtered satellites, scores are performed based on common attributes, and one or more (i.e., two or more) satellites with the highest scores are selected. For example, each satellite attribute is scored based on CPU, memory, disk idleness, temperature, and energy availability, with higher scores assigned to better attributes. Finally, the satellite or satellites with the highest overall scores are selected for scheduling to achieve load balancing.

[0035] In addition, after scheduling a satellite, if it no longer meets the requirements of the remote sensing data processing mission, other satellites will be rescheduled.

[0036] Considering that a ground station is connected to more than one satellite in a constellation, different satellites may be dispatched for processing remote sensing data received subsequently. However, the remote sensing image data processing method remains the same. Therefore, in one optional implementation, the optimized model parameters (at the ground station) are uploaded to all satellites in the constellation with which it can communicate. This allows for batch updating of the model parameters of the first processing model on all satellites after optimizing the first processing model, thereby improving the efficiency and accuracy of satellite remote sensing data processing. For example, after the ground station optimizes the first processing model, when dispatching other satellites to perform the next remote sensing data processing task, the optimized model parameters are uploaded to that satellite to update the first processing model, and then that satellite processes the remote sensing data task.

[0037] For satellite-to-ground relay processing, in an optional implementation, as Figure 2 As shown, the method for scheduling satellites to cooperate in completing the processing of collected remote sensing image data includes: Receive first remote sensing image data transmitted by a satellite, and the processing results retained after the satellite processes the remote sensing data processing task. The first remote sensing image data at least includes remote sensing image data for which the processing results have not been transmitted by the satellite; The first remote sensing image data is processed according to the remote sensing data processing task.

[0038] The satellite first processes the remote sensing data processing task (using the first processing model) to obtain all processing results of all remote sensing image data corresponding to the task, and then screens out some processing results based on corresponding screening conditions (such as credibility, accuracy, etc.) and retains them, and discards the remaining processing results; the retained processing results and the first remote sensing image data corresponding to the discarded processing results are transmitted to the ground station, and then the ground station relays the processing of the first remote sensing image data according to the remote sensing data processing task.

[0039] Taking the example of the ground station feeding back the processing results of the remote sensing data processing task to the user in the previous embodiment, the ground station fuses the processing results transmitted by the satellite and the processing results of the first remote sensing image data by the ground station, and feeds them back to the user as the final processing results of the remote sensing data processing task.

[0040] Regarding the processing of the first remote sensing image data, in some specific embodiments, the first remote sensing image data is processed at the ground station using a configured second processing model. The second processing model can also be a machine learning model, and can also be trained using sample data to obtain an initial second processing model.

[0041] Regarding the initial configuration of the first processing model, in some specific embodiments, the first processing model is obtained by pruning the second processing model, so as to run a lightweight model on the satellite side, save satellite energy consumption and computing power consumption, extend the satellite's working time and service life, and avoid excessive temperature rise.

[0042] As an optional implementation for optimizing the first processing model, the first processing model is optimized at the ground station based on at least the received first remote sensing image data.

[0043] For example, when the satellite-to-ground link lacks redundancy, the satellite transmits only the first remote sensing image data to the ground station. The ground station processes the first remote sensing image data and, based on the processing results, optimizes the first processing model using the first remote sensing image data as samples and the processing results as labels, obtaining optimized model parameters. However, when the satellite-to-ground link has redundancy, the satellite can, in addition to transmitting the first remote sensing image data, also transmit the remaining (or as much as possible) remote sensing image data for the remote sensing data processing task to the ground station. The ground station uses all received remote sensing image data as samples and the processing results as labels to optimize the first processing model and obtain optimized model parameters. Optimization of the first processing model can involve either direct optimization of its parameters or optimization of the second processing model followed by pruning to obtain the optimized first processing model. The pruned model parameters are then used as the optimized model parameters. Optimizing the first processing model using real-time processed remote sensing image data improves the accuracy of the first processing model without consuming additional storage resources.

