CN113220920A - Satellite remote sensing image sample labeling system and method based on micro-service architecture - Google Patents
Satellite remote sensing image sample labeling system and method based on micro-service architecture Download PDFInfo
- Publication number
- CN113220920A CN113220920A CN202110609043.XA CN202110609043A CN113220920A CN 113220920 A CN113220920 A CN 113220920A CN 202110609043 A CN202110609043 A CN 202110609043A CN 113220920 A CN113220920 A CN 113220920A
- Authority
- CN
- China
- Prior art keywords
- image
- labeling
- remote sensing
- marking
- subsystem
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/5866—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, manually generated location and time information
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
- G06F16/5838—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using colour
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Library & Information Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a satellite remote sensing image sample labeling system and method based on a micro-service architecture, belonging to the technical field of satellite remote sensing influence labeling; the system comprises a sample marking subsystem, a data management subsystem, an information statistics subsystem, an audit management subsystem and an intelligent algorithm integration subsystem, wherein each subsystem is designed and realized by adopting a micro-service idea, is not influenced mutually and can be deployed and operated independently; on the basis, an intelligent labeling algorithm is fully utilized, and an intelligent labeling method for the satellite remote sensing image sample is provided.
Description
Technical Field
The invention relates to the technical field of satellite remote sensing image labeling, in particular to a satellite remote sensing image sample labeling system and method based on a micro-service architecture.
Background
With the continuous improvement of the satellite observation capability in China, the quantity of satellite remote sensing data which can be acquired is multiplied, the power of interpreters is relatively unchanged, and the quantity of limited interpreters and labels limits the full play of data benefits. Meanwhile, with the rapid development of artificial intelligence technology, the intelligent remote sensing image detection, identification and classification method has become an important information extraction means in the field of satellite remote sensing image application, and the method mainly benefits from the construction of various large-scale finely labeled reference data sets, such as ImageNet, COCO and the like. As the fundamental work of the deep learning technology, the result of sample labeling directly determines the effect of the application of the deep learning technology.
However, most of the currently mainstream sample annotation software is client software, which is backward in architecture, generally cannot adapt to remote sensing image data and can only support manual completion, such as labelImg, labelme, and the like, and not only is the annotation efficiency low, but also multi-user collaborative annotation cannot be realized, and the satellite remote sensing image cannot be effectively supported. Secondly, the current labeling software generally simply labels image data, only pays attention to category and position information, and along with continuous deepening of an artificial intelligence technology, the sample data cannot meet the rich attribute requirements of a novel intelligent algorithm on the sample data. Finally, after the labeling of the current labeling software is finished, the unified management capability of the sample data condition is lacked, the whole condition of the sample data is difficult to master, and more errors or sample data with low quality are easy to occur.
In summary, the current sample labeling system has the problems of low sample labeling efficiency, low quality, small number of effective samples, insufficient feature diversity and the like, and particularly, the number of the data sets for a target in a specific area or a specific category is not sufficient, and the content diversity of the available data sets is not sufficient, so that the capability of effectively mining the remote sensing image by the existing advanced intelligent algorithm is limited to a great extent.
