Remote sensing image processing system, method and device

Abstract

The invention discloses a remote sensing image processing system, method and device. The system comprises a data management platform, a plurality of data processing units and a data correction unit, each data processing unit is in communication connection with the data management platform and the plurality of data correction units; the data management platform is used for distributing a to-be-processed data set SET1 in the collected original remote sensing image data to the data processing unit; the data processing unit is used for performing image-text interpretation on the data set SET1 to obtain a data set SET2 and then distributing the data set SET2 to the data correction unit; and the data correction unit is used for correcting the data set SET2 to obtain a target data set. Therefore, after the original image data set is screened through the data management platform, the original image data set is distributed to the data processing unit for automatic image-text interpretation, and then the original image data set is distributed to the data correction unit for correction, so that closed-loop management of online interpretation and correction is realized, screening, interpretation and correction are carried out step by step, and the interpretation quality and the processing efficiency are improved.

Description

Technical Field

[0001] The present application relates to the field of remote sensing image technology, and in particular to a remote sensing image processing system, method and device. Background Art

[0002] In the field of remote sensing, remote sensing images are interpreted graphically to obtain information about ground objects and stored in a remote sensing image sample library, so that the remote sensing image sample library can be applied to scenarios such as environmental monitoring, urban planning, natural disaster response, and scientific research.

[0003] At present, the establishment of a remote sensing image sample library mainly includes steps such as image screening, graphic interpretation, and interpretation result inspection, and many steps are implemented manually offline. In this way, the data between different processing personnel are separated from each other, and the processing efficiency is low. It is impossible to track historical operations and it is difficult to meet the processing needs of massive remote sensing images. In addition, the quality of the results of manual graphic interpretation is low, which leads to the low quality of the remote sensing image sample library, which cannot meet the business needs of multiple scenarios.

[0004] It can be seen that how to improve the quality and processing efficiency of remote sensing image text interpretation, thereby improving the quality of the remote sensing image sample library, is an urgent problem to be solved by technical personnel in this field. Summary of the invention

[0005] One aspect of the present application provides a remote sensing image processing system, the system comprising: a data management platform, a data processing unit and a data correction unit; wherein the data processing unit comprises a plurality of;

[0006] The data processing unit is communicatively connected to the data management platform; and each of the data processing units is communicatively connected to a plurality of the data correction units;

[0007] The data management platform is used to collect raw remote sensing image data and distribute the data set SET1 to be processed in the raw remote sensing image data to the data processing unit;

[0008] The data processing unit is used to perform graphic interpretation on the data set SET1 to obtain a data set SET2; and distribute the data set SET2 to the data correction unit;

[0009] The data correction unit is used to correct the image-text interpretation result corresponding to the data set SET2 to obtain a target data set.

[0010] Optionally, the data management platform includes an analysis module, a statistics module, a data storage module and a display module;

[0011] The analysis module is used to obtain the target data set and perform image analysis on the target data set to obtain analysis results; wherein the image analysis includes image classification, image area division and target detection;

[0012] The statistical module is used to collect statistics on data processing information of the data processing unit and the data correction unit; wherein the data processing information includes data processing volume, data processing results, data processing quality and data processing progress;

[0013] The data storage module is used to store the original remote sensing image data, the target data set, the analysis results and the data processing information;

[0014] The display module is used to display the data processing information.

[0015] Optionally, the data management platform further includes a first user management module and a first screening module;

[0016] The first user management module is used to perform user management on the number of the communication-connected data processing units; wherein the user management includes user deletion, user addition and user modification;

[0017] The first screening module is used to screen the original remote sensing image data through an image screening model to obtain the data set SET1.

[0018] Optionally, the data processing unit includes a second user management module and a second screening module;

[0019] The second user management module is used to perform user management on the number of the communication-connected data correction units; wherein the user management includes user deletion, user addition and user modification;

[0020] The second screening module is used to screen the data set SET1 through an image screening model to obtain a data set SET10.

[0021] Optionally, the data processing unit includes a first human-computer interaction module, a graphic interpretation module and a data distribution module;

[0022] The image-text interpretation module is used to perform image-text interpretation on the data set SET10 through an image-text interpretation model to obtain the data set SET11;

[0023] The first human-computer interaction module is used to receive a first processing instruction, and perform corresponding processing on the data set SET11 according to the first instruction to obtain a data set SET2; wherein the first processing instruction includes image deletion and interpretation quality annotation;

[0024] The data distribution module is used to distribute the data set SET2 to the data correction unit.

[0025] Optionally, the image-text interpretation module is further used to obtain the target data set and optimize the image-text interpretation model through the target data set.

[0026] Optionally, the data correction unit includes: a correction module and a second human-computer interaction module;

[0027] The correction module is used to perform preliminary correction on the data set SET2 to obtain a data set SET20;

[0028] The second human-computer interaction module is used to receive a second processing instruction, and perform corresponding processing on the data set SET20 according to the second processing instruction to obtain the target data set; wherein the second processing instruction includes image deletion, interpretation result modification, historical interpretation result tracing and interpretation quality annotation.

[0029] Optionally, the system further comprises a data correction subunit, each of the data correction units being communicatively connected with a plurality of the data correction subunits;

[0030] The data correction unit is further used to distribute the data set SET20 to the data correction subunit;

[0031] The data correction subunit is used to correct the data set SET20 and transmit the corrected data set SET20 to the second human-computer interaction module, so that the second human-computer interaction module executes the second processing instruction on the corrected data set SET20 to obtain the target data set.

[0032] Another aspect of the present application provides a remote sensing image processing method, which is applied to the remote sensing image processing system, and the method comprises:

[0033] Acquire a data set SET1; wherein the data set SET1 is a data set to be processed acquired by the data management platform from the collected original remote sensing image data;

[0034] Performing graphic and text interpretation on the data set SET1 to obtain a data set SET2;

[0035] The data set SET2 is distributed to a data correction unit so that the data correction unit corrects the graphic and text interpretation result corresponding to the data set SET2 to obtain a target data set.

[0036] Another aspect of the present application provides a remote sensing image processing device, which is applied to the remote sensing image processing system, and the device includes:

[0037] An acquisition module is used to acquire a data set SET1; wherein the data set SET1 is a data set to be processed acquired by the data management platform from the collected original remote sensing image data;

[0038] An interpretation module, used for performing graphic interpretation on the data set SET1 to obtain a data set SET2;

[0039] The distribution module is used to distribute the data set SET2 to the data correction unit, so that the data correction unit corrects the graphic and text interpretation result corresponding to the data set SET2 to obtain the target data set.

[0040] The present application provides a remote sensing image processing system, method and device, which produce the following beneficial effects: after screening the original image data set through a data management platform, it is distributed to a data processing unit for automatic graphic interpretation, and then distributed to a data correction unit for correction, thereby realizing closed-loop management of online interpretation and correction of remote sensing image data sets, screening, interpretation and correction are performed step by step, which improves the interpretation quality and greatly improves the processing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A schematic diagram of the structure of a remote sensing image processing system provided in an embodiment of the present application;

[0042] Figure 2 A schematic diagram of the structure of a data management platform provided in an embodiment of the present application;

[0043] Figure 3 A structural diagram of a data processing unit provided in an embodiment of the present application;

[0044] Figure 4 A schematic diagram of a correction interface of a data correction unit provided in an embodiment of the present application;

[0045] Figure 5 A schematic diagram of a remote sensing image processing process provided in this application;

[0046] Figure 6 A schematic diagram of a process flow of a remote sensing image processing method provided in an embodiment of the present application;

[0047] Figure 7 The present invention provides a structural intention of a remote sensing image processing device provided in an embodiment of the present application.

[0048] The accompanying drawings are marked as follows: 1 is a data management platform, 2 is a data processing unit, 3 is a data correction unit, 10 is an analysis module, 11 is a statistics module, 12 is a data storage module, 13 is a display module, 14 is a first user management module, 15 is a first screening module, 20 is a second user management module, 21 is a second screening module, 22 is a first human-computer interaction module, 23 is a graphic interpretation module, and 24 is a data distribution module. DETAILED DESCRIPTION

[0049] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0050] It should be understood that although the terms first, second, third, etc. may be used in the present application to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0051] Figure 1 A schematic diagram of the structure of a remote sensing image processing system provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, the system includes: a data management platform 1, a data processing unit 2 and a data correction unit 3, wherein the data processing unit 2 includes a plurality of units.

[0052] like Figure 1 As shown, the data processing units 2 include n numbers, and each data processing unit 2 is connected to the data management platform 1 in communication, and each data processing unit 2 is connected to multiple data correction units 3. It should be noted that the present application does not limit the number of data processing units 2, and the number of data correction units 3 connected to each data processing unit 2 is also not limited. Of course, the number of data correction units 3 connected to different data processing units 2 can be the same or different, and can be set according to actual business needs.

[0053] In a specific embodiment, the data management platform 1 provided in the present application is used to collect original remote sensing image data. It can be understood that the remote sensing image refers to the image of the earth's surface or other celestial body surface obtained by remote sensing technology. The data management platform 1 can obtain remote sensing image data from satellites, airplanes, other aircraft and ground stations, etc., and of course it can also be data directly uploaded by managers, which is not limited by the present application.

[0054] After the data management platform 1 collects the original remote sensing image data, in order to avoid wasting resources, in an optional embodiment, the original remote sensing image data can be screened at regular intervals, for example, every three days, so as to eliminate low-quality remote sensing image data such as low image clarity, and obtain the data set SET1 to be processed.

[0055] In another optional embodiment, whether to perform screening can also be determined based on the amount of original remote sensing image data currently collected. When the collection amount reaches a preset amount, the data management platform 1 screens the collected original remote sensing image data to obtain the data set SET1. This application does not limit this.

[0056] Of course, in an optional embodiment, in order to improve processing efficiency, if the original remote sensing image data obtained is already high-quality image and text data, there is no need to filter it, and the original remote sensing image data can be directly distributed to the data processing unit 2 as a data set SET1.

[0057] Further, after the data management platform 1 screens and obtains the data set SET1, it distributes the data set SET1 to the data processing unit 2. When distributing, in an optional embodiment, the data management platform 1 can distribute evenly according to the number of data processing units 2 currently connected in communication, or can automatically distribute according to the remaining processing capacity of the data processing unit 2, so as to avoid a certain data processing unit 2 from reducing the processing efficiency due to the sum of the newly added task volume and the remaining processing capacity being too large.

[0058] Of course, in another optional embodiment, the administrator of the data management platform 1 may also distribute the data according to the data processing efficiency, data processing strengths, data processing quality, etc. of each data processing unit 2, and this application does not limit this.

[0059] In a specific embodiment, after receiving the data set SET1 sent by the data management platform 1, the data processing unit 2 performs graphic interpretation on the data set SET1 to obtain the data set SET2. In an optional embodiment, automatic interpretation can be performed using a graphic interpretation model.

[0060] Furthermore, the data processing unit 2 distributes the data set SET2 including the image-text interpretation result to the data correction unit 3, so that the data correction unit 3 corrects the image-text interpretation result to obtain an accurate target data set.

[0061] It should be noted that when the data processing unit 2 receives the data set SET1 issued by the data management platform 1, the system automatically prompts the team leader of the data processing unit 2 that there is currently an image data set that needs to be processed. Similarly, after the data processing unit 2 distributes the data set SET2 to the data correction unit 3, the correctors of each data correction unit 3 will receive a prompt signal, that is, there is currently an image data set that needs to be corrected.

[0062] Similarly, when the data processing unit 2 distributes the data set SET2, in an optional embodiment, the data processing unit 2 may distribute it evenly according to the number of data correction units 3 currently connected for communication, or may automatically distribute it according to the remaining processing capacity of the data correction unit 3, so as to avoid affecting the correction efficiency of a certain data correction unit 3 due to the sum of the newly added correction amount and the remaining correction amount being too large.

[0063] Of course, in another optional embodiment, the team leader of the data processing unit 2 may also distribute the corrections according to the correction efficiency, correction strengths, correction quality, etc. of each data correction unit 3, and this application does not limit this.

[0064] Furthermore, after the data correction unit 3 receives the data set SET2, in an optional embodiment, the data set SET2 may be initially corrected by using a correction model, and then manually corrected by a corrector.

[0065] During manual correction, the image and text interpretation results corresponding to the data set SET2 can be deleted, modified, and quality-labeled. Specifically, if the corrector believes that the current image quality is low or the image and text interpretation result is poor, the image can be deleted. If it is believed that there are some errors in the interpretation result, the interpretation result can be edited and modified. After completing the above correction, the image can be quality-labeled. For example, the quality labeling includes high quality, medium quality, and low quality. This application does not limit the quality labeling method and classification.

[0066] Further, the data correction unit 3 corrects the data set SET2 to obtain a target data set, and the data correction unit 3 uploads the target data set to the data processing unit 2 so that the team leader of the data processing unit 2 can review the correction result.

[0067] If the reviewed images are found to have errors or are of low quality, they can be sent back to the corresponding data correction unit 3 for repeated correction. Of course, the team leader of the data processing unit 2 can also directly delete, modify, etc. the images with poor correction results.

[0068] After the review is completed, the data processing unit 2 uploads the obtained target data set to the data management platform 1 so as to summarize, analyze and store all target data sets. Thus, high-quality remote sensing image data can be applied to different fields such as environmental monitoring, urban planning, scientific research, etc.

[0069] In an optional embodiment, the administrator corresponding to the data management platform 1 can view all data sets, while the staff corresponding to the data processing unit 2 and the data correction unit 3 can only view the data assigned to them, thereby achieving task isolation.

[0070] Therefore, the remote sensing image processing system provided in the embodiment of the present application, after screening the original image data set through the data management platform 1, distributes it to the data processing unit 2 for automatic graphic interpretation, and then distributes it to the data correction unit 3 for correction, thereby realizing closed-loop management of online interpretation and correction of remote sensing image data sets, screening, interpretation and correction are performed step by step, which improves the interpretation quality while greatly improving the processing efficiency.

[0071] Figure 2 A schematic diagram of the structure of a data management platform provided in an embodiment of the present application. In an optional embodiment, as Figure 2 As shown, the data management platform 1 includes an analysis module 10 , a statistics module 11 , a data storage module 12 and a display module 13 .

[0072] In a specific embodiment, after the data correction unit 3 corrects the data set SET2 to obtain the target data set, the target data set is transmitted to the data management platform 1. At this time, the analysis module 10 in the data management platform 1 obtains the target data set and performs image analysis on the target data set to obtain the analysis result. In an optional embodiment, the image analysis includes image classification, image area division and target detection.

[0073] Among them, image classification refers to the classification and division of image types. The classification results can include farmland, cities and villages, etc. This application does not specifically limit the classification standards, number of classifications and types of classifications. Image area division refers to the identification of specified areas in an image and marking them with box diagrams. For example, the identification and circle of a city area in an image. It should be noted that the specified area can be set, modified and deleted according to actual business needs, and this application does not limit this.

[0074] Target detection refers to identifying a specified target in an image and marking it with a box diagram. For example, a ship in an image is identified and circled. Similarly, the target setting can be set, modified, and deleted according to the recognition business needs, and this application does not limit this.

[0075] like Figure 2 As shown, in an optional embodiment, the data management platform 1 also includes a statistical module 11, which is used to perform statistics on data processing information of the data processing unit 2 and the data correction unit 3, wherein the data processing information includes data processing volume, data processing results, data processing quality and data processing progress.

[0076] It is understandable that in order to facilitate the administrator of the data management platform 1 to monitor the current processing status of each data processing unit 2 and data correction unit 3 in real time and facilitate the subsequent distribution of image data. In a specific embodiment, the data processing unit 2 and the data correction unit 3 upload their own data processing information in real time, so that the statistical module 11 in the data management platform 1 can collect statistics on the data processing information of all data processing units 2 and data correction units 3.

[0077] Furthermore, the data processing information is transmitted to the data storage module 12 so that the administrator can call the historical processing information, determine the processing efficiency of different data processing units 2 and other information based on the historical processing information, and adjust the data allocation in time, thereby improving the image data processing efficiency.

[0078] In an optional embodiment, the data storage module 12 can store not only data processing information but also original remote sensing image data, target data set and analysis results. In a specific embodiment, after the data storage module 12 obtains the corrected high-quality target data set, it analyzes and processes the image data through the analysis module 10, and then archives, packages and stores the image data.

[0079] Of course, the statistics module 11 will also transmit the data processing information to the display module 13, so that the display module 13 can display the data processing status of each data processing unit 2 and the data correction unit 3, so as to facilitate the administrator to view and schedule tasks in time.

[0080] In an optional embodiment, in addition to displaying data processing information, the display module 13 can also display the task allocation, weekly processing volume, processing rankings, and correction quality rankings of the data processing unit 2 and the data correction unit 3. It should be noted that the display method can be a chart or other method, which is not limited in this application.

[0081] like Figure 2 As shown, based on the above embodiment, as an optional embodiment, the data management platform 1 further includes a first user management module 14 and a first screening module 15 .

[0082] It can be understood that the data management platform 1 is connected to multiple data processing units 2. The data management platform 1 can be understood as a platform that can be controlled by an administrator, and the data processing unit 2 is the team leader of image data processing. Different team leaders manage multiple image data correctors (i.e., data correction units 3).

[0083] In a specific embodiment, if the data processing business pressure is high and a new team leader needs to be added, or if the business pressure is small and a team leader needs to be reduced, or the team leader's information needs to be modified, the data management platform 1 needs to perform user management on the connected data processing unit 2.

[0084] Specifically, it is implemented through the first user management module 14 in the data management platform 1. It can be understood that the first user management module 14 is mainly used to manage the user information corresponding to the data processing unit that is communicated with the data management platform 1. Specifically, user management includes user deletion, user addition and user modification.

[0085] In an optional embodiment, the data management platform 1 also includes a first screening module 15, which is used to screen the original remote sensing image data through an image screening model, thereby identifying images whose clarity does not meet the requirements, that is, images with low clarity, thereby obtaining a data set SET1 to be interpreted.

[0086] It should be noted that the present application does not specifically limit the image screening model. In an optional embodiment, after the data correction unit 3 corrects the target data set, the target data set is uploaded to the data processing unit 2, and finally transmitted to the data management platform 1. The first screening module 15 uses the corrected high-quality target data set as a training sample to train the image screening model, thereby optimizing the image screening model, so as to quickly and accurately screen out a high-quality data set SET1 in the future, thereby improving the interpretation quality of the remote sensing image data.

[0087] Figure 3 A structural diagram of a data processing unit provided in an embodiment of the present application is provided as an optional embodiment. Figure 3 As shown, the data processing unit 2 includes a second user management module 20 and a second screening module 21 .

[0088] like Figure 1As shown, it can be understood that the data processing unit 2 is connected to multiple data correction units 3. The data processing unit 2 can be understood as a team leader, and different data correction units 3 can be understood as independent correctors. If the data processing business pressure is high and new correctors are needed, or the business pressure is low and the number of business personnel needs to be reduced, or the corrector information needs to be modified, etc., the data processing unit 2 is required to perform user management on the connected data correction units 3.

[0089] Specifically, it is implemented through the second user management module 20 in the data processing unit 2. It can be understood that the second user management module 20 is mainly used to manage the user information corresponding to the data correction platform that is communicatively connected to the data processing unit 2. Specifically, user management includes user deletion, user addition and user modification.

[0090] In an optional embodiment, if Figure 3 As shown, the data processing unit 2 also includes a second screening module 21. It can be understood that in order to improve the accuracy of the image text interpretation, before the data processing unit 2 interprets the data set SET1, the data set SET1 is screened again by the second screening module 21 so as to extract low-quality image pictures and obtain the data set SET10.

[0091] Specifically, the second screening module 21 also screens the data set SET1 through the image screening model. It should be noted that the image screening models in the first screening module 15 and the second screening module 21 can be the same model or different models. However, in an optional embodiment, the image screening model needs to be optimized through the target data set to improve the image interpretation quality.

[0092] Based on the above embodiments, Figure 3 As shown, as an optional embodiment, the data processing unit 2 may further include a first human-computer interaction module 22 , a graphic interpretation module 23 and a data distribution module 24 .

[0093] In a specific embodiment, the image-text interpretation module 23 interprets the data set SET10 through the image-text interpretation model to obtain the data set SET11. At this time, the team leader corresponding to the data processing unit 2 can perform image deletion, image modification, and interpretation quality marking on the interpreted data processing unit 2.

[0094] Specifically, the team leader types the first processing instruction through a device such as a mouse and a keyboard. After receiving the first processing instruction, the first human-computer interaction module 22 performs corresponding processing on the data set SET11 according to the first instruction to obtain the data set SET2. It can be understood that in order to improve data processing efficiency, the team leader may not modify the interpretation result, and distribute the data set SET2 to the data correction unit 3 for correction one by one. Therefore, the first processing instruction includes image deletion and interpretation quality annotation. Of course, the team leader also has the authority to modify the data set SET2. Therefore, in an optional embodiment, the first instruction may also include but is not limited to image modification.

[0095] Further, the first human-computer interaction module 22 transmits the obtained data set SET2 to the data distribution module 24, so that the data distribution module 24 distributes the data set SET2 to the data correction unit 3 for correction.

[0096] It should be noted that when the quality of the interpretation of the graphic interpretation model is high and the team leader does not enter any first processing instructions, the graphic interpretation module 23 directly transmits the interpreted data set SET11 to the data distribution module 24, so that the data distribution module 24 distributes the data set SET11 as the data set SET2 to the data correction unit 3.

[0097] In an optional embodiment, the data processing unit 2 may also include a display module for displaying the data processing information of the data correction unit 3 to which it is connected in communication. Similarly, the data processing information may include the data processing volume, data processing results, data processing quality and data processing progress, and may also include the task allocation volume of the data correction unit 3, the weekly processing volume, the processing ranking list, the correction quality ranking list, etc. The present application does not specifically limit the specific display content and display method.

[0098] In another optional embodiment, in order to improve the accuracy of image-text interpretation, the image-text interpretation module 23 is also used to obtain a target data set and optimize the image-text interpretation model through the target data set. Specifically, after the data correction unit 3 corrects the data set SET2 to obtain a high-quality target data set, the target data set is transmitted to the data processing unit 2, so that the data processing unit 2 optimizes the image screening model and the image-text interpretation model through the target data set.

[0099] As an optional embodiment, the data correction unit 3 includes: a correction module and a second human-computer interaction module. In a specific embodiment, after receiving the data set SET2 distributed by the data processing unit 2, the data correction unit 3 performs preliminary correction on the data set SET2 through the correction module to obtain the data set SET20. Specifically, automatic correction can be performed through a correction model.

[0100] Furthermore, the corrector corresponding to the data correction unit 3 manually corrects the image text one by one to improve the accuracy of the image text interpretation. Specifically, the corrector enters the second processing instruction through a keyboard or a mouse light device, and after the second human-computer interaction module receives the second processing instruction, it performs corresponding processing on the data set SET20 according to the second processing instruction to obtain the target data set. Among them, the second processing instruction includes image deletion, interpretation result modification, historical interpretation result tracing and interpretation quality annotation.

[0101] Figure 4 A correction interface diagram of a data correction unit provided in an embodiment of the present application. In an optional embodiment, during manual correction, the correction staff opens the image pictures that need to be corrected by graphic interpretation one by one. After opening, Figure 4 As shown, the correction interface includes the currently opened image and the interpretation file displayed on the right.

[0102] In an optional embodiment, if Figure 4 As shown, the interface also includes data management, which is mainly used to store all image data sets that need to be corrected by the current data correction unit 3. Notifications are mainly used to notify the corrector whether there is a new correction task. When the data processing unit 2 issues a new correction task, the notification pops up the task message. User information is used to store relevant information of the corrector, such as user name, password, etc.

[0103] When the image is opened, the corresponding interpretation file (e.g., XML file) is displayed on the right side of the image. This makes it convenient for the proofreader to check whether the content in the interpretation file is correct based on the image content. It should be noted that the interpretation file obtained by the graphic interpretation model can be in XML format or other formats, which is not limited in this application.

[0104] In an optional embodiment, if the corrector believes that the image quality is not high or the interpretation result is poor, such as Figure 4 As shown, the corrector can click the "Discard" option to delete it. If the corrector believes that the interpretation result corresponding to the image is partially incorrect or has typos, the corrector can directly edit and modify the interpretation file. After completion, click the "Save" option to obtain the target dataset.

[0105] Of course, during the calibration process, Figure 4 As shown, the corrector can also mark the quality of the image according to the interpretation quality and image clarity of the current image. If you are satisfied with the image interpretation result, or the image quality and clarity are relatively high, you can click the "Quality Marking" button to select the current quality and save it.

[0106] Among them, the quality labeling can be a classification of high quality, medium quality and low quality, or it can be a classification of first-level quality, second-level quality and third-level quality, and the first-level quality is higher than the second-level quality, which is higher than the third-level quality. This application does not limit the specific form of quality labeling.

[0107] In an optional embodiment, if the corrector believes that the current interpretation quality is poor, he can click the "View Log" option to trace the historical interpretation versions of the current image in order to select a higher quality version. If there is no satisfactory version among all versions, you can choose to click the "Discard" option to delete it.

[0108] It can be understood that the technical solution provided by this application can realize historical version backtracking by clicking "View Log", that is, the historical process of image processing can be traced to facilitate the location of the responsible person, thereby making the online processing of remote sensing image data visible, manageable and controllable.

[0109] In an optional embodiment, in order to further improve the processing efficiency of image data, as an optional embodiment, the remote sensing image processing system provided in the present application may also include a data correction subunit, and each data correction unit 3 is communicatively connected to multiple data correction subunits.

[0110] It can be understood that the data correction unit 3 can also be understood as a team leader, and manages multiple team members (i.e., data correction sub-units). In a specific embodiment, the data correction unit 3 distributes the data set SET20 to the data correction sub-units, and the team members corresponding to the data correction sub-units correct the data set SET20. For the specific correction process, refer to the correction process of the data correction unit 3 in the above embodiment.

[0111] Further, the data correction subunit transmits the corrected data set SET20 to the second human-computer interaction module in the data correction unit 3, so that the second human-computer interaction module executes the second processing instruction on the corrected data set SET20 to obtain the target data set.

[0112] It is worth noting that in order to improve the correction accuracy, in an optional embodiment, the team leader corresponding to the data correction unit 3 can be only responsible for distributing the data set SET20 to the data correction sub-unit for correction. After the data correction sub-unit completes the correction, the data correction unit 3 verifies the correction results one by one, thereby improving the correction accuracy.

[0113] Of course, it should be noted that the present application does not limit the number of data processing units 2, the number of data correction units 3, and the number of data correction units 3 that are connected to the data processing unit 2. In addition, the present application does not limit the level of image data set processing. It can be three levels including data management platform 1, data processing unit 2 and data correction unit 3, or it can be four levels including data management platform 1, data processing unit 2, data correction unit 3 and data correction subunit. The present application does not limit the number of processing levels.

[0114] Figure 5 A flowchart of a remote sensing image processing process provided by the present application is provided. In order to facilitate those skilled in the art to better understand the technical solution of the present application, the following will be combined with Figure 5 Further detailed description is given.

[0115] like Figure 5 As shown, in an optional embodiment, the data management platform 1 preliminarily screens the original remote sensing image data through the image screening model to obtain the data set SET1 to be processed. Further, the data set SET1 is distributed to the data processing unit 2 connected to the communication.

[0116] Take one of the data processing units 2 as an example for explanation. Figure 5 As shown, the data processing unit 2 first deletes low-quality images through an image screening model and / or manual screening to obtain a data set SET10. Further, the data set SET10 is interpreted through a text-image interpretation model to obtain a data set SET11.

[0117] After the data processing unit 2 obtains the data set SET11, it distributes the data set SET11 to the data correction unit 3 connected to the communication, and the data correction unit 3 performs correction to obtain the target data set. The specific correction includes image deletion, interpretation result modification, historical interpretation result tracing and interpretation quality annotation, etc.

[0118] Therefore, through the remote sensing image processing system provided in this application, a series of processing such as image storage, image screening, manual screening, image graphic interpretation, graphic interpretation correction, and correction content review can be completed in an online closed loop, thereby improving the interpretation quality while greatly improving the processing efficiency.

[0119] In the above embodiments, the remote sensing image processing system is described in detail. The present application also provides an embodiment corresponding to a remote sensing image processing method, which is applied to the remote sensing image processing system described in any of the above embodiments.

[0120] Figure 6 A flowchart of a remote sensing image processing method provided in an embodiment of the present application is shown in FIG. Figure 6As shown, the method includes:

[0121] S60: Acquire data set SET1; wherein data set SET1 is a data set to be processed acquired by the data management platform from the collected original remote sensing image data;

[0122] S61: interpreting the data set SET1 in pictures and texts to obtain the data set SET2;

[0123] S62: Distribute the data set SET2 to the data correction unit, so that the data correction unit corrects the image-text interpretation result corresponding to the data set SET2 to obtain the target data set.

[0124] Since the remote sensing image processing method provided in the present application corresponds to the remote sensing image processing system provided in the above embodiment, the specific image data processing process refers to the description of the above embodiment.

[0125] It should be noted that the remote sensing image processing method provided in the embodiment of the present application is applied to the remote sensing image processing system provided in any of the above embodiments. Therefore, the method is applied to a system including a data management platform, a data processing unit, and a data correction unit. In a specific embodiment, the execution subject of the method provided in the present application can be any one of the data management platform, the data processing unit, and the data correction unit. In the embodiment of the present application, the data processing unit is taken as the execution subject for explanation.

[0126] In the above embodiments, the remote sensing image processing method is described in detail. The present application also provides an embodiment corresponding to a remote sensing image processing device, which is applied to the remote sensing image processing system described in any of the above embodiments.

[0127] Figure 7 The structural intention of a remote sensing image processing device provided in an embodiment of the present application is as follows: Figure 7 As shown, the device comprises:

[0128] The acquisition module 70 is used to acquire a data set SET1; wherein the data set SET1 is a data set to be processed acquired by the data management platform from the collected original remote sensing image data;

[0129] The interpretation module 71 is used to interpret the data set SET1 in pictures and texts to obtain the data set SET2;

[0130] The distribution module 72 is used to distribute the data set SET2 to the data correction unit, so that the data correction unit corrects the graphic text interpretation result corresponding to the data set SET2 to obtain the target data set.

[0131] Since the remote sensing image processing device provided in the present application corresponds to the remote sensing image processing method and remote sensing image processing system provided in the above embodiments, similarly, the specific description refers to the above embodiments.

[0132] It should be noted that, although the operations are depicted in a specific order in the accompanying drawings, this should not be understood as requiring these operations to be performed in the specific order shown or to be performed sequentially, or requiring all illustrated operations to be performed to achieve the desired results. In some cases, multitasking and parallel processing may be advantageous. In addition, the separation of various system modules and components in the above-described embodiments should not be understood as requiring such separation in all embodiments, and it should be understood that the described program components and systems can generally be integrated together in a single software product, or packaged into multiple software products.

Claims

1. A remote sensing image processing system, characterized in that: The system comprises: a data management platform, a data processing unit and a data correction unit; wherein the data processing unit comprises a plurality of; The data processing unit is communicatively connected to the data management platform; and each of the data processing units is communicatively connected to a plurality of the data correction units; The data management platform is used to collect raw remote sensing image data and distribute the data set SET1 to be processed in the raw remote sensing image data to the data processing unit; The data processing unit is used to perform graphic interpretation on the data set SET1 to obtain a data set SET2; and distribute the data set SET2 to the data correction unit; The data correction unit is used to correct the image-text interpretation result corresponding to the data set SET2 to obtain a target data set.

2. The remote sensing image processing system as claimed in claim 1, characterized in that: The data management platform includes an analysis module, a statistics module, a data storage module and a display module; The analysis module is used to obtain the target data set and perform image analysis on the target data set to obtain analysis results; wherein the image analysis includes image classification, image area division and target detection; The statistical module is used to collect statistics on data processing information of the data processing unit and the data correction unit; wherein the data processing information includes data processing volume, data processing results, data processing quality and data processing progress; The data storage module is used to store the original remote sensing image data, the target data set, the analysis results and the data processing information; The display module is used to display the data processing information.

3. The remote sensing image processing system as claimed in claim 2, characterized in that: The data management platform also includes a first user management module and a first screening module; The first user management module is used to perform user management on the number of the communication-connected data processing units; wherein the user management includes user deletion, user addition and user modification; The first screening module is used to screen the original remote sensing image data through an image screening model to obtain the data set SET1.

4. The remote sensing image processing system as claimed in claim 1, characterized in that: The data processing unit includes a second user management module and a second screening module; The second user management module is used to perform user management on the number of the communication-connected data correction units; wherein the user management includes user deletion, user addition and user modification; The second screening module is used to screen the data set SET1 through an image screening model to obtain a data set SET10.

5. The remote sensing image processing system as claimed in claim 4, characterized in that: The data processing unit includes a first human-computer interaction module, a graphic interpretation module and a data distribution module; The image-text interpretation module is used to perform image-text interpretation on the data set SET10 through an image-text interpretation model to obtain the data set SET11; The first human-computer interaction module is used to receive a first processing instruction, and perform corresponding processing on the data set SET11 according to the first instruction to obtain a data set SET2; wherein the first processing instruction includes image deletion and interpretation quality annotation; The data distribution module is used to distribute the data set SET2 to the data correction unit.

6. The remote sensing image processing system as claimed in claim 5, characterized in that: The image-text interpretation module is further used to obtain the target data set and optimize the image-text interpretation model through the target data set.

7. The remote sensing image processing system as claimed in claim 1, characterized in that: The data correction unit comprises: a correction module and a second human-computer interaction module; The correction module is used to perform preliminary correction on the data set SET2 to obtain a data set SET20; The second human-computer interaction module is used to receive a second processing instruction, and perform corresponding processing on the data set SET20 according to the second processing instruction to obtain the target data set; wherein the second processing instruction includes image deletion, interpretation result modification, historical interpretation result tracing and interpretation quality annotation.

8. The remote sensing image processing system as claimed in claim 7, characterized in that: The system further comprises a data correction subunit, each of the data correction units being communicatively connected with a plurality of the data correction subunits; The data correction unit is further used to distribute the data set SET20 to the data correction subunit; The data correction subunit is used to correct the data set SET20 and transmit the corrected data set SET20 to the second human-computer interaction module, so that the second human-computer interaction module executes the second processing instruction on the corrected data set SET20 to obtain the target data set.

9. A remote sensing image processing method, characterized in that: A remote sensing image processing system applied to any one of claims 1 to 8, wherein the method comprises: Acquire a data set SET1; wherein the data set SET1 is a data set to be processed acquired by the data management platform from the collected original remote sensing image data; Performing graphic and text interpretation on the data set SET1 to obtain a data set SET2; The data set SET2 is distributed to a data correction unit so that the data correction unit corrects the graphic and text interpretation result corresponding to the data set SET2 to obtain a target data set.

10. A remote sensing image processing device, characterized in that: A remote sensing image processing system applied to any one of claims 1 to 8, the device comprising: An acquisition module is used to acquire a data set SET1; wherein the data set SET1 is a data set to be processed acquired by the data management platform from the collected original remote sensing image data; An interpretation module, used for performing graphic interpretation on the data set SET1 to obtain a data set SET2; The distribution module is used to distribute the data set SET2 to the data correction unit, so that the data correction unit corrects the graphic and text interpretation result corresponding to the data set SET2 to obtain the target data set.

Images