[0044] In some embodiments, as Figure 3 As shown, the present application provides another remote sensing image processing method, which is applied to satellites and includes the following steps: Receive dispatch instructions, which are usually sent by the ground station.

[0045] In response to receiving the scheduling instruction, the satellite uses the first processing model to cooperate with the ground station to complete the processing of the collected remote sensing image data according to the remote sensing data processing task indicated by the scheduling instruction, thereby relaying the completion of the remote sensing data processing task. The remote sensing image data is collected by the satellite according to the instructions of the remote sensing data processing task, such as the area to be collected, the type of data, the amount of data to be collected, and the frequency of collection.

[0046] The receiving ground station optimizes the model parameters of the first processing model.

[0047] The so-called scheduling instructions, in an optional embodiment, are generated by the ground station based on the received remote sensing data processing task. The scheduling instructions carry an indication of the remote sensing image data required for the remote sensing data processing task, and the satellite obtains the collected remote sensing image data from the corresponding remote sensor according to the instruction. As for the way in which the ground station schedules satellites according to the remote sensing data processing task, please refer to the previous embodiment and will not be repeated here. Through the relay processing method between the satellite and the ground station, the requirements for the satellite-to-ground link communication bandwidth can be reduced, the data processing efficiency can be improved, the satellite energy consumption and resource utilization rate can be reduced, the temperature can be prevented from being too high, and the satellite working time and service life can be extended. The optimization of model parameters by the ground station can further reduce the resource utilization rate of the satellite and improve the update efficiency of the first processing model.

[0048] As an optional implementation, the first processing model on the satellite is a lightweight model, for example, see the lightweight model obtained by pruning the full machine learning model in the previous embodiment.

[0049] As an optional implementation, Figure 4 As shown, the method for processing the collected remote sensing image data in cooperation with the ground station includes: The first remote sensing image data and the processing results retained after the remote sensing data processing task are transmitted to the ground station. The first remote sensing image data only includes the remote sensing data for which the processing results have not been transmitted.

[0050] In some specific embodiments, the satellite selects, based on a preconfigured confidence or accuracy threshold, a processing result that reaches the (confidence or accuracy) threshold from all processing results after performing the remote sensing data processing task on the collected remote sensing image data, and the processing result is retained.

[0051] As an optional method, the above threshold can be configured based on the satellite-to-ground link communication bandwidth and the computing resource occupancy on the satellite. For example, when the satellite-to-ground link communication bandwidth is high, the threshold can be configured higher to retain fewer preliminary processing results and transmit more first remote sensing image data to the ground station for processing, thereby ensuring the real-time nature of data processing and ensuring higher data processing accuracy. Similarly, when the satellite computing resource occupancy rate is high, more first remote sensing image data is transmitted to the ground station for processing. When the satellite-to-ground link communication bandwidth is low, or when the satellite computing resource occupancy rate is low, the threshold can be configured lower to reduce the first remote sensing image data transmitted to the ground station. In other words, the size of the threshold is positively correlated with the satellite-to-ground link communication bandwidth or the computing resource occupancy rate on the satellite.

[0052] Taking the classification of remote sensing image data as an example, when the first processing model processes the remote sensing data, it obtains a number of probability values ​​corresponding to the number of classification dimensions. The category corresponding to the highest probability value is taken as the classification result. Based on a configured threshold, when the highest probability value reaches the threshold, the confidence level of the remote sensing image classification is considered high. Conversely, when the highest probability value falls below the configured threshold, the confidence level of the remote sensing image classification is considered low. Finally, all high-confidence classification results are retained, and the low-confidence remote sensing image data in the classification results is used as the first remote sensing image data and transmitted to the ground station along with the high-confidence classification results.

[0053] As described in the previous embodiment, the satellite transmits first remote sensing image data to a ground station, which can then optimize the first processing model based on the first remote sensing image data. Furthermore, in an optional embodiment, the satellite transmits all collected remote sensing image data to the ground station when there is redundancy in the satellite-to-ground link. In this way, the ground station can utilize all received remote sensing image data to optimize the first processing model. While optimizing the first processing model, the ground station can also optimize its own data processing model. For example, in an embodiment where the ground station processes the first remote sensing image data using a second processing model, the ground station can optimize the second processing model based on the received remote sensing image data (either the first remote sensing image data or all remote sensing image data). The ground station's real-time optimization of the first and / or second processing models can improve the accuracy of remote sensing data processing tasks.

[0054] In some embodiments, as Figure 5 As shown, another remote sensing image processing method provided by this application includes: Receive a remote sensing data processing task, which is usually input by a user.

[0055] Based on the remote sensing data processing mission, the ground station controls satellites to cooperate in processing the remote sensing image data collected by the satellites.

[0056] The control ground station optimizes the first processing model used by the satellite to process remote sensing data processing tasks.

[0057] The optimized model parameters are uploaded to at least the satellite that performs the remote sensing data processing task.

[0058] In the remote sensing image processing method of the embodiment of the present application, a lightweight model can still be selected as the first processing model configured for the satellite to improve the remote sensing data processing efficiency and quickly determine the remote sensing data that needs to be transmitted downlink.

[0059] As an optional implementation, similar to the previous embodiment, satellite scheduling is performed by selecting at least one satellite from the constellation for scheduling based on the current satellite status information of each satellite in the constellation to which the ground station is connected and the satellite requirements of the remote sensing data processing task. Specific feasible scheduling methods can be found in the previous embodiment and will not be further described here.

[0060] More than one satellite may be dispatched to perform remote sensing data processing tasks. Furthermore, different remote sensing data processing tasks are typically received sequentially. Therefore, different satellites may be dispatched at the ground station for different remote sensing data processing tasks, all of which process remote sensing image data using the first processing model. To improve the efficiency of satellite updates to the first processing model while ensuring accuracy in processing remote sensing image data during the next mission, in one optional embodiment, the model parameters optimized by the ground station are uploaded to all satellites in the constellation that can communicate with the ground station, effectively and synchronously updating the model parameters of the first processing model across multiple satellites.

[0061] As an optional implementation, Figure 6 As shown, the method for controlling the ground station to dispatch satellites to cooperate in completing the processing of remote sensing image data collected by the satellites includes: Control the ground station to receive the first remote sensing image data transmitted by the satellite and the processing results retained by the remote sensing data processing task, where the first remote sensing image data only includes the remote sensing image data for which the processing results have not been transmitted by the satellite; and control the ground station to process the first remote sensing image data according to the remote sensing data processing task (the related operations indicated).

[0062] As an optional implementation, the above process of scheduling satellites includes: According to the current satellite status information of each satellite in the constellation connected to the ground station and based on the requirements of the remote sensing data processing task for the satellite, at least one satellite is selected from the constellation for the ground station to schedule; Based on the remote sensing data processing task, a scheduling instruction is generated and sent to the scheduled satellite; the scheduling instruction controls the satellite to process the collected remote sensing image data using the first processing model according to the remote sensing data processing task, and transmit the retained processing results and the first remote sensing image data to the ground station. The first remote sensing image data only includes the remote sensing data for which the processing results have not been transmitted; controlling the ground station to process the first remote sensing image data; The control ground station fuses the processing results transmitted by the satellite with its own processing results of the first remote sensing image data as the final processing result of the remote sensing data processing task. The fusion method can be a simple combination of the processing results of the two, or the processing object and the processed object can be marked separately based on the combination.

[0063] In one feasible implementation, a confidence or accuracy threshold is preconfigured on the satellite. Based on the threshold, the satellite selects the retained processing results from all processing results of the remote sensing data processing task. The example of selecting the retained processing results is described above and will not be repeated here.

[0064] The embodiment of the present application also optimizes the first processing model of the satellite in real time. In an optional embodiment, at least the ground station is controlled to optimize the first processing model based on the first remote sensing image data. This includes optimizing the first processing model with the first remote sensing image data, or when the satellite is still transmitting more remote sensing image data, optimizing the first processing model with all the transmitted remote sensing image data (including the corresponding processing results). Similarly, if the ground station uses the second processing model to process the first remote sensing image data, the ground station can also be controlled to optimize the second processing model using the first remote sensing image data, or to optimize the second processing model using more remote sensing image data transmitted by the satellite.

[0065] Through the above-mentioned method, this application achieves the unification of constellation management and task scheduling. Faced with complex satellite constellation architectures and diverse onboard resources, unified task scheduling and resource management effectively coordinates satellites with different configurations and avoids conflicts in task allocation. This unified management approach ensures optimal utilization of resources within the satellite constellation, improving the efficiency of task scheduling and the success rate of task execution.

[0066] In addition, for remote sensing data processing tasks, in some optional implementations, when controlling the ground station to dispatch satellites to cooperate in completing remote sensing data processing tasks, the specific matters of satellite processing of remote sensing data processing tasks are prioritized. It can be understood that the priority sorting of specific matters of remote sensing data processing tasks, such as the priority sorting of the geographical location of remote sensing image data, the priority sorting of the data size of different remote sensing image data, etc., are included in the remote sensing data processing tasks indicated by the scheduling instructions. Of course, the same priority processing requirements can also be formulated in the control of the ground station. In the embodiment of the present application, the execution status of the remote sensing data processing task is also monitored, such as the operating status of the satellite, the working status of the remote sensor, the progress of remote sensing image data processing, and the stability of the satellite-to-ground link, to ensure that a timely response can be made when the task execution is abnormal.

[0067] In some optional implementations, if any anomalies or faults are identified during monitoring, the ground station is immediately controlled to take remedial measures. This may involve switching to a backup satellite, reconfiguring sensor parameters, or repairing the satellite-to-ground link. Through timely fault handling, the system can reduce the risk of mission failure and ensure mission continuation.

[0068] Furthermore, necessary mission adjustments and optimizations can be made based on monitoring data. This may include re-prioritizing specific mission items, optimizing satellite computing resource utilization, or adjusting remote sensing data processing procedures, ensuring the mission is executed smoothly and completed as planned.

[0069] In some embodiments, as Figure 7As shown, the system structure for controlling the ground station to dispatch satellites to cooperate in completing the remote sensing data processing task can be achieved by the ground station dispatching multiple satellites at the same time.

[0070] Based on the ideas of the above embodiments, in a feasible implementation manner, the present application further provides a computer-readable storage medium storing computer instructions, which can execute the remote sensing image processing method of any of the above embodiments.

[0071] As a feasible implementation method, Figure 8 As shown, an embodiment of the present application provides a remote sensing image processing device for use in a ground station. The device is configured with an information management module, a task scheduling module, a ground-side algorithm module, and a ground communication module. In addition, a ground database can be configured for data storage. In another embodiment of a remote sensing image processing device for use in a satellite, the device is configured with an onboard communication module, a data acquisition module, and a satellite-side algorithm module. Alternatively, a satellite database can be configured for data storage. The remote sensing image processing devices of the two embodiments are described below.

[0072] For the remote sensing image processing device applied to the ground station, its information management module is used to receive the remote sensing data processing task; the task scheduling module is used to schedule the satellite to cooperate based on the remote sensing data processing task to complete the processing of the remote sensing image data collected by the satellite, including: instructing the scheduled satellite to downlink the first remote sensing image data, and processing the processing results retained by the remote sensing data processing task, and the first remote sensing image data only includes the remote sensing image data whose processing results are not downlinked by the satellite; the ground-end algorithm module is used to process the first remote sensing image data according to the remote sensing data processing task, and optimize the first processing model used by the satellite when processing the remote sensing data processing task; the ground communication module is used to receive the first remote sensing image data, and upload the optimized model parameters to at least the satellite.

[0073] For a remote sensing image processing device applied to a satellite, its data acquisition module is used to collect remote sensing image data (according to the requirements of the remote sensing data processing task indicated by the scheduling instruction); the satellite-side algorithm module is used to cooperate with the ground station to complete the processing of the remote sensing image data using the first processing model according to the remote sensing data processing task indicated by the scheduling instruction, including: determining the processing results retained by the remote sensing data processing task and the first remote sensing image data, and the first remote sensing image data only includes the remote sensing image data for which the processing results are not retained; the onboard communication module is used to receive the scheduling instruction and the model parameters of the first processing model after the ground station optimizes the model, and is also used to transmit the first remote sensing image data and the retained processing results to the ground station that sends the scheduling instruction.

[0074] The specific configuration data of each module in the two remote sensing image processing devices mentioned above can refer to the corresponding process features in the above embodiments of the remote sensing image processing method.

[0075] For ease of understanding, the following describes the operating procedures of the system consisting of the two remote sensing image processing devices described above: The information management module receives user-initiated task instructions containing remote sensing data processing tasks, retrieves the current satellite status information of each satellite in the connected constellation from the ground database, and transmits it to the task scheduling module for satellite scheduling. The scheduling instructions are then transmitted to the scheduled satellite via the satellite-to-ground link established by the ground communication module and the onboard communication module. After receiving the scheduling instructions, the satellite (and its onboard communication module) uses the data acquisition module to collect the required remote sensing image data from the corresponding remote sensor according to the requirements of the remote sensing data processing task indicated in the scheduling instructions and stores it in the satellite database. The data acquisition module also periodically collects satellite status information and stores it in the satellite database. During the satellite-to-ground communication window, the satellite transmits this satellite status information to the ground station via the onboard communication module. The satellite-side algorithm module retrieves the remote sensing image data required for this remote sensing data processing task from the satellite database. Based on the instructions of the remote sensing data processing task, it processes the remote sensing image data using the configured first processing model to obtain all processing results and selects the retained processing results based on the configured threshold. The retained processing results, along with the first remote sensing image data corresponding to the unretained processing results, are transmitted to the ground station via the onboard communication module. The transmitted processing results are stored in the ground database. The transmitted first remote sensing image data is further processed by the ground-side algorithm module using the configured second processing model, and the processing results are also stored in the ground database. The information management module obtains the processing results of the ground-side algorithm module and the processing results transmitted by the satellite from the ground database, integrating them into the final processing results and feeding them back to the user. In addition, the ground-side algorithm module also uses the first remote sensing image data (or all the remote sensing image data received when the satellite is also transmitting more remote sensing image data) to optimize the first processing model (the first processing model is also stored in the ground station), and uploads the optimized model parameters to the satellite (the currently scheduled satellite or all satellites that can communicate in the connected constellation) through the ground communication module to update the model parameters of the first processing model in the satellite-side algorithm module.

[0076] The present invention is not limited to the aforementioned specific embodiments, but extends to any new features or any new combination disclosed in this specification, as well as any new method or process steps or any new combination disclosed.

Claims

1. A remote sensing image processing method, characterized in that: Applicable to ground stations, including: Receive remote sensing data processing tasks; Scheduling satellites to cooperate in processing remote sensing image data collected by the satellite based on the remote sensing data processing task includes: receiving first remote sensing image data transmitted by the satellite, and processing a processing result retained by the remote sensing data processing task, wherein the first remote sensing image data only includes remote sensing image data for which the processing result has not been transmitted by the satellite; and processing the first remote sensing image data according to the remote sensing data processing task; Optimizing a first processing model used by the satellite to process the remote sensing data processing task; The optimized model parameters are uploaded to at least the satellite.

2. The remote sensing image processing method according to claim 1, wherein: Methods for scheduling satellites include: According to the current satellite status information of each satellite in the connected constellation and based on the requirements of the remote sensing data processing task on the satellite, at least one satellite is selected from the constellation for scheduling.

3. The remote sensing image processing method according to claim 2, wherein: Uploading the optimized model parameters to at least the satellite includes: The optimized model parameters are uploaded to all satellites in the constellation that can communicate with it.

4. The remote sensing image processing method according to claim 1, wherein: The method of scheduling satellites to cooperatively complete processing of collected remote sensing image data also includes: The processing result of the satellite downlink and the processing result of the first remote sensing image data are integrated as the final processing result of the remote sensing data processing task.

5. The remote sensing image processing method according to claim 1, wherein: Optimizing a first processing model used by the satellite to process the remote sensing data processing task includes: The first processing model is optimized based on at least the first remotely sensed image data.

6. A remote sensing image processing method, characterized in that: Applications in satellites include: Receive dispatch instructions; According to the remote sensing data processing task indicated by the scheduling instruction, the first processing model is used to cooperate with the ground station to complete processing of the collected remote sensing image data, including: transmitting the first remote sensing image data and the processing results retained by the remote sensing data processing task to the ground station, wherein the first remote sensing image data only includes the remote sensing image data for which the processing results have not been transmitted; Receive model parameters after the ground station optimizes the first processing model.

7. The remote sensing image processing method according to claim 6, wherein: Methods for selecting which processing results to retain include: Based on a preconfigured threshold, a processing result that reaches the threshold is selected from all processing results after the remote sensing data processing task is performed on the collected remote sensing image data.

8. The remote sensing image processing method according to claim 6, wherein: Also includes: When the satellite-to-ground link is redundant, all remote sensing image data are downloaded to the ground station.

9. A remote sensing image processing method, characterized in that: include: Receive remote sensing data processing tasks; Controlling a ground station to dispatch a satellite to cooperate in processing remote sensing image data collected by the satellite based on the remote sensing data processing task, including: controlling the ground station to receive first remote sensing image data transmitted by the satellite, and processing a processing result retained by the remote sensing data processing task, wherein the first remote sensing image data only includes remote sensing image data for which the processing result has not been transmitted by the satellite; and controlling the ground station to process the first remote sensing image data according to the remote sensing data processing task; controlling the ground station to optimize a first processing model used by the satellite to process the remote sensing data processing task; The optimized model parameters are uploaded to at least the satellite.

10. A computer-readable storage medium storing computer instructions, characterized in that: Running the computer instructions can execute the remote sensing image processing method according to any one of claims 1 to 9.

11. A remote sensing image processing device, characterized in that: Applicable to ground stations, including: Information management module, used to receive remote sensing data processing tasks; a task scheduling module, configured to schedule satellites to cooperate in processing remote sensing image data collected by the satellites based on the remote sensing data processing task, including: instructing the scheduled satellites to downlink first remote sensing image data, and processing results retained by the remote sensing data processing task, wherein the first remote sensing image data only includes remote sensing image data for which the satellites have not downlinked processing results; a ground-side algorithm module, configured to process the first remote sensing image data according to a remote sensing data processing task, and optimize a first processing model used by the satellite to process the remote sensing data processing task; The ground communication module is used to receive the first remote sensing image data and upload the optimized model parameters to at least the satellite.

12. A remote sensing image processing device, characterized in that: For satellites, including: Data acquisition module, used for collecting remote sensing image data; The satellite-side algorithm module is configured to cooperate with the ground station to complete processing of the remote sensing image data using a first processing model according to the remote sensing data processing task indicated by the scheduling instruction, including: determining a processing result retained by the remote sensing data processing task and first remote sensing image data, where the first remote sensing image data only includes remote sensing image data for which the processing result is not retained; The satellite communication module is used to receive the scheduling instruction and the model parameters after the ground station optimizes the first processing model, and is also used to transmit the first remote sensing image data and the retained processing results to the ground station that sends the scheduling instruction.

Images