Disclosure of Invention
In view of the above, the present invention provides a system and a method for labeling a satellite remote sensing image sample based on a micro service architecture. The system supports containerized deployment, improves the operation efficiency of the intelligent algorithm, and realizes multi-user online collaborative labeling.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
a satellite remote sensing image sample labeling system based on a micro-service architecture is realized by a sample labeling subsystem, a data management subsystem, an information statistics subsystem and an intelligent algorithm integration subsystem;
the sample marking subsystem is used for loading and displaying the remote sensing image and the marking result and providing the functions of automatic marking and manual marking of the remote sensing image;
the data management subsystem is used for cleaning and organizing the remote sensing image data and the labeling result data;
the information statistics subsystem is used for displaying the statistics information of the remote sensing image data and the labeling result;
the intelligent algorithm integration subsystem integrates various intelligent marking algorithms and calls the algorithms in a web service mode, and the intelligent marking algorithms comprise target-level intelligent marking algorithms and pixel-level intelligent marking algorithms;
furthermore, the sample marking subsystem comprises an image loading display module, an automatic marking module, a manual marking module and an information display and image basic operation module;
the image loading display module is used for realizing loading display of the remote sensing image and integrates a remote sensing image basic algorithm of image stretching and image resampling in a function calling mode;
the automatic labeling module is used for realizing the functions of automatic labeling of the whole graph and automatic labeling of the area; the intelligent marking algorithm in the intelligent algorithm integration subsystem is called in a service calling mode through a uniform service calling interface to realize automatic marking of the whole image or the designated area of the current remote sensing image;
the manual marking module is used for marking the target in the remote sensing image in a man-machine interaction mode;
the information display module is used for displaying image list information in a current database, currently loaded remote sensing image element information, real-time marking result information, an image list information display sequence number, an image name and a current marker;
the image basic operation module is used for completing adjustment of contrast, saturation and brightness of the remote sensing image and providing a palette function for adjusting colors of all the marking frames in the current image.
Furthermore, the data management subsystem comprises a data cleaning and warehousing module, a data retrieval module, an image catalog setting module and a labeling result management module;
the data cleaning and warehousing module is used for scanning a specified folder path at regular time, automatically completing the analysis and warehousing of the remote sensing image data when new remote sensing image data is obtained, cleaning the meta information of the remote sensing image data, and removing repeated fields and error value data in the meta information;
the data retrieval module is used for providing retrieval functions for image data and sample labeling data in a database;
the image directory setting module is used for setting image path information and adjusting the scanned specified folder path;
the marking result management module is used for providing functions of submitting, checking, importing and exporting the marking result and supporting batch operation.
A satellite remote sensing image sample labeling method based on a micro-service architecture is disclosed, and the system is realized by the following steps:
step one, the data management subsystem scans the set catalog at regular time, automatically analyzes the scanned image and cleans and stores the meta information;
secondly, the sample labeling subsystem acquires information of an image to be labeled from a database, loads the image to be labeled and selects a waveband of the multiband image;
step three, calling a background intelligent labeling algorithm to automatically label the whole image or the local area of the image;
step four, the intelligent algorithm integration subsystem completes target-level automatic labeling according to the request, completes pixel-level automatic labeling on the basis of the target-level automatic labeling, and then returns the result to the sample labeling subsystem;
step five, the sample labeling subsystem adjusts the contrast, saturation, brightness and color of the target labeling frame of the image according to the image condition;
step six, manually utilizing a marking tool to correct the automatic marking result, editing relevant attribute information, and storing the marking result in a warehouse;
and seventhly, counting and displaying the marking conditions of the images and the samples by the marking information stored in the database in the information counting subsystem.
The invention adopts the technical scheme to produce the beneficial effects that:
1. the intelligent labeling system is based on the micro-service architecture idea, adopts B/S architecture design, is easy to develop and deploy, has no mutual influence on subsystems, supports containerized deployment when a certain micro-service has a problem and does not influence the operation of other functions, and originally supports a cloud computing environment, thereby effectively realizing multi-user online collaborative labeling;
2. according to the invention, an intelligent marking algorithm is fully utilized, the traditional manual marking mode is improved into an intelligent auxiliary marking mode, the marking efficiency is greatly improved, two modes of full-image marking and area marking are provided in the marking process by utilizing the intelligent algorithm, the problem of large data volume of a single remote sensing image is effectively solved, and the operating efficiency of the intelligent algorithm is greatly improved;
3. in the marking process, the attribute information of the remote sensing image data sample can be more comprehensively displayed through setting of multiple dimensional attributes such as small samples, difficult samples, incomplete samples and the like, and the follow-up learning training of an intelligent algorithm is better assisted.
4. The invention facilitates the understanding of the characteristics of the data and the samples from multiple dimensions by cleaning, organizing, managing and statistically analyzing the remote sensing images and the sample data, and provides more flexible training label data by supporting the import and export of various label formats.
Drawings
FIG. 1 is a block diagram of an embodiment of the present invention
Fig. 2 is a block diagram of a system according to an embodiment of the present invention.
Fig. 3 is a diagram of the interface relationship between subsystems according to the embodiment of the present invention.
FIG. 4 is a flowchart illustrating the labeling process according to the embodiment of the present invention.
Detailed Description
The present invention will be further described with reference to the accompanying drawings and specific embodiments.
The invention provides a satellite remote sensing image sample intelligent marking system and method based on a micro-service architecture, and aims at satellite remote sensing image data. The satellite remote sensing image sample intelligent labeling system based on the micro-service architecture is based on the micro-service idea, adopts B/S architecture design, designs integrated remote sensing image basic processing and intelligent labeling algorithm, constructs a man-machine combined intelligent sample labeling tool, and realizes intelligent auxiliary labeling, cleaning and organization management of remote sensing image data.
The following description of the embodiments and the basic principles of the present invention are further described with reference to the accompanying drawings.
Referring to fig. 1 to 4, the satellite remote sensing image sample intelligent labeling system based on the microservice architecture of the embodiment is implemented by adopting a B/S architecture design, one or more servers or workstations can be used as a server, after the server finishes deployment and starts service, any PC connected with the server is used as a client, access can be finished through a browser, remote sensing image labeling is performed, and simultaneous online collaborative labeling of multiple people is supported.
The architecture diagram of the satellite remote sensing image sample intelligent marking system based on the micro-service architecture is shown in fig. 1 and is divided into 5 layers, namely a basic resource layer, a database layer, a system service layer, an application supporting layer and an application layer from bottom to top.
(1) Base resource layer
The basic resource layer comprises computing resources, storage resources and network resources and provides basic hardware support for the system.
(2) Database layer
The database layer mainly comprises a basic support library, an image resource library and a marking result library, wherein the basic support library comprises user information and authority information, the image resource library completes maintenance and management of satellite remote sensing image data, the marking result library comprises automatic marking result data and marking result data after manual correction, and the automatic marking result and the marking result data after manual correction are respectively stored in two database tables.
(3) System service layer
The system service provides services such as user management, authority management, message management, log management and the like, and provides basic system service for the system.
(4) Using a supporting layer
The support layer comprises a front-end display frame, WEB services, a remote sensing image basic processing algorithm and an intelligent marking algorithm.
(5) Application layer
The application layer comprises applications such as automatic labeling, manual labeling, statistical analysis, data management, auditing management and the like.
A block diagram of a satellite remote sensing image sample intelligent labeling system based on a micro-service architecture is shown in fig. 2, and comprises 5 subsystems of sample labeling, data management, information statistics, audit management and intelligent algorithm integration, and interface relations among the subsystems are shown in fig. 3.
The work flow chart of the satellite remote sensing image sample intelligent labeling system based on the micro-service architecture is shown in fig. 4, and the specific steps are as follows:
the method comprises the following steps that firstly, a data management subsystem scans regularly according to a set catalog, automatically analyzes and cleans meta information and stores the meta information in a warehouse after an image is scanned;
secondly, the sample marking subsystem loads the image to be marked from the database, selects the wave band, and can perform roaming and zooming operations on the image under necessary conditions;
step three, after the image loading is finished, selecting a whole image or a local image, and calling a background automatic annotation algorithm to perform automatic annotation;
step four, the intelligent algorithm integration subsystem completes target-level automatic labeling according to the request, completes pixel-level automatic labeling on the basis of the target-level automatic labeling, and returns the result to the sample labeling subsystem;
step five, the sample labeling subsystem can adjust the contrast, saturation, brightness and color of the target labeling frame of the image according to the image condition;
step six, manually utilizing a marking tool to correct the automatic marking result, editing relevant attribute information, and storing the marking result in a warehouse;
step seven, a user can utilize the data management subsystem to search and inquire the data and submit the searched images and the labeled information to the auditing operation;
and step eight, in the auditing management subsystem, auditing the labeling result by the user with the auditing authority, if the auditing is passed, warehousing the labeling information, updating the labeling result, if the auditing is not passed, returning to the step six, and modifying the sample labeling subsystem according to the auditing suggestion.
And finally, counting and displaying the labeling conditions of the images and the samples in the information counting subsystem.
The following describes the components of the system:
(1) sample annotation subsystem
The sample labeling subsystem loads and displays the remote sensing image and the labeling result and provides automatic and manual labeling functions of the remote sensing image, and the functions of the modules comprise an image loading display module, an automatic labeling module, a manual labeling module, an information display module and an image basic operation module, and the functions of the modules are as follows:
the image loading display module integrates remote sensing image basic algorithms such as image stretching, image resampling and the like in a function calling mode, realizes loading display of the remote sensing image on the premise of not issuing service to the remote sensing image, realizes scaling and roaming of the remote sensing image through continuous interaction of a front platform and a back platform, and provides a wave band selection function for a user to select a corresponding wave band to display the remote sensing image with the wave band number more than 1 when the remote sensing image is displayed;
the automatic labeling module comprises a full-image and area automatic labeling function, a unified service calling interface is designed, a background intelligent labeling algorithm is called in a WEB service calling mode to realize automatic labeling of the full image or the designated area of the current remote sensing image, intelligent algorithm execution progress information can be displayed in the automatic labeling process, and target-level automatic labeling and pixel-level automatic labeling are supported; and in the calling process, information such as the ID, the image path, the labeling type and the like of the image is transmitted to the intelligent labeling algorithm through the web service interface, and the labeling result is returned after the automatic labeling is finished.
The manual labeling module is used for completing rectangular labeling and polygonal labeling of an interested target in the remote sensing image by selecting a rectangular frame or a polygonal frame, wherein the labeled attribute information comprises position information, target large type, target model, target state, whether a small sample is available, whether a difficult sample is available or not and whether an incomplete sample is available or not, and meanwhile, the editing of the existing labeling result is supported;
the information display module is used for displaying image list information, currently loaded remote sensing image element information and real-time marking result information in a current database, wherein the image list information displays a serial number, an image name and a current marker, the remote sensing image element information comprises a satellite code, a sensor type, a spatial resolution, an image size, a central store longitude and latitude, image content and imaging time, the image content can be edited, and the real-time marking result information comprises a serial number, a multi-level label name, a marking person and detail information;
and the image basic operation module is used for completing the adjustment of the contrast, the saturation and the brightness of the remote sensing image and providing a palette function for adjusting the color of each marking frame in the current image.
(2) Data management subsystem
The data management subsystem finishes the cleaning and organization management of the remote sensing image data and the labeling result data, and comprises a data cleaning and warehousing module, a data retrieval module, an image catalog setting module and a labeling result management module, wherein the functions of the modules are as follows:
the data cleaning and warehousing module scans the specified folder path at regular time, automatically completes the analysis and warehousing of the remote sensing image data when new remote sensing image data is obtained, cleans the meta information of the remote sensing image data, and removes repeated fields and error value data in the meta information;
the data retrieval module is used for providing a retrieval function for image data and sample marking data in the database, and the retrieval conditions comprise satellite code numbers, imaging time, marking persons, sensor types, target types, marking modes, image resolution and image states;
the image directory setting module is used for setting image path information and adjusting the scanned specified folder path;
and the marking result management module provides functions of submitting, checking, importing and exporting the marking result, supports batch operation, and imports and exports file types including txt, json and xml formats.
(3) Information statistics subsystem
And the information statistics subsystem displays the statistics information of the remote sensing image data and the labeling result in the forms of a histogram, a pie chart and a curve chart, wherein the statistics information comprises an image sensor, a target type, an image resolution, a satellite code number, the number of newly added targets every day and a labeling person.
(4) Auditing management subsystem
And the auditing management subsystem can manage all the annotated images submitted for auditing, and auditors can audit the annotated results of the images submitted for auditing one by one to confirm whether the annotated results pass or not and need to give out reason instructions when the annotated results do not pass.
(5) Intelligent algorithm integration subsystem
The intelligent algorithm integration subsystem integrates various intelligent marking algorithms in a service calling mode, the intelligent marking algorithms comprise Mask R-CNN-based airplane target-level and pixel-level automatic marking algorithms, the intelligent marking algorithms are extensible, each intelligent marking algorithm is an independent service, and currently, target-level and pixel-level intelligent marking algorithms of airplane targets are preset in the system.
The specific working flow of the intelligent labeling method for the satellite remote sensing image sample based on the micro-service architecture is shown in fig. 3, and after a user finishes user registration and login through a client and sets a designated directory of a scanned image, the flow is as follows:
(101) scanning the appointed directory in real time according to the set appointed directory information;
(102) after a new remote sensing image is scanned, automatically analyzing the remote sensing image, extracting pixel information of the remote sensing image, and storing the pixel information in a database;
meta-information is extracted from the image file name and the image itself, including satellite code, image name, sensor type, spatial resolution, image size, and latitude and longitude information.
(103) After the image is put in a warehouse, automatically calling a related intelligent marking algorithm according to image load and resolution information to automatically mark the remote sensing image in a full picture manner, and putting the automatic marking result in the warehouse for storage;
the integrated various intelligent algorithms are classified according to the types and resolution ranges of the supported image sensors, and the intelligent algorithms are matched through the types and the resolution ranges of the sensors.
(104) After the automatic labeling is finished, automatically loading the automatic labeling result of the image after the image is loaded and displayed;
(105) according to the requirement, the automatic labeling result can be hidden or the intelligent algorithm can be called again for the counterweight region to perform target-level or pixel-level labeling;
(106) the automatic marking result is corrected manually and other attribute information is supplemented;
(107) after manual correction, the corrected labeling result is stored in a warehouse;
(108) and auditing the labeling result by an auditor, and forming a final labeling result after the auditing.
The embodiment is realized by JAVA, Javascript, standard C + + and Python languages, wherein an interface part is realized by the Javascript language, a background service main body is realized by the JAVA language, related algorithms related to remote sensing image processing are realized by the standard C + + and are integrally called in a dynamic library mode, intelligent processing algorithms already comprise airplane target-level and pixel-level intelligent labeling algorithms of sub-meter-level visible light images and are realized by the Python and the standard C + + languages, and the integrated calling is realized in a web service mode. Meanwhile, based on the high-resolution second remote sensing image, the automatic labeling and manual labeling capabilities of the airplane are verified in the graphic workstation environment, the target-level automatic labeling result is that the false alarm rate is 6.98%, the accuracy rate is 89.56%, and the error rate is 10.89% in the three high-resolution second remote sensing images, and on the basis, accurate labeling of the airplane can be completed through manual correction.
Claims (4)
1. A satellite remote sensing image sample labeling system based on a micro-service architecture is characterized by comprising a sample labeling subsystem, a data management subsystem, an information statistics subsystem and an intelligent algorithm integration subsystem;
the sample marking subsystem is used for loading and displaying the remote sensing image and the marking result and providing the functions of automatic marking and manual marking of the remote sensing image;
the data management subsystem is used for cleaning and organizing the remote sensing image data and the labeling result data;
the information statistics subsystem is used for displaying the statistics information of the remote sensing image data and the labeling result;
the intelligent algorithm integration subsystem integrates various intelligent marking algorithms and calls the algorithms in a web service mode, and the intelligent marking algorithms comprise target-level intelligent marking algorithms and pixel-level intelligent marking algorithms.
2. The satellite remote sensing image sample labeling system based on the micro-service architecture as claimed in claim 1, wherein the sample labeling subsystem comprises an image loading display module, an automatic labeling module, a manual labeling module, an information display and image basic operation module;
the image loading display module is used for realizing loading display of the remote sensing image and integrates a remote sensing image basic algorithm of image stretching and image resampling in a function calling mode;
the automatic labeling module is used for realizing the functions of automatic labeling of the whole graph and automatic labeling of the area; the intelligent marking algorithm in the intelligent algorithm integration subsystem is called in a service calling mode through a uniform service calling interface to realize automatic marking of the whole image or the designated area of the current remote sensing image;
the manual marking module is used for marking the target in the remote sensing image in a man-machine interaction mode;
the information display module is used for displaying image list information in a current database, currently loaded remote sensing image element information, real-time marking result information, an image list information display sequence number, an image name and a current marker;
the image basic operation module is used for completing adjustment of contrast, saturation and brightness of the remote sensing image and providing a palette function for adjusting colors of all the marking frames in the current image.
3. The satellite remote sensing image sample labeling system based on the micro-service architecture as claimed in claim 1, wherein the data management subsystem comprises a data cleaning and warehousing module, a data retrieval module, an image catalog setting module and a labeling result management module;
the data cleaning and warehousing module is used for scanning a specified folder path at regular time, automatically completing the analysis and warehousing of the remote sensing image data when new remote sensing image data is obtained, cleaning the meta information of the remote sensing image data, and removing repeated fields and error value data in the meta information;
the data retrieval module is used for providing retrieval functions for image data and sample labeling data in a database;
the image directory setting module is used for setting image path information and adjusting the scanned specified folder path;
the marking result management module is used for providing functions of submitting, checking, importing and exporting the marking result and supporting batch operation.
4. A satellite remote sensing image sample labeling method based on a micro-service architecture is characterized in that based on the system of claim 1, the method is realized through the following steps:
step one, the data management subsystem scans the set catalog at regular time, automatically analyzes the scanned image and cleans and stores the meta information;
secondly, the sample labeling subsystem acquires information of an image to be labeled from a database, loads the image to be labeled and selects a waveband of the multiband image;
step three, calling a background intelligent labeling algorithm to automatically label the whole image or the local area of the image;
step four, the intelligent algorithm integration subsystem completes target-level automatic labeling according to the request, completes pixel-level automatic labeling on the basis of the target-level automatic labeling, and then returns the result to the sample labeling subsystem;
step five, the sample labeling subsystem adjusts the contrast, saturation, brightness and color of the target labeling frame of the image according to the image condition;
step six, manually utilizing a marking tool to correct the automatic marking result, editing relevant attribute information, and storing the marking result in a warehouse;
and seventhly, counting and displaying the marking conditions of the images and the samples by the marking information stored in the database in the information counting subsystem.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110609043.XA CN113220920B (en) | 2021-06-01 | 2021-06-01 | Satellite remote sensing image sample labeling system and method based on micro-service architecture |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110609043.XA CN113220920B (en) | 2021-06-01 | 2021-06-01 | Satellite remote sensing image sample labeling system and method based on micro-service architecture |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113220920A true CN113220920A (en) | 2021-08-06 |
| CN113220920B CN113220920B (en) | 2022-08-12 |
Family
ID=77082239
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110609043.XA Active CN113220920B (en) | 2021-06-01 | 2021-06-01 | Satellite remote sensing image sample labeling system and method based on micro-service architecture |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113220920B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113989678A (en) * | 2021-11-18 | 2022-01-28 | 生态环境部卫星环境应用中心 | Method and device for constructing ecosystem classification sample library |
| CN114489829A (en) * | 2021-12-22 | 2022-05-13 | 北京市遥感信息研究所 | ArcMap-based remote sensing image sample labeling method |
| CN115170924A (en) * | 2022-07-22 | 2022-10-11 | 陕西航天技术应用研究院有限公司 | Intelligent interpretation system for air, space and ground big data |
| CN117315494A (en) * | 2023-11-29 | 2023-12-29 | 中国科学院空天信息创新研究院 | Collaborative concurrency labeling method and system based on regional association |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107273928A (en) * | 2017-06-14 | 2017-10-20 | 上海海洋大学 | A kind of remote sensing images automatic marking method based on weight Fusion Features |
| CN107886125A (en) * | 2017-11-09 | 2018-04-06 | 南京大学 | MODIS satellite remote sensing images mask methods based on local spectral factorization marking |
| CN109670060A (en) * | 2018-12-10 | 2019-04-23 | 北京航天泰坦科技股份有限公司 | A kind of remote sensing image semi-automation mask method based on deep learning |
| CN110851630A (en) * | 2019-10-14 | 2020-02-28 | 武汉市慧润天成信息科技有限公司 | Management system and method for deep learning labeled samples |
| CN111081353A (en) * | 2019-12-17 | 2020-04-28 | 浙江明峰智能医疗科技有限公司 | Method and system for automatically learning online and intelligently assisting in labeling medical images |
| CN111339800A (en) * | 2018-12-18 | 2020-06-26 | 中科星图股份有限公司 | Method and device for producing remote sensing sample data and computer readable storage medium |
| CN111580947A (en) * | 2020-04-29 | 2020-08-25 | 中国科学院空天信息创新研究院 | Online collaborative remote sensing image annotation system based on artificial intelligence |
| CN112329751A (en) * | 2021-01-06 | 2021-02-05 | 北京道达天际科技有限公司 | Deep learning-based multi-scale remote sensing image target identification system and method |
| CN112396128A (en) * | 2020-12-08 | 2021-02-23 | 中国铁路设计集团有限公司 | Automatic labeling method for railway external environment risk source sample |
| CN112613397A (en) * | 2020-12-21 | 2021-04-06 | 中国人民解放军战略支援部队航天工程大学 | Method for constructing target recognition training sample set of multi-view optical satellite remote sensing image |
-
2021
- 2021-06-01 CN CN202110609043.XA patent/CN113220920B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107273928A (en) * | 2017-06-14 | 2017-10-20 | 上海海洋大学 | A kind of remote sensing images automatic marking method based on weight Fusion Features |
| CN107886125A (en) * | 2017-11-09 | 2018-04-06 | 南京大学 | MODIS satellite remote sensing images mask methods based on local spectral factorization marking |
| CN109670060A (en) * | 2018-12-10 | 2019-04-23 | 北京航天泰坦科技股份有限公司 | A kind of remote sensing image semi-automation mask method based on deep learning |
| CN111339800A (en) * | 2018-12-18 | 2020-06-26 | 中科星图股份有限公司 | Method and device for producing remote sensing sample data and computer readable storage medium |
| CN110851630A (en) * | 2019-10-14 | 2020-02-28 | 武汉市慧润天成信息科技有限公司 | Management system and method for deep learning labeled samples |
| CN111081353A (en) * | 2019-12-17 | 2020-04-28 | 浙江明峰智能医疗科技有限公司 | Method and system for automatically learning online and intelligently assisting in labeling medical images |
| CN111580947A (en) * | 2020-04-29 | 2020-08-25 | 中国科学院空天信息创新研究院 | Online collaborative remote sensing image annotation system based on artificial intelligence |
| CN112396128A (en) * | 2020-12-08 | 2021-02-23 | 中国铁路设计集团有限公司 | Automatic labeling method for railway external environment risk source sample |
| CN112613397A (en) * | 2020-12-21 | 2021-04-06 | 中国人民解放军战略支援部队航天工程大学 | Method for constructing target recognition training sample set of multi-view optical satellite remote sensing image |
| CN112329751A (en) * | 2021-01-06 | 2021-02-05 | 北京道达天际科技有限公司 | Deep learning-based multi-scale remote sensing image target identification system and method |
Non-Patent Citations (2)
| Title |
|---|
| 戴芹等: "综合多特征遥感图像智能检索方法的概念设计", 《地球信息科学学报》 * |
| 黄冬梅等: "基于DBNMI模型的海洋遥感影像自动标注方法", 《中国科学技术大学学报》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113989678A (en) * | 2021-11-18 | 2022-01-28 | 生态环境部卫星环境应用中心 | Method and device for constructing ecosystem classification sample library |
| CN113989678B (en) * | 2021-11-18 | 2022-06-10 | 生态环境部卫星环境应用中心 | Method and device for constructing ecosystem classification sample library |
| CN114489829A (en) * | 2021-12-22 | 2022-05-13 | 北京市遥感信息研究所 | ArcMap-based remote sensing image sample labeling method |
| CN114489829B (en) * | 2021-12-22 | 2023-04-18 | 北京市遥感信息研究所 | Remote sensing image sample labeling method based on ArcMap |
| CN115170924A (en) * | 2022-07-22 | 2022-10-11 | 陕西航天技术应用研究院有限公司 | Intelligent interpretation system for air, space and ground big data |
| CN117315494A (en) * | 2023-11-29 | 2023-12-29 | 中国科学院空天信息创新研究院 | Collaborative concurrency labeling method and system based on regional association |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113220920B (en) | 2022-08-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113220920B (en) | Satellite remote sensing image sample labeling system and method based on micro-service architecture | |
| CN112115198B (en) | Urban remote sensing intelligent service platform | |
| US12032546B2 (en) | Systems and methods for populating a structured database based on an image representation of a data table | |
| Zhou et al. | Semantic understanding of scenes through the ade20k dataset | |
| Wulder et al. | Comparison of airborne and satellite high spatial resolution data for the identification of individual trees with local maxima filtering | |
| CN111580947B (en) | Online collaborative remote sensing image annotation system based on artificial intelligence | |
| CN109344223A (en) | Building information model management system and method based on cloud computing technology | |
| CN113988794A (en) | Multi-data integrated rural agricultural information system and method | |
| CN107506499A (en) | The method, apparatus and server of logical relation are established between point of interest and building | |
| CN112102443A (en) | Marking system and marking method suitable for substation equipment inspection image | |
| CN110008296A (en) | A kind of system and method for fast construction geographic information data application | |
| CN115858526A (en) | Multidimensional visual test data management system based on uncertain data source formats | |
| CN117315494A (en) | Collaborative concurrency labeling method and system based on regional association | |
| CN116992851A (en) | Water body change investigation document batch generation method based on remote sensing data | |
| CN114780074B (en) | Information computing system for realizing big data analysis and construction method | |
| CN112632162B (en) | Manufacturing production monitoring report system | |
| CN114385740A (en) | Visualization display system and method based on uranium mine geological cloud platform | |
| CN112508667A (en) | Financial data analysis system based on cloud native micro-service architecture | |
| Pham et al. | Interactive visualization of spatial and temporal patterns of diversity and abundance in ecological data | |
| CN114116686A (en) | Data visualization method for realizing data large screen | |
| WO1997026608A1 (en) | Authoring and publishing system for interactive multimedia computer applications | |
| CN110599892A (en) | Cultural industry monitoring method | |
| CN117523417B (en) | Method and electronic equipment applied to unified right-confirming registration of natural resources | |
| CN113393216B (en) | Laboratory digital system | |
| JP2001188703A (en) | Method for obtaining page information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |