Data processing method and device based on remote sensing data

Abstract

Embodiments of the present specification provide a data processing method and device based on remote sensing data, wherein a data processing method based on remote sensing data comprises: acquiring encrypted plot data sent by a first server and inputting into a trusted execution environment; the encrypted plot data is obtained by encrypting plot data based on a public key by the first server; in the trusted execution environment, the encrypted plot data is decrypted based on a private key paired with the public key, and plot crop type marking is performed on the input remote sensing data based on the decryption result to obtain a remote sensing data sample; in the trusted execution environment, model training is performed according to the remote sensing data sample, and a crop type identification model obtained by training is encrypted; and acquiring and storing the encrypted model output by the trusted execution environment.

Description

[0001] This application is a divisional application of Chinese patent application filed on May 20, 2021, with application number 202110552120.2 and title "Data Processing Method and Apparatus Based on Remote Sensing Data". Technical Field

[0002] This document relates to the field of data processing technology based on remote sensing data, and in particular to a data processing method and apparatus based on remote sensing data. Background Technology

[0003] With the development of remote sensing technology, remote sensing data has been widely used in various industries. Remote sensing technology is a general term for a comprehensive technical system for observing the Earth and celestial bodies from the ground to space. It can acquire satellite data from remote sensing technology platforms, and receive, process, and analyze information through remote sensing instruments. Remote sensing technology is a rapidly developing high technology, and the information network it has formed is constantly providing people with a large amount of scientific data and dynamic information. Remote sensing data generally refers to remote sensing images, which are films or photographs that record the electromagnetic wave magnitude of various ground objects. They are mainly divided into aerial photographs and satellite photographs. Summary of the Invention

[0004] This specification provides one or more embodiments of a data processing method based on remote sensing data. The method includes: acquiring encrypted land parcel data sent by a first server and transmitting it to a trusted execution environment. The encrypted land parcel data is obtained by the first server encrypting land parcel data using a public key. In the trusted execution environment, the encrypted land parcel data is decrypted using a private key paired with the public key, and the incoming remote sensing data is labeled with crop types based on the decryption result to obtain remote sensing data samples. A model is trained in the trusted execution environment using the remote sensing data samples, and the trained crop type identification model is encrypted. The encrypted model output by the trusted execution environment is acquired and stored.

[0005] This specification provides one or more embodiments of a method for processing identification data based on remote sensing data, comprising: receiving a public key generated by a trusted execution environment (TEA) sent by a second server; encrypting land parcel data according to the public key and sending the encrypted land parcel data to the second server; obtaining an encrypted identification result of the crop type of the target remote sensing data sent by the second server. The crop type is obtained by the second server using a crop type identification model in the TEA to identify the crop type of the incoming target remote sensing data; the encrypted identification result is obtained by the TEA encrypting the crop type of the land parcel according to the public key; and decrypting the encrypted identification result using a private key paired with the public key to obtain the crop type of the land parcel, thereby determining the crop type of the land parcel corresponding to the user-submitted land parcel labeling data.

[0006] This specification provides one or more embodiments of a data processing apparatus based on remote sensing data, comprising: an acquisition module configured to acquire encrypted land parcel data sent by a first server and transmit it to a trusted execution environment. The encrypted land parcel data is obtained by the first server encrypting land parcel data using a public key. A decryption module configured to decrypt the encrypted land parcel data in the trusted execution environment using a private key paired with the public key, and to label the incoming remote sensing data with crop types based on the decryption result, thereby obtaining remote sensing data samples. A training module configured to train a model in the trusted execution environment based on the remote sensing data samples, and to encrypt the trained crop type identification model. A storage module configured to acquire and store the encrypted model output by the trusted execution environment.

[0007] This specification provides one or more embodiments of a remote sensing data identification processing apparatus, comprising: a receiving module configured to receive a public key generated by a trusted execution environment (TEA) sent by a second server; an encryption module configured to encrypt land parcel data according to the public key and send the encrypted land parcel data to the second server; an acquisition module configured to acquire an encrypted identification result of the crop type of the target remote sensing data sent by the second server. The crop type is obtained by the second server using a crop type identification model in the TEA to identify the crop type of the incoming target remote sensing data. The encrypted identification result is obtained by the TEA encrypting the crop type of the land parcel according to the public key; and a decryption module configured to decrypt the encrypted identification result using a private key paired with the public key to obtain the crop type of the land parcel, thereby determining the crop type of the land parcel corresponding to the user-submitted land parcel labeling data.

[0008] This specification provides one or more embodiments of a data processing device based on remote sensing data, including: a processor; and a memory configured to store computer-executable instructions, which, when executed, cause the processor to: acquire encrypted land parcel data sent by a first server and transmit it to a trusted execution environment. The encrypted land parcel data is obtained by the first server encrypting land parcel data based on a public key. In the trusted execution environment, the encrypted land parcel data is decrypted based on a private key paired with the public key, and the incoming remote sensing data is labeled with crop types based on the decryption result to obtain remote sensing data samples. In the trusted execution environment, a model is trained based on the remote sensing data samples, and the trained crop type identification model is encrypted. The encrypted model output by the trusted execution environment is acquired and stored.

[0009] This specification provides one or more embodiments of a remote sensing data identification processing device, comprising: a processor; and a memory configured to store computer-executable instructions, which, when executed, cause the processor to: receive a public key generated by a trusted execution environment from a second server; encrypt land parcel data according to the public key and send the encrypted land parcel data to the second server; obtain an encrypted identification result of the crop type of the target remote sensing data sent by the second server. The crop type is obtained by the second server using a crop type identification model in the trusted execution environment to identify the crop type of the incoming target remote sensing data; the encrypted identification result is obtained by the trusted execution environment encrypting the crop type of the land parcel according to the public key; and decrypt the encrypted identification result using a private key paired with the public key to obtain the crop type of the land parcel, thereby determining the crop type of the land parcel corresponding to the user-submitted land parcel labeling data.

[0010] This specification provides one or more embodiments of a storage medium for storing computer-executable instructions that, when executed, implement the following process: acquiring encrypted land parcel data sent by a first server and transmitting it to a trusted execution environment. The encrypted land parcel data is obtained by the first server encrypting land parcel data based on a public key. In the trusted execution environment, the encrypted land parcel data is decrypted based on a private key paired with the public key, and the incoming remote sensing data is labeled with crop types based on the decryption result to obtain remote sensing data samples. In the trusted execution environment, a model is trained based on the remote sensing data samples, and the trained crop type identification model is encrypted. The encrypted model output by the trusted execution environment is acquired and stored.

[0011] This specification provides one or more embodiments of a storage medium for storing computer-executable instructions that, when executed, implement the following process: receiving a public key generated by a trusted execution environment from a second server; encrypting land parcel data according to the public key and sending the encrypted land parcel data to the second server; obtaining an encrypted identification result of the crop type of the target remote sensing data sent by the second server. The crop type is obtained by the second server using a crop type identification model in the trusted execution environment to identify the crop type of the incoming target remote sensing data; the encrypted identification result is obtained by the trusted execution environment encrypting the crop type of the land parcel according to the public key; and decrypting the encrypted identification result using a private key paired with the public key to obtain the crop type of the land parcel, thereby determining the crop type of the land parcel corresponding to the user-submitted land parcel labeling data. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in one or more embodiments of this specification or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 A flowchart illustrating a data processing method based on remote sensing data, provided for one or more embodiments of this specification;

[0014] Figure 2 A timing diagram of a data processing method based on remote sensing data for model training scenarios, provided in one or more embodiments of this specification;

[0015] Figure 3 A timing diagram of a data processing method based on remote sensing data for application in a remote sensing image recognition scenario, provided by one or more embodiments of this specification;

[0016] Figure 4 A flowchart illustrating a remote sensing data processing method for identification, provided in one or more embodiments of this specification;

[0017] Figure 5 A schematic diagram of a data processing device based on remote sensing data provided for one or more embodiments of this specification;

[0018] Figure 6 A schematic diagram of an identification data processing device based on remote sensing data, provided for one or more embodiments of this specification;

[0019] Figure 7 A schematic diagram of the structure of a data processing device based on remote sensing data provided for one or more embodiments of this specification;

[0020] Figure 8 This is a schematic diagram of the structure of a remote sensing data identification processing device provided for one or more embodiments of this specification. Detailed Implementation

[0021] To enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the technical solutions in one or more embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this specification, and not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of this document.

[0022] This specification provides an example of a data processing method based on remote sensing data:

[0023] Reference Figure 1 It shows a flowchart of a data processing method based on remote sensing data provided in this embodiment, with reference to... Figure 2 It shows a timeline diagram of a data processing method based on remote sensing data applied to a model training scenario provided in this embodiment, with reference to... Figure 3 The diagram shows a timing diagram of a data processing method based on remote sensing data for application in remote sensing image recognition scenarios, as provided in this embodiment.

[0024] Reference Figure 1 The data processing method based on remote sensing data provided in this embodiment is applied to a second server and specifically includes steps S102 to S108.

[0025] Step S102: Obtain the encrypted land parcel data sent by the first server and pass it into the trusted execution environment.

[0026] In practical applications, the combination of satellite remote sensing and artificial intelligence has promoted the development and progress of agricultural technology. The application of agricultural technology often involves farmers' remote sensing data. To protect farmers' privacy data in compliance with regulations, this embodiment provides a data processing method based on remote sensing data. This method establishes a Trusted Execution Environment (TEE) in the institutional domain storing the remote sensing data, creating a joint data computation method between the institutional domain and the service domain. This completes the training and identification of remote sensing data, protecting the institutional domain's remote sensing data from leakage. Furthermore, the service domain's land parcel data is placed in encrypted form within the institutional domain, preventing the institutional domain from viewing the plaintext of the service domain's land parcel data. Calculations are performed on the institutional domain's remote sensing data and the service domain's land parcel data within the institutional domain, and the calculation results are sent to the service domain in encrypted form. Specifically, a model is trained in the institutional domain using the institutional domain's remote sensing data and the service domain's land parcel data to obtain a crop type identification model. This model is then used to identify remote sensing data of unknown crop types, and the identification results are encrypted before being sent to the service domain. This assures the institutional domain that no remote sensing data has been leaked, thus enhancing the protection of privacy data.

[0027] In this embodiment, the first server is a server deployed in the service domain; the first server stores the land parcel data; the encrypted land parcel data is obtained by the first server encrypting the land parcel data based on a public key; the land parcel data includes the location data of the land parcel and the crop type data corresponding to the land parcel; the first server encrypts the stored land parcel data according to the public key generated and sent by the TEE deployed in the institutional domain, and sends the encrypted land parcel data to the second server; the second server is a server deployed in the institutional domain, specifically, the second server includes a data server and a trusted server; the data server refers to a server that stores remote sensing data; the trusted server refers to a server that deploys the TEE; furthermore, the second server may only include the data server or the trusted server. When the second server includes the data server, the TEE is deployed on the data server, and the remote sensing data is stored in a location outside the TEE in the data server; when the second server includes the trusted server, the remote sensing data is stored in a location outside the TEE in the trusted server.

[0028] In specific implementation, to ensure data privacy and prevent other institutional domains from viewing the input TEE and TEE output data, ensuring that only designated institutional domains can view the data, in an optional implementation of this embodiment, the following steps are performed before obtaining the encrypted land parcel data:

[0029] A key pair is generated in the trusted execution environment; the key pair includes the private key and the public key.

[0030] The trusted server obtains the public key output by the trusted execution environment and forwards the public key to the first server through the data server;

[0031] The key pair is sent by the second server to the key server for storage after it is generated.

[0032] Specifically, the TEE generates a key pair containing a public key and a private key and outputs it to a trusted server. The trusted server sends the generated public key to the first server in the service domain, so that the first server can encrypt the stored land parcel data based on the public key. In order for the first server to be able to decrypt the encrypted identification result after receiving the encrypted identification result output by the TEE, the trusted server sends the generated key to a third-party server (key server) jointly selected by the organization domain and the service domain for storage, so that the first server can call it.

[0033] During the process of the first server sending encrypted land parcel data to the second server, the encrypted land parcel data is first sent to the data server, and then the data server sends it to the trusted server. The trusted server then transmits the encrypted land parcel data to the TEE. In one optional implementation of this embodiment, the encrypted land parcel data sent by the first server is obtained and transmitted to the trusted execution environment by performing the following steps: First, the data server obtains the encrypted land parcel data sent by the first server and forwards the encrypted land parcel data to the trusted server; then the trusted server transmits the encrypted land parcel data to the trusted execution environment through the connection interface.

[0034] For example, the TEE generates an asymmetric public-private key pair and outputs it to a trusted server. The trusted server sends the asymmetric public key to a data server, which then sends it to a server in the service domain. To prevent the data server from sending other asymmetric public keys that could lead to the leakage of land parcel data on the service domain server, the TEE and the service domain server jointly monitor the transmission process of the asymmetric public key. Upon receiving the asymmetric public key, the service domain server encrypts the land parcel data using the asymmetric public key and sends the encrypted land parcel data to the data server, which then forwards it to the trusted server. The server then transmits the encrypted land parcel data to the TEE via the PCI (Peripheral Component Interconnect) interface.

[0035] In practice, model training is performed based on remote sensing data labeled with crop types. To improve the efficiency of model training and reduce data processing during the training process, it is necessary to preprocess the remote sensing data of unlabeled crop types stored on the data server before transmitting the preprocessed remote sensing data to the TEE. This reduces the complexity of data processing during model training. In one optional implementation method provided in this embodiment, the remote sensing data is transmitted to the TEE in the following way:

[0036] The initial remote sensing data is preprocessed in the data server to obtain the remote sensing data, and the remote sensing data is then sent to the trusted server.

[0037] The trusted server transmits the remote sensing data to the trusted execution environment through a connection interface;

[0038] The preprocessing of the initial remote sensing data includes at least one of the following:

[0039] The initial remote sensing data is subjected to radiometric correction, geometric correction, parameter extraction, and / or classification.

[0040] For example, the data server in the institutional domain eliminates or corrects distortions in remote sensing images caused by radiometric errors. It also uses a series of models to correct and eliminate distortions caused by factors such as photographic material deformation, objective lens distortion, atmospheric refraction, Earth curvature, Earth rotation, and topographic relief, which result in inconsistencies between the geometric position, shape, size, orientation, and other features of various objects in the original image and the representation requirements in the reference system. Parameters are extracted from the remote sensing images obtained after radiometric and geometric corrections, and the cultivated land portion in the remote sensing images is extracted. The final remote sensing images are then sent to a trusted server, which in turn transmits them to a TEE via the PCI interface.

[0041] It should be noted that the process of the data server sending encrypted land parcel data to the trusted server, and the process of the data server sending remote sensing data to the trusted server, can be that the data server first sends the encrypted land parcel data to the trusted server, and then sends the remote sensing data; or the data server, upon receiving the encrypted land parcel data, preprocesses the initial remote sensing data, and then sends the preprocessed remote sensing data together with the encrypted land parcel data to the trusted server.

[0042] Step S104: In the trusted execution environment, the encrypted land parcel data is decrypted based on the private key paired with the public key, and the crop type of the land parcel is marked on the incoming remote sensing data based on the decryption result to obtain a remote sensing data sample.

[0043] The crop type labeling of the plot refers to the process of matching the crop types contained in the plot data with the remote sensing data based on the location information of the remote sensing data and the location information of the plot data.

[0044] In practice, after receiving the encrypted land parcel data, the TEE needs to decrypt the encrypted land parcel data to obtain the land parcel data. In one optional implementation method provided in this embodiment, the encrypted land parcel data is decrypted in the following way:

[0045] A private key query request is sent to the key server; the key server generates a private key viewing reminder based on the private key query request and sends it to the first server;

[0046] Obtain the private key sent by the key server when the first server submits a confirmation instruction for the private key viewing reminder;

[0047] The encrypted land parcel data is decrypted based on the private key.

[0048] Specifically, to ensure data privacy, the key pair generated by the TEE is stored in a key server jointly selected by the organization domain and the service domain. To prevent other servers from accessing the key pair and causing privacy data leakage, the key pair can only be accessed after confirmation by the organization domain and the service domain. It should be noted that the key pair is generated by the TEE and can also be stored in both the TEE and the key server. When the TEE needs to use the key pair, it can perform decryption, encryption, and other processing based on the stored key pair. When the server on the service side needs to use the key pair, it queries the key server.

[0049] For example, when the TEE receives incoming encrypted land parcel data, it generates and outputs a private key query request. The trusted server sends this request to a third-party server via the data server. The third-party server, based on the request, sends a private key viewing notification to the server in the service domain. Upon receiving confirmation from the server in the service domain regarding the notification, the third-party server sends the asymmetric private key to the trusted server, which then transmits it to the TEE. To prevent the data server from using this private key to decrypt the encrypted land parcel data, the encrypted land parcel data can be marked and set to be automatically deleted after detecting that the data server has sent it to the trusted server. Alternatively, access permissions can be set for the private key when it is sent by the key server, allowing only those with the required permissions to access it.

[0050] In practical applications, for service domains providing agricultural technology services, in order to reduce service costs and promote the popularization and dissemination of agricultural technology, service domains need to cooperate with data operation agency domains such as resource satellite centers. Specifically, service domains use full-volume download servers established abroad to download remote sensing images, and then use OSS (Object Storage Service, Alibaba Cloud Object Storage Service) to accelerate the copying to China, and then combine it with land parcel data for analysis. However, data downloading requires storage and bandwidth costs. In the domestic satellite scenario, there is no commercial download platform for remote sensing data. Therefore, obtaining remote sensing data outside of the agency domain that manages remote sensing data is non-compliant. Even if it is possible to obtain remote sensing data outside of the agency domain, it still requires resources, bandwidth, and data purchase costs. In addition, if the service domain's land parcel data is sent to the agency domain for calculation, it violates the service domain's data privacy rights and is still non-compliant.

[0051] In this embodiment, by deploying a TEE in the institutional domain, the land parcel data of the service domain is sent to the TEE in encrypted form, and the remote sensing data of the institutional domain is sent to the TEE. A model with crop type recognition capability is trained in the TEE, and the obtained crop type recognition model is used to identify the remote sensing data. Finally, the encrypted recognition result is sent to the service domain, thereby ensuring that the data of the institutional domain does not leave the domain and ensuring the security of the data leaving the domain.

[0052] In specific implementation, after decrypting the encrypted plot data to obtain the plot data, in order to train a model with crop type recognition capabilities, it is necessary to determine remote sensing data samples based on the plot data and remote sensing data, and then train the model based on the remote sensing data samples. In one optional implementation method provided in this embodiment, the remote sensing data samples are obtained in the following way:

[0053] Based on the location information in the decrypted land parcel data and the location information in the remote sensing data, the land parcel data and the remote sensing data are matched for location.

[0054] Crop type is marked at the successfully matched remote sensing plots based on the crop type information in the plot data.

[0055] Specifically, the first step is to match the location of the land parcel data with the location of the remote sensing data. In other words, the location in the land parcel data is matched with the location in the remote sensing data. If the location matches, the crop type information in the land parcel information is marked as the crop type information in the remote sensing data corresponding to the location to obtain the remote sensing data sample.

[0056] It should be noted that the labeling of crop types for plots in the incoming remote sensing data based on the decryption results can be done by the TEE; after the TEE decrypts the encrypted plot data, it labels the crop types for plots in the incoming remote sensing data based on the decryption results to obtain remote sensing data samples; alternatively, the TEE can decrypt the encrypted plot data, input the decryption results and remote sensing data into the model, and the model can label the crop types for plots in the incoming remote sensing data based on the decryption results to obtain remote sensing data samples, and then train the model based on the remote sensing data samples.

[0057] For example, the TEE decrypts the encrypted land parcel data using the private key sent by the key server to obtain the land parcel data. The land parcel data and remote sensing data are then input into the U-Net model. The U-Net model performs semantic segmentation on the remote sensing image based on the land parcel data to obtain remote sensing data samples.

[0058] Step S106: In the trusted execution environment, model training is performed based on the remote sensing data samples, and the crop type identification model obtained from the training is encrypted.

[0059] The crop species identification model refers to a model trained using the remote sensing data samples that meets preset conditions in terms of accuracy and recall, such as the trained U-Net model.

[0060] In practical implementation, to make the crop species identification model more accurate and effective, after training the model to meet certain accuracy and recall rates, the model is used as the crop species identification model. In one optional implementation method provided in this embodiment, the model is trained in the following way:

[0061] The remote sensing data samples are divided into a training sample set and a test sample set;

[0062] At least one candidate model is obtained by training the model based on the training sample set;

[0063] The test sample set is input into the candidate model, and the evaluation parameters of each candidate model are determined based on the recognition results of each candidate model.

[0064] Candidate models whose evaluation parameters meet preset conditions are selected as crop type identification models.

[0065] In order to improve the efficiency of model training, the AI ​​accelerator in TEE is called to make the model training process more efficient.

[0066] For example, remote sensing data samples are divided into training sample set and test sample set according to a certain ratio. At least one candidate U-Net model is trained using the training sample set. The test sample set is input into each candidate U-Net model to obtain the accuracy and recall of each candidate U-Net model. The candidate U-Net model with higher accuracy and recall than other candidate U-Net models is determined as the crop type identification model for crop type identification.

[0067] It should be noted that, in order to further ensure the security of privacy data and prevent privacy data leakage, the obtained crop type identification model is also stored in an encrypted form. In order to facilitate the reloading of the trained crop type identification model after a power outage, and to avoid the loss of the crop type identification model due to power outage and to avoid resource consumption caused by repeated model training, after the crop type identification model is trained by TEE, the crop type identification model is first encrypted using the generated public key, and then the encrypted model is output.

[0068] Step S108: Obtain and store the encryption model output by the trusted execution environment.

[0069] In practice, after obtaining the encrypted model, the trusted server stores it. In addition, after obtaining the encrypted model, the trusted server can send it to the server in the service domain via the data server, enabling the service domain to identify crop types from the acquired remote sensing data. Specifically, after obtaining the encrypted model, the server in the service domain uses the private key stored on the key server to decrypt the encrypted model and stores the decrypted crop type identification model.

[0070] After training and obtaining the crop species identification model, the model can be used to identify crop species in remote sensing data that is not labeled with crop species. In one optional implementation of this embodiment, if remote sensing data without crop species is detected, the following steps are performed:

[0071] The initial target remote sensing data is preprocessed to obtain target remote sensing data, and the encryption model is then loaded.

[0072] The encryption model and the target remote sensing data are then transmitted to the trusted execution environment.

[0073] The encryption model is decrypted in the trusted execution environment, and the target remote sensing data is identified based on the crop type identification model obtained from the decryption.

[0074] In one optional implementation of this embodiment, during the process of preprocessing the initial remote sensing data to be identified to obtain the remote sensing data to be identified and loading the encrypted model, the initial target remote sensing data is first preprocessed in the data server, and the preprocessed target remote sensing data is sent to the trusted server; upon receiving the target remote sensing data, the trusted server loads the encrypted model; after the trusted server loads the encrypted model, it transmits both the encrypted model and the target remote sensing data to the TEE, the TEE decrypts the encrypted model using a private key, and inputs the target remote sensing data into the decrypted crop type identification model.

[0075] Furthermore, in an optional implementation of this embodiment, after inputting the target remote sensing data into the decrypted crop type identification model, the identified crop types of the plots are encrypted in the trusted execution environment and output to the trusted server; the trusted server receives the encrypted identification result and forwards the encrypted identification result to the first server through the data server.

[0076] For example, during the model training phase, the input remote sensing data samples are obtained by semantic segmentation of remote sensing data and land parcel data corresponding to province P. After obtaining the crop type identification model, it is necessary to identify crop types from remote sensing data across the country. The data server preprocesses the remote sensing data across the country and sends the preprocessed remote sensing data to the trusted server. After receiving the remote sensing data, the trusted server loads the encrypted model and transmits the loaded encrypted model and remote sensing data to the TEE through the PCI interface. The TEE decrypts the encrypted model using the generated asymmetric private key and inputs the remote sensing data into the decrypted crop type identification model. After the crop type identification module outputs the identification result, to prevent the data server from viewing the identification result, the TEE uses the generated asymmetric public key to encrypt and output the identified crop types of the land parcels. After obtaining the encrypted identification result, the trusted server sends the encrypted identification result to the server in the service domain through the data server.

[0077] In practical implementation, the first server cannot view the actual recognition result after receiving the encrypted recognition result; it needs to decrypt the encrypted recognition result before viewing it. In one optional implementation provided in this embodiment, the first server also performs the following operations after receiving the encrypted recognition result:

[0078] Send a private key query request to the key server;

[0079] Obtain the private key generated by the trusted execution environment, which is sent by the key server upon receiving a confirmation instruction from the second server;

[0080] The encrypted identification result is decrypted based on the private key, and the crop type of the plot corresponding to the target remote sensing data obtained after decryption is stored.

[0081] Specifically, after the first server stores the target remote sensing data and the corresponding crop types for the plot, if it detects target plot annotation data submitted by a target user, it queries the target crop type corresponding to the target plot annotation data based on the crop type of the plot; then, it determines the target service quota based on the target crop type and distributes it to the target user. Specifically, first, the corresponding remote sensing location is determined based on the location information of the target plot annotation data; then, the crop type corresponding to that remote sensing location is queried; the queried crop type is used as the crop type corresponding to the target plot annotation data; and the target service quota corresponding to that crop type is determined according to preset service rules and distributed to the target user.

[0082] The following description uses the application of a remote sensing data processing method provided in this embodiment in a model training scenario as an example to further illustrate the remote sensing data processing method provided in this embodiment. (See also...) Figure 2 A data processing method based on remote sensing data, applied to model training scenarios, includes the following steps.

[0083] Step S206: Upon receiving the encrypted land parcel data, the data server preprocesses the initial remote sensing image.

[0084] Prior to this, the service domain server encrypts the land parcel data using the asymmetric public key generated by the trusted server and sends the encrypted land parcel data to the data server.

[0085] In step S208, the data server sends the encrypted land parcel data and the pre-processed remote sensing image to the trusted server, which then transmits it to the TEE.

[0086] The TEE is deployed in a trusted server; both the trusted server and the data server are deployed in the organization domain.

[0087] In step S210, the TEE decrypts the encrypted land parcel data based on the generated asymmetric private key.

[0088] Step S212: Input the decryption result and remote sensing image into the model, and perform semantic segmentation of the remote sensing image in the model based on the decryption result to obtain remote sensing image samples.

[0089] Step S214: Train the model based on remote sensing image samples to obtain a crop species identification model.

[0090] Step S216: Encrypt the crop type identification model using the generated asymmetric public key.

[0091] Step S218: Send the encrypted model obtained through encryption to a trusted server and store it.

[0092] The TEE trains a crop species identification model, encrypts the crop species identification model to obtain an encrypted model, and outputs the encrypted model to a trusted server for storage.

[0093] The following description uses the application of a data processing method based on remote sensing data provided in this embodiment in a remote sensing image recognition scenario as an example to further illustrate the data processing method based on remote sensing data provided in this embodiment. (See also...) Figure 3 A data processing method based on remote sensing data, applicable to remote sensing image recognition scenarios, includes the following steps.

[0094] In step S302, when the data server detects the target remote sensing image, it preprocesses the target remote sensing image and sends the processed target remote sensing image to the trusted server.

[0095] In step S304, the trusted server loads the encrypted model and transmits the target remote sensing image and the encrypted model to the TEE.

[0096] In step S306, the TEE uses the asymmetric private key to decrypt the encrypted model and obtain the crop type identification model.

[0097] Step S308: Input the target remote sensing image into the crop type identification model to identify the crop type and obtain the crop type of the plot.

[0098] Step S310: Encrypt the crop type of the plot using an asymmetric public key, and send the encrypted identification result obtained through encryption to the data server through a trusted server.

[0099] In step S312, the data server sends the encrypted identification result to the service domain server.

[0100] Subsequently, the service domain server uses an asymmetric private key to decrypt the encrypted identification result and stores the crop types of the plots obtained from the decryption.

[0101] In summary, the data processing method based on remote sensing data provided in this embodiment first obtains encrypted land parcel data obtained by encrypting the land parcel data with a public key generated by a trusted execution environment (TEA) sent by a first server, and then transmits it to the TEA. Next, the TEA decrypts the encrypted land parcel data using a private key paired with the public key, and uses the decryption result to label the crop types of the incoming remote sensing data to obtain remote sensing data samples. Then, the TEA trains a model based on the remote sensing data samples and encrypts the obtained crop type identification model. Finally, the TEA obtains and stores the encrypted model output by the TEA, and uses this encrypted model to identify crop types from the remote sensing data. This method protects the security of remote sensing data by preventing it from leaving the institutional domain, reduces the cost of purchasing remote sensing data, and uses encrypted data transmission during data processing to avoid data leakage.

[0102] This specification provides an example of a remote sensing data identification data processing method:

[0103] Reference Figure 4 It shows a flowchart of a remote sensing data identification data processing method provided in this embodiment, with reference to... Figure 2 It shows a timeline diagram of a data processing method based on remote sensing data applied to a model training scenario provided in this embodiment, with reference to... Figure 3 The diagram shows a timing diagram of a data processing method based on remote sensing data for application in remote sensing image recognition scenarios, as provided in this embodiment.

[0104] Reference Figure 4 The identification data processing method based on remote sensing data provided in this embodiment is applied to a first server and specifically includes the following steps S402 to S408.

[0105] Step S402: Receive the public key generated by the trusted execution environment sent by the second server.

[0106] In practical applications, the combination of satellite remote sensing and artificial intelligence has promoted the development and progress of agricultural technology. The application of agricultural technology often involves farmers' remote sensing data. To protect farmers' privacy data in compliance with regulations, this embodiment provides a data processing method based on remote sensing data. This method establishes a Trusted Execution Environment (TEE) in the institutional domain storing the remote sensing data, creating a joint data computation method between the institutional domain and the service domain. This completes the training and identification of remote sensing data, protecting the institutional domain's remote sensing data from leakage. Furthermore, the service domain's land parcel data is placed in encrypted form within the institutional domain, preventing the institutional domain from viewing the plaintext of the service domain's land parcel data. Calculations are performed on the institutional domain's remote sensing data and the service domain's land parcel data within the institutional domain, and the calculation results are sent to the service domain in encrypted form. Specifically, a model is trained in the institutional domain using the institutional domain's remote sensing data and the service domain's land parcel data to obtain a crop type identification model. This model is then used to identify remote sensing data of unknown crop types, and the identification results are encrypted before being sent to the service domain. This assures the institutional domain that no remote sensing data has been leaked, thus enhancing the protection of privacy data.

[0107] In this embodiment, the first server is a server deployed in the service domain; the first server stores the land parcel data; the first server encrypts the stored land parcel data using a public key generated and sent by the TEE deployed in the institutional domain, and sends the encrypted land parcel data to the second server; the second server is a server deployed in the institutional domain, specifically, the second server includes a data server and a trusted server; the data server refers to a server that stores remote sensing data; the trusted server refers to a server that stores the TEE; in addition, the second server may also include only one of the data server and the trusted server, the TEE may be directly stored in the data server storing remote sensing data; the remote sensing data may also be stored in the trusted server storing the TEE.

[0108] In practical implementation, to ensure data privacy and prevent other institutional domains from viewing the input TEE and TEE output data, ensuring that only designated institutional domains can view the data, in an optional implementation method provided in this embodiment, the key pair is obtained in the following way:

[0109] A key pair is generated in the trusted execution environment; the key pair includes the private key and the public key.

[0110] The trusted server obtains the public key output by the trusted execution environment and forwards the public key to the first server through the data server;

[0111] The key pair is sent by the second server to the key server for storage after it is generated.

[0112] Specifically, the TEE generates a key pair containing a public key and a private key and outputs it to a trusted server. The trusted server sends the generated public key to the first server in the service domain, so that the first server can encrypt the stored land parcel data based on the public key. In order for the first server to be able to decrypt the encrypted identification result after receiving the encrypted identification result output by the TEE, the trusted server sends the generated key to a third-party server (key server) jointly selected by the organization domain and the service domain for storage, so that the first server can call it.

[0113] Step S404: Encrypt the land parcel data according to the public key, and send the encrypted land parcel data to the second server.

[0114] The encrypted land parcel data is obtained by the first server encrypting the land parcel data based on a public key; the land parcel data includes the location data of the land parcel and the crop type data corresponding to the land parcel.

[0115] In specific implementation, after the encrypted plot data is sent to the second server, the second server trains a model based on the encrypted plot data and pre-stored remote sensing data, enabling the identification of crop types using the trained crop type recognition model. In one optional implementation of this embodiment, the crop type recognition model is obtained in the following manner:

[0116] The encrypted land parcel data sent by the first server is obtained and transmitted to the trusted execution environment; the encrypted land parcel data is obtained by the first server encrypting the land parcel data based on the public key;

[0117] In the trusted execution environment, the encrypted land parcel data is decrypted based on the private key paired with the public key, and the crop type of the land parcel is marked on the incoming remote sensing data based on the decryption result to obtain a remote sensing data sample.

[0118] The crop species identification model is obtained by training the model based on the remote sensing data samples in the trusted execution environment.

[0119] During the process of the first server sending encrypted land parcel data to the second server, the encrypted land parcel data is first sent to the data server, and then the data server sends it to the trusted server. The trusted server then transmits the encrypted land parcel data to the TEE. In one optional implementation of this embodiment, the encrypted land parcel data sent by the first server is obtained and transmitted to the trusted execution environment by performing the following steps: First, the data server obtains the encrypted land parcel data sent by the first server and forwards the encrypted land parcel data to the trusted server; then the trusted server transmits the encrypted land parcel data to the trusted execution environment through the connection interface.

[0120] For example, the TEE generates an asymmetric public-private key pair and outputs it to a trusted server. The trusted server sends the asymmetric public key to a data server, which then sends it to a server in the service domain. To prevent the data server from sending other asymmetric public keys that could lead to the leakage of land parcel data on the service domain server, the TEE and the service domain server jointly monitor the transmission process of the asymmetric public key. Upon receiving the asymmetric public key, the service domain server encrypts the land parcel data using the asymmetric public key and sends the encrypted land parcel data to the data server, which then forwards it to the trusted server. The server then transmits the encrypted land parcel data to the TEE via the PCI (Peripheral Component Interconnect) interface.

[0121] In practice, the trusted server trains the model based on remote sensing data labeled with crop types. To improve the efficiency of model training and reduce data processing during training, the remote sensing data of unlabeled crop types stored on the data server needs to be preprocessed before being passed to the TEE. This reduces the complexity of data processing during model training. The remote sensing data is passed to the TEE in the following way:

[0122] The initial remote sensing data is preprocessed in the data server to obtain the remote sensing data, and the remote sensing data is then sent to the trusted server.

[0123] The trusted server transmits the remote sensing data to the trusted execution environment through a connection interface;

[0124] The preprocessing of the initial remote sensing data includes at least one of the following:

[0125] The initial remote sensing data is subjected to radiometric correction, geometric correction, parameter extraction, and / or classification.

[0126] For example, the data server in the institutional domain eliminates or corrects distortions in remote sensing images caused by radiometric errors. It also uses a series of models to correct and eliminate distortions caused by factors such as photographic material deformation, objective lens distortion, atmospheric refraction, Earth curvature, Earth rotation, and topographic relief, which result in inconsistencies between the geometric position, shape, size, orientation, and other features of various objects in the original image and the representation requirements in the reference system. Parameters are extracted from the remote sensing images obtained after radiometric and geometric corrections, and the cultivated land portion in the remote sensing images is extracted. The final remote sensing images are then sent to a trusted server, which in turn transmits them to a TEE via the PCI interface.

[0127] It should be noted that the process of the data server sending encrypted land parcel data to the trusted server and the process of sending remote sensing data to the trusted server can be either to send the encrypted land parcel data to the trusted server first and then send the remote sensing data; or, upon receiving the encrypted land parcel data, to preprocess the initial remote sensing data and then send the preprocessed remote sensing data and the encrypted land parcel data together to the trusted server.

[0128] In practice, after receiving the encrypted land parcel data, the TEE needs to decrypt the encrypted land parcel data to obtain the actual land parcel data. Specifically, the encrypted land parcel data is decrypted in the following manner:

[0129] A private key query request is sent to the key server; the key server generates a private key viewing reminder based on the private key query request and sends it to the first server;

[0130] Obtain the private key sent by the key server when the first server submits a confirmation instruction for the private key viewing reminder;

[0131] The encrypted land parcel data is decrypted based on the private key.

[0132] Specifically, to ensure data privacy, the key pair generated by the TEE is stored in a key server jointly selected by the organization domain and the service domain. To prevent other servers from accessing the key pair and causing privacy data leakage, the key pair can only be accessed after confirmation by the organization domain and the service domain. It should be noted that the key pair is generated by the TEE and can also be stored in both the TEE and the key server. When the TEE needs to use the key pair, it can perform decryption, encryption, and other processing based on the stored key pair. When the server on the service side needs to use the key pair, it queries the key server.

[0133] For example, when the TEE receives incoming encrypted land parcel data, it generates and outputs a private key query request. The trusted server sends this request to a third-party server via the data server. The third-party server, based on the request, sends a private key viewing notification to the server in the service domain. Upon receiving confirmation from the server in the service domain regarding the notification, the third-party server sends the asymmetric private key to the trusted server, which then transmits it to the TEE. To prevent the data server from using this private key to decrypt the encrypted land parcel data, the encrypted land parcel data can be marked and set to be automatically deleted after detecting that the data server has sent it to the trusted server. Alternatively, access permissions can be set for the private key when it is sent by the key server, allowing only those with the required permissions to access it.

[0134] In practice, after decrypting the encrypted land parcel data to obtain the land parcel data, in order to train a model with crop type recognition capabilities, it is necessary to determine remote sensing data samples based on the land parcel data and remote sensing data, and then train the model based on the remote sensing data samples. Specifically, the remote sensing data samples are obtained in the following way:

[0135] Based on the location information in the decrypted land parcel data and the location information in the remote sensing data, the land parcel data and the remote sensing data are matched for location.

[0136] Crop type is marked at the successfully matched remote sensing plots based on the crop type information in the plot data.

[0137] Specifically, the first step is to match the location of the land parcel data with the location of the remote sensing data. In other words, the location in the land parcel data is matched with the location in the remote sensing data. If the location matches, the crop type information in the land parcel information is marked as the crop type information in the remote sensing data corresponding to the location to obtain the remote sensing data sample.

[0138] It should be noted that the labeling of crop types for plots in the incoming remote sensing data based on the decryption results can be done by the TEE; after the TEE decrypts the encrypted plot data, it labels the crop types for plots in the incoming remote sensing data based on the decryption results to obtain remote sensing data samples; alternatively, the TEE can decrypt the encrypted plot data, input the decryption results and remote sensing data into the model, and the model can label the crop types for plots in the incoming remote sensing data based on the decryption results to obtain remote sensing data samples, and then train the model based on the remote sensing data samples.

[0139] For example, the TEE decrypts the encrypted land parcel data using the private key sent by the key server to obtain the land parcel data. The land parcel data and remote sensing data are then input into the U-Net model. The U-Net model performs semantic segmentation on the remote sensing image based on the land parcel data to obtain remote sensing data samples.

[0140] In practical implementation, to make the crop species identification model more accurate and effective, after training the model to meet certain accuracy and recall rates, this model is used as the crop species identification model. Specifically, the model is trained in the following way:

[0141] The remote sensing data samples are divided into a training sample set and a test sample set;

[0142] At least one candidate model is obtained by training the model based on the training sample set;

[0143] The test sample set is input into the candidate model, and the evaluation parameters of each candidate model are determined based on the recognition results of each candidate model.

[0144] Candidate models whose evaluation parameters meet preset conditions are selected as crop type identification models.

[0145] In order to improve the efficiency of model training, the AI ​​accelerator in TEE is called to make the model training process more efficient.

[0146] For example, remote sensing data samples are divided into training sample set and test sample set according to a certain ratio. At least one candidate U-Net model is trained using the training sample set. The test sample set is input into each candidate U-Net model to obtain the accuracy and recall of each candidate U-Net model. The candidate U-Net model with higher accuracy and recall than other candidate U-Net models is determined as the crop type identification model for crop type identification.

[0147] It should be noted that, in order to further ensure the security of privacy data and prevent privacy data leakage, the obtained crop type identification model is also stored in an encrypted form. In order to facilitate the reloading of the trained crop type identification model after a power outage, and to avoid the loss of the crop type identification model due to power outage and to avoid resource consumption caused by repeated model training, after the crop type identification model is trained by TEE, the crop type identification model is first encrypted using the generated public key, and then the encrypted model is output.

[0148] In practice, after obtaining the encrypted model, the trusted server stores it. In addition, after obtaining the encrypted model, the trusted server can send it to the server in the service domain via the data server, enabling the service domain to identify crop types from the acquired remote sensing data. Specifically, after obtaining the encrypted model, the server in the service domain uses the private key stored on the key server to decrypt the encrypted model and stores the decrypted crop type identification model.

[0149] After training and obtaining the crop species identification model, the model can be used to identify crop species in remote sensing data that is not labeled with crop species. In one optional implementation of this embodiment, if the second server detects the target remote sensing data, the second server obtains the encrypted identification result of the crop species of the plot in the target remote sensing data in the following manner:

[0150] The initial target remote sensing data is preprocessed to obtain target remote sensing data, and the encryption model is then loaded.

[0151] The encryption model and the target remote sensing data are then transmitted to the trusted execution environment.

[0152] The encryption model is decrypted in the trusted execution environment, and the target remote sensing data is identified based on the crop type identification model obtained from the decryption.

[0153] The identified crop types for the plots are encrypted within the trusted execution environment and then output to the trusted server.

[0154] The trusted server receives the encrypted identification result and forwards it to the first server through the data server.

[0155] In the process of preprocessing the initial remote sensing data to be identified to obtain the target remote sensing data and loading the encrypted model, the second server first preprocesses the initial target remote sensing data in the data server and sends the preprocessed target remote sensing data to the trusted server. Upon receiving the target remote sensing data, the trusted server loads the encrypted model. After loading the encrypted model, the trusted server transmits both the encrypted model and the target remote sensing data to the TEE. The TEE uses a private key to decrypt the encrypted model and inputs the target remote sensing data into the decrypted crop type identification model.

[0156] Furthermore, after inputting the target remote sensing data into the decrypted crop type identification model, the trusted execution environment encrypts the identified crop types of the plots and outputs them to the trusted server; the trusted server receives the encrypted identification result and forwards the encrypted identification result to the first server through the data server.

[0157] For example, during the model training phase, the input remote sensing data samples are obtained by semantic segmentation of remote sensing data and land parcel data corresponding to province P. After obtaining the crop type identification model, it is necessary to identify crop types from remote sensing data across the country. The data server preprocesses the remote sensing data across the country and sends the preprocessed remote sensing data to the trusted server. After receiving the remote sensing data, the trusted server loads the encrypted model and transmits the loaded encrypted model and remote sensing data to the TEE through the PCI interface. The TEE decrypts the encrypted model using the generated asymmetric private key and inputs the remote sensing data into the decrypted crop type identification model. After the crop type identification module outputs the identification result, to prevent the data server from viewing the identification result, the TEE uses the generated asymmetric public key to encrypt and output the identified crop types of the land parcels. After obtaining the encrypted identification result, the trusted server sends the encrypted identification result to the server in the service domain through the data server.

[0158] Step S406: Obtain the encrypted identification result of the crop type of the plot in the target remote sensing data sent by the second server.

[0159] The crop type of the plot is obtained by the second server using a crop type identification model to identify the crop type of the incoming target remote sensing data in the trusted execution environment; the encrypted identification result is obtained by the trusted execution environment encrypting the crop type of the plot according to the public key.

[0160] Step S408: Decrypt the encrypted identification result using the private key paired with the public key to obtain the crop type of the plot, so as to determine the crop type of the plot corresponding to the plot labeling data submitted by the user.

[0161] In practice, the first server cannot view the actual recognition result after receiving the encrypted recognition result; it needs to decrypt the encrypted recognition result before viewing it. In one optional implementation of this embodiment, the process of decrypting the encrypted recognition result using the private key paired with the public key is achieved by executing the following steps:

[0162] Send a private key query request to the key server;

[0163] Obtain the private key generated by the trusted execution environment, which is sent by the key server upon receiving confirmation instructions from the second server;

[0164] The encrypted identification result is decrypted based on the private key, and the crop type of the plot obtained from the decryption is stored.

[0165] Specifically, after the first server stores the target remote sensing data and the corresponding crop types for the plots, in an optional implementation provided in this embodiment, if target plot annotation data submitted by a target user is detected, the target crop type corresponding to the target plot annotation data is queried based on the crop type of the plot; then, a target service quota is determined based on the target crop type and issued to the target user. Specifically, firstly, the corresponding remote sensing location is determined based on the location information of the target plot annotation data; then, the crop type corresponding to that remote sensing location is queried; the queried crop type is used as the crop type corresponding to the target plot annotation data; and the target service quota corresponding to that crop type is determined according to preset service rules and issued to the target user.

[0166] The following example illustrates the application of a remote sensing data-based identification data processing method in a model training scenario, further illustrating the remote sensing data-based identification data processing method provided in this embodiment. (See also...) Figure 2 A remote sensing-based identification data processing method applied to model training scenarios includes the following steps.

[0167] In step S202, the service domain server encrypts the land parcel data using the asymmetric public key generated by the trusted server.

[0168] Step S204: Send the encrypted land parcel data to the data server.

[0169] Subsequently, upon receiving the encrypted plot data, the data server preprocesses the initial remote sensing image. The data server then sends the encrypted plot data and the preprocessed remote sensing image to the trusted server, which transmits it to the TEE (Trusted Execution Environment). The TEE decrypts the encrypted plot data using the generated asymmetric private key. The decryption result and the remote sensing image are input into the model. The model performs semantic segmentation on the remote sensing image based on the decryption result to obtain remote sensing image samples. The model is then trained based on the remote sensing image samples to obtain a crop type identification model. The crop type identification model is then encrypted using the generated asymmetric public key, and the encrypted model is sent to the trusted server for storage.

[0170] The following description uses the application of a remote sensing data-based identification data processing method provided in this embodiment in a remote sensing image recognition scenario as an example to further illustrate the remote sensing data-based identification data processing method provided in this embodiment. (See also...) Figure 3 A remote sensing data processing method for remote sensing image recognition scenarios includes the following steps.

[0171] In step S314, the service domain server uses an asymmetric private key to decrypt the encrypted identification result and stores the crop types of the plot obtained from the decryption.

[0172] Prior to this, upon detecting the target remote sensing image, the data server preprocesses the target remote sensing image and sends the processed image to the trusted server. The trusted server loads the encryption model and transmits the target remote sensing image and the encryption model to the TEE. The TEE decrypts the encryption model using an asymmetric private key to obtain a crop type identification model. The target remote sensing image is then input into the crop type identification model to identify the crop type of the plot. The crop type of the plot is then encrypted using an asymmetric public key, and the encrypted identification result is sent to the data server through the trusted server. The data server then sends the encrypted identification result to the service domain server.

[0173] In summary, the remote sensing data identification processing method provided in this embodiment first receives a public key generated by a trusted execution environment sent by a second server. Then, it encrypts the land parcel data according to the public key and sends the encrypted land parcel data to the second server. Next, it obtains the encrypted identification result of the crop type of the target remote sensing data sent by the second server. Finally, it decrypts the encrypted identification result using the private key paired with the public key to obtain the crop type of the land parcel, thereby determining the crop type of the land parcel corresponding to the user-submitted land parcel labeling data. In this way, it performs joint calculations on the data of the institutional domain and the service domain while protecting the data of the institutional domain and the service domain from being leaked.

[0174] This specification provides an embodiment of a data processing device based on remote sensing data, as follows:

[0175] In the above embodiments, a data processing method based on remote sensing data is provided, and correspondingly, a data processing device based on remote sensing data is also provided, which will be described below with reference to the accompanying drawings.

[0176] Reference Figure 5 The diagram shows a data processing device based on remote sensing data provided in this embodiment.

[0177] Since the apparatus embodiments correspond to the method embodiments, the descriptions are relatively simple. For relevant parts, please refer to the corresponding descriptions of the method embodiments provided above. The apparatus embodiments described below are merely illustrative.

[0178] This embodiment provides a data processing device based on remote sensing data, including:

[0179] The acquisition module 502 is configured to acquire encrypted land parcel data sent by the first server and pass it into the trusted execution environment; the encrypted land parcel data is obtained by the first server encrypting the land parcel data based on a public key;

[0180] The decryption module 504 is configured to decrypt the encrypted land parcel data based on the private key paired with the public key in the trusted execution environment, and to mark the crop type of the land parcel in the incoming remote sensing data based on the decryption result to obtain a remote sensing data sample.

[0181] Training module 506 is configured to train a model based on the remote sensing data samples in the trusted execution environment and to encrypt the crop species identification model obtained through training.

[0182] Storage module 508 is configured to acquire and store the encryption model output by the trusted execution environment.

[0183] This specification provides an embodiment of a remote sensing data identification and processing device:

[0184] In the above embodiments, a method for identifying data processing based on remote sensing data is provided, and correspondingly, a device for identifying data processing based on remote sensing data is also provided, which will be described below with reference to the accompanying drawings.

[0185] Reference Figure 6 The diagram shows a schematic of a remote sensing data identification processing device provided in this embodiment.

[0186] Since the apparatus embodiments correspond to the method embodiments, the descriptions are relatively simple. For relevant parts, please refer to the corresponding descriptions of the method embodiments provided above. The apparatus embodiments described below are merely illustrative.

[0187] This embodiment provides a remote sensing data identification processing device, including:

[0188] The receiving module 602 is configured to receive a public key generated by a trusted execution environment sent by a second server;

[0189] The encryption module 604 is configured to encrypt the land parcel data according to the public key and send the encrypted land parcel data to the second server.

[0190] The acquisition module 606 is configured to acquire the encrypted identification result of the crop type of the plot in the target remote sensing data sent by the second server; the crop type is obtained by the second server using a crop type identification model to identify the crop type of the incoming target remote sensing data in the trusted execution environment; the encrypted identification result is obtained by the trusted execution environment encrypting the crop type of the plot according to the public key.

[0191] The decryption module 608 is configured to decrypt the encrypted identification result using a private key paired with the public key to obtain the crop type of the plot, so as to determine the crop type of the plot corresponding to the plot labeling data submitted by the user.

[0192] This specification provides an embodiment of a data processing device based on remote sensing data as follows:

[0193] Corresponding to the data processing method based on remote sensing data described above, and based on the same technical concept, one or more embodiments of this specification also provide a data processing device based on remote sensing data, which is used to execute the data processing method based on remote sensing data provided above. Figure 7 This is a schematic diagram of the structure of a data processing device based on remote sensing data, provided for one or more embodiments of this specification.

[0194] This embodiment provides a data processing device based on remote sensing data, comprising:

[0195] like Figure 7 As shown, data processing devices based on remote sensing data can vary significantly due to differences in configuration or performance. They may include one or more processors 701 and a memory 702, where one or more application programs or data can be stored. The memory 702 can be temporary or persistent storage. The application programs stored in the memory 702 may include one or more modules (not shown), each module including a series of computer-executable instructions for the remote sensing data processing device. Furthermore, the processor 701 may be configured to communicate with the memory 702, executing the series of computer-executable instructions in the memory 702 on the remote sensing data processing device. The remote sensing data processing device may also include one or more power supplies 703, one or more wired or wireless network interfaces 704, one or more input / output interfaces 705, one or more keyboards 706, etc.

[0196] In one specific embodiment, the data processing device based on remote sensing data includes a memory and one or more programs, wherein the one or more programs are stored in the memory, and the one or more programs may include one or more modules, and each module may include a series of computer-executable instructions for the data processing device based on remote sensing data, and is configured to be executed by one or more processors. The one or more programs include computer-executable instructions for performing the following:

[0197] The encrypted land parcel data sent by the first server is obtained and transmitted to the trusted execution environment; the encrypted land parcel data is obtained by the first server encrypting the land parcel data based on the public key;

[0198] In the trusted execution environment, the encrypted land parcel data is decrypted based on the private key paired with the public key, and the crop type of the land parcel is marked on the incoming remote sensing data based on the decryption result to obtain a remote sensing data sample.

[0199] In the trusted execution environment, the model is trained based on the remote sensing data samples, and the crop species identification model obtained from the training is encrypted.

[0200] Acquire and store the encryption model output by the trusted execution environment.

[0201] This specification provides an embodiment of a remote sensing data identification and processing device as follows:

[0202] Corresponding to the remote sensing data-based identification data processing method described above, and based on the same technical concept, one or more embodiments of this specification also provide a remote sensing data-based identification data processing device, which is used to execute the remote sensing data-based identification data processing method described above. Figure 8 This is a schematic diagram of the structure of a remote sensing data identification processing device provided for one or more embodiments of this specification.

[0203] This embodiment provides a remote sensing data processing device for identification, including:

[0204] like Figure 8As shown, identification data processing devices based on remote sensing data can vary significantly due to differences in configuration or performance. They may include one or more processors 801 and a memory 802, where one or more application programs or data can be stored. The memory 802 can be temporary or persistent storage. The application programs stored in the memory 802 may include one or more modules (not shown), each module including a series of computer-executable instructions for the identification data processing device. Furthermore, the processor 801 may be configured to communicate with the memory 802, executing the series of computer-executable instructions in the memory 802 on the identification data processing device. The identification data processing device may also include one or more power supplies 803, one or more wired or wireless network interfaces 804, one or more input / output interfaces 805, one or more keyboards 806, etc.

[0205] In one specific embodiment, the identification data processing device based on remote sensing data includes a memory and one or more programs, wherein one or more programs are stored in the memory, and one or more programs may include one or more modules, and each module may include a series of computer-executable instructions for the identification data processing device based on remote sensing data, and is configured to be executed by one or more processors. The one or more programs include computer-executable instructions for performing the following:

[0206] Receive the public key generated by the trusted execution environment sent by the second server;

[0207] The land parcel data is encrypted using the public key, and the encrypted land parcel data is sent to the second server.

[0208] The encrypted identification result of the crop type of the target remote sensing data sent by the second server is obtained; the crop type of the plot is obtained by the second server using a crop type identification model to identify the crop type of the incoming target remote sensing data in the trusted execution environment; the encrypted identification result is obtained by the trusted execution environment encrypting the crop type of the plot according to the public key.

[0209] The encrypted identification result is decrypted using the private key paired with the public key to obtain the crop type of the plot, thereby determining the crop type of the plot corresponding to the plot labeling data submitted by the user.

[0210] This specification provides an example of a storage medium as follows:

[0211] Corresponding to the data processing method based on remote sensing data described above, and based on the same technical concept, one or more embodiments of this specification also provide a storage medium.

[0212] The storage medium provided in this embodiment is used to store computer-executable instructions, which, when executed, implement the following process:

[0213] The encrypted land parcel data sent by the first server is obtained and transmitted to the trusted execution environment; the encrypted land parcel data is obtained by the first server encrypting the land parcel data based on the public key;

[0214] In the trusted execution environment, the encrypted land parcel data is decrypted based on the private key paired with the public key, and the crop type of the land parcel is marked on the incoming remote sensing data based on the decryption result to obtain a remote sensing data sample.

[0215] In the trusted execution environment, the model is trained based on the remote sensing data samples, and the crop species identification model obtained from the training is encrypted.

[0216] Acquire and store the encryption model output by the trusted execution environment.

[0217] It should be noted that the embodiments concerning storage media in this specification and the embodiments concerning data processing methods based on remote sensing data in this specification are based on the same inventive concept. Therefore, the specific implementation of this embodiment can be referred to the implementation of the corresponding method described above, and the repeated parts will not be described again.

[0218] This specification provides an example of a storage medium as follows:

[0219] Corresponding to the above-described method for identifying data based on remote sensing data, and based on the same technical concept, one or more embodiments of this specification also provide a storage medium.

[0220] The storage medium provided in this embodiment is used to store computer-executable instructions, which, when executed, implement the following process:

[0221] Receive the public key generated by the trusted execution environment sent by the second server;

[0222] The land parcel data is encrypted using the public key, and the encrypted land parcel data is sent to the second server.

[0223] The encrypted identification result of the crop type of the target remote sensing data sent by the second server is obtained; the crop type of the plot is obtained by the second server using a crop type identification model to identify the crop type of the incoming target remote sensing data in the trusted execution environment; the encrypted identification result is obtained by the trusted execution environment encrypting the crop type of the plot according to the public key.

[0224] The encrypted identification result is decrypted using the private key paired with the public key to obtain the crop type of the plot, thereby determining the crop type of the plot corresponding to the plot labeling data submitted by the user.

[0225] It should be noted that the embodiments concerning storage media in this specification and the embodiments concerning identification data processing methods based on remote sensing data in this specification are based on the same inventive concept. Therefore, the specific implementation of this embodiment can be referred to the implementation of the corresponding method described above, and the repeated parts will not be described again.

[0226] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0227] In the 1930s, improvements to a technology could be clearly distinguished as either hardware improvements (e.g., improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the methodology). However, with technological advancements, many improvements to the methodology today can be considered direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved methodology into the hardware circuit. Therefore, it cannot be said that an improvement to the methodology cannot be implemented using a hardware physical module. For example, a Programmable Logic Device (PLD) (e.g., a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logic function is determined by the user programming the device. Designers can program a digital system themselves to "integrate" it onto a PLD, without needing chip manufacturers to design and manufacture dedicated integrated circuit chips. Furthermore, nowadays, instead of manually manufacturing integrated circuit chips, this programming is mostly implemented using "logic compiler" software. Similar to the software compiler used in program development, the original code before compilation must be written in a specific programming language, called a Hardware Description Language (HDL). There are many HDLs, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, and RHDL (Ruby Hardware Description Language). Currently, the most commonly used are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should understand that by simply performing some logic programming on the method flow using one of these hardware description languages ​​and programming it into an integrated circuit, the hardware circuit implementing the logical method flow can be easily obtained.

[0228] The controller can be implemented in any suitable manner. For example, it can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicon Labs C8051F320. A memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also recognize that, in addition to implementing the controller in purely computer-readable program code form, the same functionality can be achieved by logically programming the method steps to make the controller take the form of logic gates, switches, ASICs, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0229] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0230] For ease of description, the above apparatus is described by dividing it into various functional units. Of course, when implementing the embodiments of this specification, the functions of each unit can be implemented in one or more software and / or hardware.

[0231] Those skilled in the art will understand that one or more embodiments of this specification can be provided as a method, system, or computer program product. Therefore, one or more embodiments of this specification may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this specification may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0232] This specification is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this specification. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable remote sensing data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable remote sensing data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0233] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable remote sensing data processing device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0234] These computer program instructions can also be loaded onto a computer or other programmable data processing device based on remote sensing data, causing a series of operational steps to be executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable device for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0235] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, a network interface, and memory. Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0236] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0237] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. One or more embodiments of this specification can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a particular task or implement a particular abstract data type. One or more embodiments of this specification can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In a distributed computing environment, program modules can reside in local and remote computer storage media, including storage devices. The various embodiments in this specification are described in a progressive manner, with reference to each other for similar or identical parts. Each embodiment focuses on describing the differences from other embodiments. In particular, the system implementation is basically similar to the method implementation, so the description is relatively simple. For relevant details, please refer to the description of the method implementation.

[0238] The above description is merely an embodiment of this document and is not intended to limit the scope of this document. Various modifications and variations can be made to this document by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this document should be included within the scope of the claims of this document.

Claims

1. A data processing method based on remote sensing data, applied to a second server deployed in an institutional domain, the second server comprising a data server and a trusted server, the data server and the trusted server being deployed in the institutional domain, the method comprising: Obtain the encrypted land parcel data sent by the first server and pass it into the trusted execution environment; The first server is deployed in the service domain; In the trusted execution environment, the encrypted plot data is decrypted, and the incoming remote sensing data is labeled with plot crop types based on the decryption result to obtain remote sensing data samples. The plot crop type labeling includes labeling the remote sensing data that matches the location of the plot data according to the crop type information in the plot data. In the trusted execution environment, the model is trained based on the remote sensing data samples, and the crop species identification model obtained from the training is encrypted. Acquire and store the encryption model output by the trusted execution environment; Prior to the step of obtaining the encrypted land parcel data sent by the first server and transmitting it to the trusted execution environment, the process also includes: A key pair is generated in the trusted execution environment; the key pair includes a private key and a public key. The trusted server obtains the public key output by the trusted execution environment and forwards the public key to the first server through the data server. The key pair is then sent by the second server to the key server for storage after it is generated.

2. The data processing method based on remote sensing data according to claim 1, wherein obtaining the encrypted land parcel data sent by the first server and transmitting it to the trusted execution environment includes: The data server obtains the encrypted land parcel data sent by the first server and forwards the encrypted land parcel data to the trusted server; The trusted server transmits the encrypted land parcel data to the trusted execution environment through a connection interface.

3. The data processing method based on remote sensing data according to claim 1, before the step of decrypting the encrypted plot data in the trusted execution environment and marking the crop types of the incoming remote sensing data based on the decryption result to obtain a remote sensing data sample, further includes: The initial remote sensing data is preprocessed in the data server to obtain the remote sensing data, and the remote sensing data is then sent to the trusted server. The trusted server transmits the remote sensing data to the trusted execution environment through a connection interface; The preprocessing of the initial remote sensing data includes at least one of the following: The initial remote sensing data is subjected to radiometric correction, geometric correction, parameter extraction, and / or classification.

4. The data processing method based on remote sensing data according to claim 1, before the step of obtaining the encrypted land parcel data sent by the first server and transmitting it to the trusted execution environment, further includes: A key pair is generated in the trusted execution environment; the key pair includes a private key and a public key. The trusted server obtains the public key output by the trusted execution environment and forwards the public key to the first server through the data server; The key pair is sent by the second server to the key server for storage after it is generated.

5. The data processing method based on remote sensing data according to claim 1, wherein the step of training the model based on the remote sensing data samples in the trusted execution environment includes: The remote sensing data samples are divided into a training sample set and a test sample set; At least one candidate model is obtained by training the model based on the training sample set; The test sample set is input into the candidate model, and the evaluation parameters of each candidate model are determined based on the recognition results of each candidate model. Candidate models whose evaluation parameters meet preset conditions are selected as crop type identification models.

6. The data processing method based on remote sensing data according to claim 1, wherein decrypting the encrypted land parcel data in the trusted execution environment includes: Send a private key query request to the key server; The key server generates a private key viewing reminder based on the private key query request and sends it to the first server. Obtain the private key sent by the key server when the first server submits a confirmation instruction for the private key viewing reminder; The encrypted land parcel data is decrypted based on the private key.

7. The data processing method based on remote sensing data according to claim 1, after the step of acquiring and storing the encryption model output by the trusted execution environment is performed, it further includes: The initial target remote sensing data is preprocessed to obtain target remote sensing data, and the encryption model is then loaded. The encryption model and the target remote sensing data are then transmitted to the trusted execution environment. The encryption model is decrypted in the trusted execution environment, and the target remote sensing data is identified based on the crop type identification model obtained from the decryption.

8. The data processing method based on remote sensing data according to claim 7, after the step of decrypting the encryption model in the trusted execution environment and identifying the crop type of the target remote sensing data according to the decrypted crop type identification model, the method further includes: The identified crop types for the plots are encrypted within the trusted execution environment and then output to the trusted server. The trusted server receives the encrypted identification result and forwards it to the first server through the data server.

9. The data processing method based on remote sensing data according to claim 7, wherein the step of preprocessing the initial target remote sensing data to obtain the target remote sensing data and loading the encryption model includes: The initial target remote sensing data is preprocessed in the data server, and the preprocessed target remote sensing data is sent to the trusted server. The trusted server loads the encryption model upon receiving the target remote sensing data.

10. The data processing method based on remote sensing data according to claim 8, after the first server receives the encrypted identification result, it performs the following operations: Send a private key query request to the key server; Obtain the private key generated by the trusted execution environment, which is sent by the key server upon receiving a confirmation instruction from the second server; The encrypted identification result is decrypted based on the private key, and the crop type of the plot corresponding to the target remote sensing data obtained after decryption is stored.

11. A method for processing identification data based on remote sensing data, applied to a first server deployed in a service domain, the method comprising: Receive the public key generated by the trusted execution environment sent by the second server; The second server is deployed in the organization domain, and the second server includes a data server and a trusted server, wherein the data server and the trusted server are deployed in the organization domain; The land parcel data is encrypted using the public key, and the encrypted land parcel data is sent to the second server. The encrypted identification result of the crop type of the target remote sensing data sent by the second server is obtained; the crop type of the plot is obtained by the second server using a crop type identification model to identify the crop type of the incoming target remote sensing data in the trusted execution environment; the encrypted identification result is obtained by the trusted execution environment encrypting the crop type of the plot according to the public key; the remote sensing data sample of the crop type identification model is obtained by marking the remote sensing data that matches the location of the plot data with crop type information in the plot data. The encrypted identification result is decrypted using the private key paired with the public key to obtain the crop type of the plot, thereby determining the crop type of the plot corresponding to the plot labeling data submitted by the user; Before the step of receiving the public key generated by the trusted execution environment sent by the second server is executed, the second server performs the following operations: A key pair is generated in the trusted execution environment; the key pair includes the private key and the public key. The trusted server obtains the public key output by the trusted execution environment and forwards the public key to the first server through the data server.

12. The identification data processing method based on remote sensing data according to claim 11 further includes: If target plot labeling data submitted by the target user is detected, the target crop type corresponding to the target plot labeling data is queried based on the crop type of the plot. The target service quota is determined based on the target crop type and then disbursed to the target user.

13. The identification data processing method based on remote sensing data according to claim 11, wherein decrypting the encrypted identification result using a private key paired with the public key to obtain the crop type of the plot includes: Send a private key query request to the key server; Obtain the private key generated by the trusted execution environment, which is sent by the key server upon receiving confirmation instructions from the second server; The encrypted identification result is decrypted based on the private key, and the crop type of the plot obtained from the decryption is stored.

14. The identification data processing method based on remote sensing data according to claim 11, wherein the key pair is sent by the second server to the key server for storage after it is generated.

15. The identification data processing method based on remote sensing data according to claim 13, wherein the second server obtains and sends the encrypted identification result in the following manner: Preprocess the initial target remote sensing data to obtain target remote sensing data, and load the encrypted model; The encryption model and the target remote sensing data are then transmitted to the trusted execution environment. The encryption model is decrypted in the trusted execution environment, and the target remote sensing data is identified based on the crop type identification model obtained from the decryption. The identified crop types for the plots are encrypted within the trusted execution environment and then output to the trusted server. The trusted server receives the encrypted identification result and forwards it to the first server through the data server.

16. The crop species identification model based on remote sensing data according to claim 11 is obtained in the following manner: The encrypted land parcel data sent by the first server is obtained and transmitted to the trusted execution environment; the encrypted land parcel data is obtained by the first server encrypting the land parcel data based on the public key; In the trusted execution environment, the encrypted land parcel data is decrypted based on the private key paired with the public key, and the crop type of the land parcel is marked on the incoming remote sensing data based on the decryption result to obtain a remote sensing data sample. The crop species identification model is obtained by training the model based on the remote sensing data samples in the trusted execution environment.

17. A data processing apparatus based on remote sensing data, disposed on a second server deployed in an institutional domain, the second server including a data server and a trusted server, the data server and the trusted server being deployed in the institutional domain, the apparatus comprising: The acquisition module is configured to acquire encrypted land parcel data sent by the first server and pass it into the trusted execution environment; The first server is deployed in the service domain; The decryption module is configured to decrypt the encrypted plot data in the trusted execution environment, and to mark the crop types of the incoming remote sensing data based on the decryption result to obtain remote sensing data samples. The crop type marking includes marking the remote sensing data that matches the location of the plot data according to the crop type information in the plot data. The training module is configured to train a model based on the remote sensing data samples in the trusted execution environment and to encrypt the crop species identification model obtained through training. The storage module is configured to acquire and store the cryptographic model output by the trusted execution environment; The device further includes: A generative model is configured to generate key pairs in the trusted execution environment; the key pairs include a private key and a public key. The forwarding module is configured so that the trusted server obtains the public key output by the trusted execution environment and forwards the public key to the first server through the data server, wherein the key pair is sent by the second server to the key server for storage after generation.

18. A remote sensing data identification processing device, disposed on a first server deployed in a service domain, the device comprising: The receiving module is configured to receive a public key generated by a trusted execution environment and sent by a second server; The second server is deployed in the organization domain, and the second server includes a data server and a trusted server, wherein the data server and the trusted server are deployed in the organization domain; The encryption module is configured to encrypt the land parcel data according to the public key and send the encrypted land parcel data to the second server. The acquisition module is configured to acquire the encrypted identification result of the crop type of the target remote sensing data sent by the second server; the crop type of the plot is obtained by the second server using a crop type identification model to identify the crop type of the incoming target remote sensing data in the trusted execution environment; the encrypted identification result is obtained by the trusted execution environment encrypting the crop type of the plot according to the public key; the remote sensing data sample of the crop type identification model is obtained by marking the remote sensing data that matches the location of the plot data with crop type information in the plot data. The decryption module is configured to decrypt the encrypted identification result using a private key paired with the public key to obtain the crop type of the plot, so as to determine the crop type of the plot corresponding to the plot labeling data submitted by the user; The device further includes: A generation module is configured to generate a key pair in the trusted execution environment; the key pair includes the private key and the public key. The forwarding module is configured to allow the trusted server to obtain the public key output by the trusted execution environment and forward the public key to the first server through the data server.

19. A data processing device based on remote sensing data, disposed on a second server deployed in an institutional domain, the second server including a data server and a trusted server, the data server and the trusted server being deployed in the institutional domain, the device comprising: processor; as well as, A memory configured to store computer-executable instructions, which, when executed, cause the processor to: The encrypted land parcel data sent by the first server is obtained and transmitted to the trusted execution environment; the first server is deployed in the service domain; In the trusted execution environment, the encrypted plot data is decrypted, and the incoming remote sensing data is labeled with plot crop types based on the decryption result to obtain remote sensing data samples. The plot crop type labeling includes labeling the remote sensing data that matches the location of the plot data according to the crop type information in the plot data. In the trusted execution environment, the model is trained based on the remote sensing data samples, and the crop species identification model obtained from the training is encrypted. Acquire and store the encryption model output by the trusted execution environment; Prior to the step of obtaining the encrypted land parcel data sent by the first server and transmitting it to the trusted execution environment, the process also includes: A key pair is generated in the trusted execution environment; the key pair includes a private key and a public key. The trusted server obtains the public key output by the trusted execution environment and forwards the public key to the first server through the data server. The key pair is then sent by the second server to the key server for storage after it is generated.

20. A remote sensing data identification processing device, configured on a first server deployed in a service domain, the device comprising: processor; as well as, A memory configured to store computer-executable instructions, which, when executed, cause the processor to: Receive the public key generated by the trusted execution environment sent by the second server; The second server is deployed in the organization domain, and the second server includes a data server and a trusted server, wherein the data server and the trusted server are deployed in the organization domain; The land parcel data is encrypted using the public key, and the encrypted land parcel data is sent to the second server. The encrypted identification result of the crop type of the target remote sensing data sent by the second server is obtained; the crop type of the plot is obtained by the second server using a crop type identification model to identify the crop type of the incoming target remote sensing data in the trusted execution environment; the encrypted identification result is obtained by the trusted execution environment encrypting the crop type of the plot according to the public key; the remote sensing data sample of the crop type identification model is obtained by marking the remote sensing data that matches the location of the plot data with crop type information in the plot data. The encrypted identification result is decrypted using the private key paired with the public key to obtain the crop type of the plot, thereby determining the crop type of the plot corresponding to the plot labeling data submitted by the user; Before the step of receiving the public key generated by the trusted execution environment sent by the second server is executed, the second server performs the following operations: A key pair is generated in the trusted execution environment; the key pair includes the private key and the public key. The trusted server obtains the public key output by the trusted execution environment and forwards the public key to the first server through the data server.

21. A storage medium disposed on a second server deployed in an institutional domain, the second server including a data server and a trusted server, the data server and the trusted server being deployed in the institutional domain, the storage medium being used to store computer-executable instructions, the computer-executable instructions, when executed, performing the following process: The encrypted land parcel data sent by the first server is obtained and transmitted to the trusted execution environment; the first server is deployed in the service domain; In the trusted execution environment, the encrypted plot data is decrypted, and the incoming remote sensing data is labeled with plot crop types based on the decryption result to obtain remote sensing data samples. The plot crop type labeling includes labeling the remote sensing data that matches the location of the plot data according to the crop type information in the plot data. In the trusted execution environment, the model is trained based on the remote sensing data samples, and the crop species identification model obtained from the training is encrypted. Acquire and store the encryption model output by the trusted execution environment; in, Before the step of obtaining the encrypted land parcel data sent by the first server and transmitting it to the trusted execution environment, the method further includes: A key pair is generated in the trusted execution environment; the key pair includes a private key and a public key. The trusted server obtains the public key output by the trusted execution environment and forwards the public key to the first server through the data server. The key pair is then sent by the second server to the key server for storage after it is generated.

22. A storage medium disposed on a first server deployed in a service domain, the storage medium being used to store computer-executable instructions, which, when executed, perform the following process: Receive the public key generated by the trusted execution environment sent by the second server; The second server is deployed in the organization domain, and the second server includes a data server and a trusted server, wherein the data server and the trusted server are deployed in the organization domain; The land parcel data is encrypted using the public key, and the encrypted land parcel data is sent to the second server. The encrypted identification result of the crop type of the target remote sensing data sent by the second server is obtained; the crop type of the plot is obtained by the second server using a crop type identification model to identify the crop type of the incoming target remote sensing data in the trusted execution environment; the encrypted identification result is obtained by the trusted execution environment encrypting the crop type of the plot according to the public key; the remote sensing data sample of the crop type identification model is obtained by marking the remote sensing data that matches the location of the plot data with crop type information in the plot data. The encrypted identification result is decrypted using the private key paired with the public key to obtain the crop type of the plot, thereby determining the crop type of the plot corresponding to the plot labeling data submitted by the user; Before the step of receiving the public key generated by the trusted execution environment sent by the second server is executed, the second server performs the following operations: A key pair is generated in the trusted execution environment; the key pair includes the private key and the public key. The trusted server obtains the public key output by the trusted execution environment and forwards the public key to the first server through the data server.

Images