CN109800671B - Target interpretation-oriented multisource remote sensing information knowledge graph construction method and system - Google Patents
Target interpretation-oriented multisource remote sensing information knowledge graph construction method and system Download PDFInfo
- Publication number
- CN109800671B CN109800671B CN201811625194.9A CN201811625194A CN109800671B CN 109800671 B CN109800671 B CN 109800671B CN 201811625194 A CN201811625194 A CN 201811625194A CN 109800671 B CN109800671 B CN 109800671B
- Authority
- CN
- China
- Prior art keywords
- target
- interpretation
- remote sensing
- information
- typical
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a target interpretation-oriented multisource remote sensing information knowledge graph construction method and a target interpretation-oriented multisource remote sensing information knowledge graph construction system, wherein the method comprises the following steps: automatically interpreting the multi-source remote sensing image to be processed according to the attribute feature recognition algorithm model group to obtain interpretation information; combining a knowledge graph framework of automatic interpretation of remote sensing images, carrying out primary classification and filing on the interpreted information to obtain a target automatic interpretation primary fusion information table; according to the target automatic interpretation primary fusion information table, by combining with a typical remote sensing target interpretation analysis ontology base, entity construction of interpretation information is carried out, and a knowledge map model is obtained; and expanding the knowledge map model, updating and completing the historical knowledge map according to the expanded knowledge map model, and completing knowledge extraction of the interpretation information. The method can complete the automatic extraction of typical targets of the multi-source remote sensing image, the comprehensive summary of the remote sensing image interpretation information and form the knowledge graph with a standardized expression form.
Description
Technical Field
The invention belongs to the technical field of remote sensing target interpretation, and particularly relates to a target interpretation-oriented multisource remote sensing information knowledge graph construction method and system.
Background
With the rapid development of earth observation technology, the observation capability of human beings on the earth reaches an unprecedented level. The satellite loads with different imaging modes, different wave bands and different resolutions enable the remote sensing data to be diversified day by day, the remote sensing data volume is obviously increased, the updating period is shortened, and the timeliness is stronger and stronger. The telemetric data exhibits significant "big data" characteristics.
In the traditional remote sensing image target interpretation process, the interpretation result can be obtained only after a large amount of data is consulted simultaneously by depending on the knowledge and experience of a remote sensing expert, so that the workload is large and the efficiency is low. It is therefore meaningful to accumulate, express, and use expert knowledge, and build a knowledge base system that assists the interpreter to solve the interpretation more quickly and efficiently.
In addition, a large amount of global satellites form massive remote sensing information, how to provide a high-efficiency technology and method for data management, information extraction, semantic analysis and the like, multi-source and heterogeneous data such as satellite remote sensing data, ground observation data, simulation models and the like and prior information in a plurality of platforms or knowledge bases are integrated, logic association expression, information semantic integration and collaborative comprehensive management of remote sensing big data are supported, semantic integration and interoperation of massive satellite data application services are finally realized, and shared platform construction is a problem which is urgently solved at present.
Disclosure of Invention
The technical problem of the invention is solved: the method and the system for constructing the multi-source remote sensing information knowledge graph for target interpretation are capable of completing automatic extraction of typical targets of multi-source remote sensing images and comprehensive summarization of remote sensing image interpretation information, forming the knowledge graph with a standardized representation form, and have the advantages of visual results, clear structure, accurate content, strong expandability and the like.
In order to solve the technical problem, the invention discloses a target interpretation-oriented multisource remote sensing information knowledge graph construction method, which comprises the following steps:
automatically interpreting the multi-source remote sensing image to be processed according to the attribute feature recognition algorithm model group to obtain interpretation information;
combining a remote sensing image automatic interpretation knowledge graph frame, and carrying out primary classification and filing on the interpretation information to obtain a target automatic interpretation primary fusion information table;
according to the target automatic interpretation primary fusion information table, combining a typical remote sensing target interpretation analysis ontology base to construct an entity of interpretation information to obtain a knowledge map model;
and expanding the knowledge map model, updating and completing the historical knowledge map according to the expanded knowledge map model, and completing knowledge extraction of the interpretation information.
Preferably, the method further comprises the following steps:
acquiring sample data;
extracting a typical target and a presence area of the typical target from the sample data;
and constructing the typical remote sensing target interpretation analysis ontology base according to the extracted typical target and the existence region of the typical target.
Preferably, the typical remote sensing target interpretation analysis ontology library carries at least one of the following information: target attributes indicating a target texture structure, a scale size, and a typical component; region attributes for indicating region typical feature type composition, administrative division name, geographical structure type, and geological structure type; and the incidence relation attribute is used for indicating main incidence relation types between the targets and the regions.
Preferably, the method further comprises the following steps:
preprocessing the multi-source remote sensing image to be processed to enable the preprocessed multi-source remote sensing image to meet the automatic interpretation quality requirement; wherein the pre-treatment comprises: image denoising, radiation correction and geometric correction.
Preferably, according to the attribute feature recognition algorithm model group, automatically interpreting the multi-source remote sensing image to be processed to obtain interpretation information, including:
according to the type of the typical remote sensing image and the related background knowledge of the interpreted target object, constructing an attribute feature recognition algorithm model group aiming at the typical target and the existing region of the typical target in the interpreted analysis ontology library of the typical remote sensing target;
and automatically interpreting the multi-source remote sensing image to be processed according to the attribute feature recognition algorithm model group, and extracting to obtain target information and background information in the multi-source remote sensing image to be processed.
Preferably, the automatically interpreting the primary fusion information table according to the target, and combining with a typical remote sensing target interpretation analysis ontology library to construct an entity of the interpretation information to obtain a knowledge map model, including:
according to the target automatic interpretation primary fusion information table, combining a typical remote sensing target interpretation analysis ontology library to construct an entity of interpretation information; and according to the target coordinates and target attribute information carried in the interpretation information, extracting the relation of 'target-target' and 'target-background' by combining spatial position judgment and the typical remote sensing target interpretation analysis ontology library, and completing the direct knowledge map model construction of the multi-source remote sensing image interpretation information to obtain a knowledge map model.
Preferably, the expanding the knowledge graph model, and updating the historical knowledge graph network according to the expanded knowledge graph model to complete knowledge extraction of the interpretation information, includes:
according to a long-chain indirect incidence relation reasoning algorithm and/or an incidence pattern learning algorithm, data mining is carried out on the knowledge graph model, and the knowledge graph model is expanded; and updating the historical knowledge map network according to the expanded knowledge map model to finish knowledge extraction of the interpretation information.
The invention also discloses a target interpretation-oriented multisource remote sensing information knowledge graph construction system, which comprises the following steps:
the interpretation module is used for automatically interpreting the multi-source remote sensing image to be processed according to the attribute feature recognition algorithm model group to obtain interpretation information;
the fusion module is used for carrying out primary classification and filing on the interpreted information by combining a knowledge graph framework of the automatic interpretation of the remote sensing image to obtain a target automatic interpretation primary fusion information table;
the map construction module is used for carrying out entity construction of interpretation information by combining a typical remote sensing target interpretation analysis ontology base according to the target automatic interpretation primary fusion information table to obtain a knowledge map model;
and the map updating module is used for expanding the knowledge map model, updating and completing the historical knowledge map according to the expanded knowledge map model, and completing knowledge extraction of the interpretation information.
The invention has the following advantages:
(1) the method can complete the automatic extraction of typical targets, the interpretation of the remote sensing images and the comprehensive collection of the interpreted information of the multi-source remote sensing images, form the knowledge graph with a standardized representation form, effectively relieve the problems of poor information integrity, insufficient information relevance, poor information accuracy and the like in the representation of the automatic interpreted information of the multi-source remote sensing images, and has the advantages of visual result, clear structure, accurate content, strong expandability and the like
(2) The invention utilizes the knowledge graph as a fusion model of multisource semi-structured data, provides an effective representation form for the high-efficiency summarization of interpretation output coordinates and attribute information for multisource remote sensing targets,
(3) The knowledge graph-based multi-source interpretation information synthesis method is beneficial to realizing the high-efficiency synthesis of large-scale multi-source remote sensing data interpretation information, and provides an effective technical solution for remote sensing big data analysis and data mining.
Drawings
FIG. 1 is a flowchart of steps of a target-oriented interpreted multisource remote sensing information knowledge graph construction method in an embodiment of the invention;
FIG. 2 is a framework diagram of a multisource target interpretation knowledge graph building process in an embodiment of the invention;
FIG. 3 is a flow chart of a multi-source remote sensing information fusion framework according to an embodiment of the present invention;
FIG. 4 is a flow chart of construction of a knowledge graph of multisource remote sensing target interpretation information in the embodiment of the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the embodiments of the present invention will be described in detail with reference to the accompanying drawings.
As shown in fig. 1, in this embodiment, the target-oriented interpreted multi-source remote sensing information knowledge graph construction method includes:
and 101, automatically interpreting the multi-source remote sensing image to be processed according to the attribute feature recognition algorithm model group to obtain interpretation information.
In this embodiment, an attribute feature identification algorithm model group for a typical target and an existing region of the typical target in an interpretation analysis ontology library of the typical remote sensing target can be constructed according to the type of the typical remote sensing image and the related background knowledge of the interpreted target object; and then, automatically interpreting the multi-source remote sensing image to be processed according to the attribute feature recognition algorithm model group, and extracting to obtain target information and background information in the multi-source remote sensing image to be processed. The target information carried in the interpretation information obtained by automatic interpretation mainly comprises the spatial position information of the target and various attribute information of the target. The spatial position information is expressed by using spatial geographical coordinates, and various attribute information of the target is displayed by using a multi-column table.
Preferably, for the difference between different targets and different image sources in data quantity and labeled quantity, a deep learning method and a classical feature modeling method can be used to complete the extraction of corresponding target detection and attribute identification information. Specifically, the method comprises the following steps:
the deep learning method comprises the following steps: and aiming at the situation that the data source and the labeled data are rich, a deep target feature modeling mode based on deep learning is used for completing model construction of target expression features and attribute features.
Classical feature modeling: and aiming at the condition that a data source and labeled data are scarce, extracting the position information and the attribute information of the target by using a shallow target model of 'SIFT + SVM classifier'.
And 102, combining the remote sensing image automatic interpretation knowledge graph framework, and carrying out primary classification and filing on the interpreted information to obtain a target automatic interpretation primary fusion information table.
In this embodiment, the targets from different remote sensing image data sources are associated through respective spatial position information, preliminary entity disambiguation is completed by using the spatial position information association, and therefore, different-source remote sensing image attribute association of each main target is completed, and a target multi-source attribute high-dimensional vector is formed.
And 103, according to the target automatic interpretation primary fusion information table, combining a typical remote sensing target interpretation analysis ontology library to construct an entity of interpretation information to obtain a knowledge map model.
In this embodiment, the entity construction of the interpretation information can be performed according to the target automatic interpretation primary fusion information table and by combining a typical remote sensing target interpretation analysis ontology library; and according to the target coordinates and target attribute information carried in the interpretation information, extracting the relation of 'target-target' and 'target-background' by combining spatial position judgment and the typical remote sensing target interpretation analysis ontology library, and completing the direct knowledge map model construction of the multi-source remote sensing image interpretation information to obtain a knowledge map model.
And 104, expanding the knowledge map model, updating and completing the historical knowledge map according to the expanded knowledge map model, and completing knowledge extraction of the interpretation information.
In this embodiment, a long-chain indirect association relationship inference algorithm and/or an association pattern learning algorithm may be adopted to perform data mining on the knowledge graph model and extend the knowledge graph model; and updating and complementing the historical knowledge map according to the expanded knowledge map model to finish knowledge extraction of the interpretation information.
In a preferred embodiment of the present invention, before performing automatic interpretation on a multi-source remote sensing image to be processed, the multi-source remote sensing image to be processed may be preprocessed to make the preprocessed multi-source remote sensing image to be processed meet the requirement of automatic interpretation quality, where the preprocessing includes, but is not limited to: and image denoising, radiation correction, geometric correction and the like.
In a preferred embodiment of the present invention, a typical remote sensing target interpretation analysis ontology library can be constructed in advance by the following steps: acquiring sample data; extracting a typical target and a presence area of the typical target from the sample data; and constructing the typical remote sensing target interpretation analysis ontology base according to the extracted typical target and the existence region of the typical target. The typical remote sensing target interpretation analysis ontology library carries at least one of the following information: target attributes (including typical attribute information such as a target texture structure, a scale size and a typical component), region attributes (including a region typical feature type composition, an administrative division name, a geographic structure type and a geological structure type and the like) and association attributes (including main association types between the target and between the target and the region and the like).
On the basis of the above embodiments, a specific example is described below.
As shown in fig. 2 to 4, taking statistical analysis tasks of types and quantities of the airplane parked and landed in the airport based on the visible light and the SAR image as an example, a construction process of a multisource remote sensing information knowledge graph interpreted towards a target is explained, which specifically includes:
(1) and constructing an airplane target interpretation analysis ontology library aiming at a typical airplane type and a typical airport based on the disclosed information data.
The airplane target interpretation analysis ontology library mainly comprises the following attribute information:
1a) object Property (P)obj): the method comprises typical attribute information of important components such as an airplane target texture structure, an airport wingspan, a nose wing and the like.
1b) Region attribute (P)reg): including the building structure of a typical airport, the name of the administrative division in which the airport is located, the geographic structure type, the geological structure type, etc. of the airport and the surrounding environment.
1c) Association attribute (L (obj, obj), L (obj, reg)): including the types of primary associations between different aircraft targets, aircraft and airports, such as adjacency, landing, taxi, etc.
(2) Obtaining a multisource remote sensing image set { IorigAnd according to the multisource remote sensing image set { I }origAutomatically identifying and interpreting the target of the airplane in the set of multi-source remote sensing images, and preprocessing various remote sensing images including image denoising, radiation correction, geometric correction and the like to obtain a preprocessed multi-source remote sensing image set (I)procAnd (4) making the standard of automatic interpretation and extraction.
(3) Establishing an attribute feature recognition algorithm model group aiming at main airplane types and distributed airports in an airplane target interpretation analysis ontology base based on various main remote sensing image types and relevant background knowledge of the interpreted target object
And finishing automatic interpretation and extraction of airplane and airport information in the multi-source remote sensing image data.
In the training process, aiming at the difference of different airplane models and different image sources in data quantity and labeled quantity, a deep learning method and a classical feature modeling method are used for completing the extraction of corresponding target detection and attribute identification information:
3a) a deep learning method is provided, which comprises the steps of,
and aiming at the situation that the image data source and the labeled data of the visible light airplane are rich, a depth target feature modeling mode based on deep learning is used for completing the model construction of the visible light representation features and the attribute features of the airplane.
3b) Classical feature modeling
And aiming at the situation that infrared, SAR and hyperspectral data sources and labeled data of the airplane are scarce, extracting target position information and attribute information by using a shallow target model of SIFT + SVM classifier.
3c) The aircraft information obtained by the automatic target interpretation model mainly comprises the spatial position information P of the aircraftobjAnd all kinds of aircraftSexual information Qobj. The spatial position information is represented by using spatial geographic coordinates, and the attribute information is displayed by using a multi-column table.
(4) According to attribute information of the airplane and the airport obtained by automatic interpretation of the multi-source remote sensing image, primary classification and filing of the interpreted information are carried out by combining an automatic interpretation knowledge map frame of the remote sensing image, and an automatic interpretation primary fusion information table of the airplane target is formed.
Airplane images from different remote sensing image data sources pass through respective geographic spatial position information PobjPerforming correlation, completing preliminary entity disambiguation by utilizing spatial position correlation, and completing attribute correlation of different-source remote sensing images of each main target to form a target multi-source attribute high-dimensional vector
(5) The method comprises the steps of automatically interpreting a primary fusion data information table based on a multi-source remote sensing image target, utilizing an already-constructed airplane target interpretation analysis ontology base to construct an airplane and airport entity for remote sensing image interpretation, and extracting 'airplane-airplane' and 'airplane-airport' relations by combining spatial position judgment and the ontology base according to extracted airplane coordinate and attribute information to complete the construction of a direct knowledge map model of multi-source remote sensing image interpretation information.
Constructed knowledge graph and
the triple mode is represented. Wherein the content of the first and second substances,
and expressing the space coordinate and the attribute vector of the relation host object in the association relation, and L expresses the type of the association relation.
(6) On the basis of a knowledge graph model directly constructed based on interpretation information, mining hidden attributes and association relations of interpretation target information is completed by using methods such as long-chain indirect association relation reasoning and association pattern learning, and new association relation triple is extracted
And expanding the existing knowledge graph structure model, and updating the existing intelligent aircraft interpretation knowledge graph network by using the expanded perfect knowledge graph to finish the knowledge extraction of interpretation information.
On the basis of the embodiment, the invention also discloses a target interpretation-oriented multisource remote sensing information knowledge graph construction system, which comprises: the interpretation module is used for automatically interpreting the multi-source remote sensing image to be processed according to the attribute feature recognition algorithm model group to obtain interpretation information; the fusion module is used for carrying out primary classification and filing on the interpreted information by combining a knowledge graph framework of the automatic interpretation of the remote sensing image to obtain a target automatic interpretation primary fusion information table; the map construction module is used for carrying out entity construction of interpretation information by combining a typical remote sensing target interpretation analysis ontology base according to the target automatic interpretation primary fusion information table to obtain a knowledge map model; and the map updating module is used for expanding the knowledge map model, updating and completing the historical knowledge map according to the expanded knowledge map model, and completing knowledge extraction of the interpretation information.
For the system embodiment, since it corresponds to the method embodiment, the description is relatively simple, and for the relevant points, refer to the description of the method embodiment section.
The embodiments in the present description are all described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.
The above description is only for the best mode of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention.
Those skilled in the art will appreciate that the invention may be practiced without these specific details.
Claims (2)
1. A target-oriented interpretation multi-source remote sensing information knowledge graph construction method is characterized by comprising the following steps:
acquiring sample data; extracting a typical target and a presence area of the typical target from the sample data; constructing a typical remote sensing target interpretation analysis ontology library according to the typical target obtained by extraction and the existing region of the typical target; the typical remote sensing target interpretation analysis ontology library carries at least one of the following information: target attributes indicating a target texture structure, a scale size, and a typical component; region attributes for indicating region typical feature type composition, administrative division name, geographical structure type, and geological structure type; an incidence relation attribute for indicating a main incidence relation type between the targets and the regions;
automatically interpreting the multi-source remote sensing image to be processed according to the attribute feature recognition algorithm model group to obtain interpretation information; the method comprises the following steps: according to the type of the typical remote sensing image and the related background knowledge of the interpreted target object, constructing an attribute feature recognition algorithm model group aiming at the typical target and the existing region of the typical target in the interpreted analysis ontology library of the typical remote sensing target; automatically interpreting the multi-source remote sensing image to be processed according to the attribute feature recognition algorithm model group, and extracting to obtain target information and background information in the multi-source remote sensing image to be processed;
combining a remote sensing image automatic interpretation knowledge graph frame, and carrying out primary classification and filing on the interpretation information to obtain a target automatic interpretation primary fusion information table; the method comprises the following steps: the method comprises the following steps that targets from different remote sensing image data sources are correlated through respective spatial position information, preliminary entity disambiguation is completed through spatial position information correlation, correlation of different-source remote sensing image attributes of main targets is completed, and target multi-source attribute high-dimensional vectors are formed;
according to the target automatic interpretation primary fusion information table, combining a typical remote sensing target interpretation analysis ontology base to construct an entity of interpretation information to obtain a knowledge map model; the method comprises the following steps: according to the target automatic interpretation primary fusion information table, combining a typical remote sensing target interpretation analysis ontology library to construct an entity of interpretation information; according to target coordinates and target attribute information carried in the interpretation information, target-target and target-background relation extraction is carried out by combining spatial position judgment and the typical remote sensing target interpretation analysis ontology library, and direct knowledge map model construction of multi-source remote sensing image interpretation information is completed to obtain a knowledge map model;
expanding the knowledge map model, updating and completing the historical knowledge map according to the expanded knowledge map model, and completing knowledge extraction of interpretation information; the method comprises the following steps: according to a long-chain indirect incidence relation reasoning algorithm and/or an incidence pattern learning algorithm, data mining is carried out on the knowledge graph model, and the knowledge graph model is expanded; updating the historical knowledge map network according to the expanded knowledge map model to finish knowledge extraction of the interpretation information;
wherein:
aiming at the difference of different targets and different image sources in data quantity and labeled quantity, a deep learning method and a classical feature modeling method are used for completing the extraction of corresponding target detection and attribute identification information; specifically, the method comprises the following steps:
the deep learning method comprises the following steps: aiming at the situation that the data source and the labeled data are rich, a deep target feature modeling mode based on deep learning is used for completing model construction of target expression features and attribute features;
classical feature modeling: and aiming at the condition that a data source and labeled data are scarce, extracting the position information and the attribute information of the target by using a shallow target model of 'SIFT + SVM classifier'.
2. The target-oriented interpreted multisource remote sensing information knowledge graph construction method of claim 1, characterized by further comprising:
preprocessing the multi-source remote sensing image to be processed to enable the preprocessed multi-source remote sensing image to meet the automatic interpretation quality requirement; wherein the pre-treatment comprises: image denoising, radiation correction and geometric correction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811625194.9A CN109800671B (en) | 2018-12-28 | 2018-12-28 | Target interpretation-oriented multisource remote sensing information knowledge graph construction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811625194.9A CN109800671B (en) | 2018-12-28 | 2018-12-28 | Target interpretation-oriented multisource remote sensing information knowledge graph construction method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109800671A CN109800671A (en) | 2019-05-24 |
| CN109800671B true CN109800671B (en) | 2021-03-02 |
Family
ID=66558089
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811625194.9A Expired - Fee Related CN109800671B (en) | 2018-12-28 | 2018-12-28 | Target interpretation-oriented multisource remote sensing information knowledge graph construction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109800671B (en) |
Families Citing this family (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110795932B (en) * | 2019-09-30 | 2021-03-30 | 中国地质大学（武汉） | Geological report text information extraction method based on geological ontology |
| CN110825882B (en) * | 2019-10-09 | 2022-03-01 | 西安交通大学 | Knowledge graph-based information system management method |
| CN112836060B (en) * | 2019-11-25 | 2023-11-24 | 中国科学技术信息研究所 | Atlas construction method and apparatus for technological innovation data |
| CN111126298B (en) * | 2019-12-25 | 2024-01-09 | 二十一世纪空间技术应用股份有限公司 | Remote sensing information interpretation method and device and electronic equipment |
| CN111428762B (en) * | 2020-03-12 | 2022-03-15 | 武汉大学 | Interpretable remote sensing image ground feature classification method combining deep data learning and ontology knowledge reasoning |
| CN111325184B (en) * | 2020-03-20 | 2023-04-18 | 宁夏回族自治区自然资源勘测调查院 | Intelligent interpretation and change information detection method for remote sensing image |
| CN111125294B (en) * | 2020-03-31 | 2020-06-26 | 武汉中科通达高新技术股份有限公司 | Spatial relationship knowledge graph data model representation method and system |
| CN111666313B (en) * | 2020-05-25 | 2023-02-07 | 中科星图股份有限公司 | Correlation construction and multi-user data matching method based on multi-source heterogeneous remote sensing data |
| CN113761971B (en) * | 2020-06-02 | 2023-06-20 | 中国人民解放军战略支援部队信息工程大学 | Remote sensing image target knowledge graph construction method and device |
| CN111639196B (en) * | 2020-06-03 | 2022-03-15 | 核工业湖州勘测规划设计研究院股份有限公司 | Multi-layer gradually-enhanced ground disaster knowledge graph and automatic completion method thereof |
| CN112445918A (en) * | 2020-11-27 | 2021-03-05 | 杭州海康威视数字技术股份有限公司 | Knowledge graph generation method and device, electronic equipment and storage medium |
| CN112507122A (en) * | 2020-12-01 | 2021-03-16 | 浙江易智信息技术有限公司 | High-resolution multi-source remote sensing data fusion method based on knowledge graph |
| CN112612902B (en) * | 2020-12-23 | 2023-07-14 | 国网浙江省电力有限公司电力科学研究院 | Knowledge graph construction method and device for power grid main equipment |
| CN112579813A (en) * | 2020-12-24 | 2021-03-30 | 上海湃星信息科技有限公司 | Remote sensing image retrieval method and device based on knowledge graph |
| CN112732963A (en) * | 2021-01-15 | 2021-04-30 | 浙江大学 | Remote sensing big data based cross-boundary service application method |
| CN113220894B (en) * | 2021-02-07 | 2023-08-18 | 国家卫星气象中心(国家空间天气监测预警中心) | Intelligent satellite remote sensing data acquisition method based on perception calculation |
| CN113448934A (en) * | 2021-06-25 | 2021-09-28 | 中科开采夫(海南)空天信息研究院有限公司 | Distribution platform, method, equipment and medium based on multi-source remote sensing data |
| CN113408663B (en) * | 2021-07-20 | 2022-04-08 | 中国科学院地理科学与资源研究所 | Fusion model construction method, fusion model using device and electronic equipment |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105373590A (en) * | 2015-10-22 | 2016-03-02 | 百度在线网络技术（北京）有限公司 | Knowledge data processing method and knowledge data processing device |
| US10314001B2 (en) * | 2016-12-22 | 2019-06-04 | Here Global B.V. | Method and apparatus for providing adaptive location sampling in mobile devices |
| CN107665252B (en) * | 2017-09-27 | 2020-08-25 | 深圳证券信息有限公司 | Method and device for creating knowledge graph |
-
2018
- 2018-12-28 CN CN201811625194.9A patent/CN109800671B/en not_active Expired - Fee Related
Non-Patent Citations (2)
| Title |
|---|
| Nurses’ knowledge and attitudes about antibiotic therapy in critical care;Cindy L 等;《Intensive and Critical Care Nursing》;20010831;第17卷(第4期);第213-218页 * |
| 遥感卫星特定领域大规模知识图谱构建关键技术;谢榕 等;《无线电工程》;20170430;第47卷(第4期);第1-6页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109800671A (en) | 2019-05-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109800671B (en) | Target interpretation-oriented multisource remote sensing information knowledge graph construction method and system | |
| Li et al. | A review of building detection from very high resolution optical remote sensing images | |
| Yin et al. | Airport detection based on improved faster RCNN in large scale remote sensing images | |
| CN103035006A (en) | High-resolution aerial image partition method based on LEGION and under assisting of LiDAR | |
| Bai et al. | Deep learning in different remote sensing image categories and applications: status and prospects | |
| Chen et al. | Object-based multi-modal convolution neural networks for building extraction using panchromatic and multispectral imagery | |
| CN112101189A (en) | SAR image target detection method and test platform based on attention mechanism | |
| Koutsoudis et al. | Multispectral aerial imagery-based 3D digitisation, segmentation and annotation of large scale urban areas of significant cultural value | |
| Jiang et al. | Arbitrary-shaped building boundary-aware detection with pixel aggregation network | |
| CN114166842A (en) | Town forest monitoring method based on cooperation of high-resolution remote sensing data and ground survey data | |
| Huang et al. | Urban Building Classification (UBC) V2-A Benchmark for Global Building Detection and Fine-grained Classification from Satellite Imagery | |
| Kodors et al. | Building recognition using LiDAR and energy minimization approach | |
| CN111639672B (en) | Deep learning city function classification method based on majority voting | |
| Liu et al. | TSCMDL: Multimodal deep learning framework for classifying tree species using fusion of 2-D and 3-D features | |
| Richards-Rissetto et al. | A 3D point cloud Deep Learning approach using Lidar to identify ancient Maya archaeological sites | |
| Dong et al. | Building extraction from high spatial resolution remote sensing images of complex scenes by combining region-line feature fusion and OCNN | |
| Gao et al. | IUNet-IF: identification of construction waste using unmanned aerial vehicle remote sensing and multi-layer deep learning methods | |
| Gallagher et al. | A Multispectral Automated Transfer Technique (MATT) for machine-driven image labeling utilizing the Segment Anything Model (SAM) | |
| CN111738201B (en) | Method and system for extracting remote sensing image of woodland based on region-of-interest network | |
| Li et al. | Measuring detailed urban vegetation with multisource high-resolution remote sensing imagery for environmental design and planning | |
| Zhou et al. | Multispecies individual tree crown extraction and classification based on BlendMask and high-resolution UAV images | |
| Shao et al. | Semantic segmentation of remote sensing image based on Contextual U-Net | |
| Wu et al. | A DCNN Geographic Object Extraction Method for National Geographic Condition Monitoring | |
| Duan et al. | Denoising and classification of urban ICESat-2 photon data fused with Sentinel-2 spectral images | |
| Wang et al. | A land-cover classification method of high-resolution remote sensing imagery based on convolution neural network |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210302 Termination date: 20211228 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |