CN112561832B - Remote sensing image data storage method and system - Google Patents
Remote sensing image data storage method and system Download PDFInfo
- Publication number
- CN112561832B CN112561832B CN202011551192.7A CN202011551192A CN112561832B CN 112561832 B CN112561832 B CN 112561832B CN 202011551192 A CN202011551192 A CN 202011551192A CN 112561832 B CN112561832 B CN 112561832B
- Authority
- CN
- China
- Prior art keywords
- image
- data
- epipolar line
- layer
- matching
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 77
- 238000013500 data storage Methods 0.000 title claims abstract description 42
- 238000003860 storage Methods 0.000 claims abstract description 68
- 238000012545 processing Methods 0.000 claims abstract description 38
- 230000009466 transformation Effects 0.000 claims abstract description 25
- 238000012937 correction Methods 0.000 claims abstract description 19
- 238000004458 analytical method Methods 0.000 claims abstract description 10
- 238000007781 pre-processing Methods 0.000 claims abstract description 9
- 230000005855 radiation Effects 0.000 claims abstract description 6
- 230000005540 biological transmission Effects 0.000 claims abstract description 5
- 238000005520 cutting process Methods 0.000 claims abstract description 4
- 238000005457 optimization Methods 0.000 claims description 34
- 238000004422 calculation algorithm Methods 0.000 claims description 24
- 238000004366 reverse phase liquid chromatography Methods 0.000 claims description 22
- 230000008569 process Effects 0.000 claims description 13
- 230000002829 reductive effect Effects 0.000 claims description 11
- 238000009499 grossing Methods 0.000 claims description 7
- 238000013507 mapping Methods 0.000 claims description 5
- 238000001514 detection method Methods 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000012217 deletion Methods 0.000 claims description 3
- 230000037430 deletion Effects 0.000 claims description 3
- 238000009825 accumulation Methods 0.000 claims description 2
- 238000011156 evaluation Methods 0.000 claims description 2
- 241001270131 Agaricus moelleri Species 0.000 claims 4
- 230000001902 propagating effect Effects 0.000 claims 1
- 230000006870 function Effects 0.000 description 16
- 238000010586 diagram Methods 0.000 description 15
- 238000007726 management method Methods 0.000 description 13
- 238000004519 manufacturing process Methods 0.000 description 9
- 230000000694 effects Effects 0.000 description 7
- 238000000605 extraction Methods 0.000 description 7
- 238000013523 data management Methods 0.000 description 6
- 230000008520 organization Effects 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 5
- 230000007246 mechanism Effects 0.000 description 5
- 238000012876 topography Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000013508 migration Methods 0.000 description 4
- 230000005012 migration Effects 0.000 description 4
- 230000001360 synchronised effect Effects 0.000 description 4
- 238000010276 construction Methods 0.000 description 3
- 238000009432 framing Methods 0.000 description 3
- 230000008676 import Effects 0.000 description 3
- 230000010354 integration Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 230000002776 aggregation Effects 0.000 description 2
- 238000004220 aggregation Methods 0.000 description 2
- 238000013075 data extraction Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000005192 partition Methods 0.000 description 2
- 238000009877 rendering Methods 0.000 description 2
- 230000007704 transition Effects 0.000 description 2
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 238000012550 audit Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 230000006854 communication Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000002354 daily effect Effects 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000006837 decompression Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 230000006740 morphological transformation Effects 0.000 description 1
- 230000036961 partial effect Effects 0.000 description 1
- 238000013439 planning Methods 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 239000000243 solution Substances 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G06T5/77—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/27—Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/10—Image enhancement or restoration by non-spatial domain filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20016—Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20036—Morphological image processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20048—Transform domain processing
- G06T2207/20064—Wavelet transform [DWT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30181—Earth observation
Abstract
The invention provides a remote sensing image data storage method and a remote sensing image data storage system, which are used for acquiring remote sensing image data; preprocessing and texture analysis are carried out on the three-dimensional remote sensing image data, an image pyramid is generated through wavelet transformation, top-layer images of the pyramid are transformed and matched based on message transmission, layer-by-layer matching is carried out, and a digital surface model is generated; carrying out radiation correction, differential correction and mosaic on the three-dimensional remote sensing image data subjected to scanning treatment by utilizing a digital elevation model, and cutting according to a set picture range to generate a digital orthographic image; according to the remote sensing image data acquisition time, storing the data in a set time period in a local storage area, storing the data outside the set time period in an external storage area, wherein each area comprises a plurality of storage nodes, and respectively transmitting the original remote sensing image data and the processing result generated based on the original data to different storage nodes for storage. The invention improves the storage efficiency of the image data.
Description
Technical Field
The invention belongs to the technical field of image data storage, and particularly relates to a remote sensing image data storage method and system.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
Along with the rapid development of remote sensing technology and computer information technology, remote sensing image products are applied in more and more fields, and the resolution requirements of partial application fields on remote sensing image data are higher and higher, and the data volume is increased. It is expected by researchers that the remote sensing data volume will increase by 20% -30% each year.
Because image data is huge and complex, in the prior art, when storing image data, the image data is generally compressed or divided into different image blocks, and then the image blocks are distributed to different processing nodes for processing. However, these methods still have problems of poor processing speed and low efficiency; in addition, the processed image blocks are easy to lose or can not be spliced because of complex processing such as displacement, rotation, affine, interpolation, distortion, offset and the like, and the problem of repeated processing is also solved. The processing speed of the remote sensing image data is seriously affected.
Disclosure of Invention
In order to solve the problems, the invention provides a remote sensing image data storage method and a remote sensing image data storage system.
According to some embodiments, the present invention employs the following technical solutions:
a remote sensing image data storage method comprises the following steps:
acquiring remote sensing image data;
preprocessing and texture analysis are carried out on the three-dimensional remote sensing image data, an image pyramid is generated through wavelet transformation, top-layer images of the pyramid are transformed and matched based on message transmission, layer-by-layer matching is carried out, and a digital surface model is generated;
carrying out radiation correction, differential correction and mosaic on the three-dimensional remote sensing image data subjected to scanning treatment by utilizing a digital elevation model, and cutting according to a set picture range to generate a digital orthographic image;
according to the remote sensing image data acquisition time, storing the data in a set time period in a local storage area, storing the data outside the set time period in an external storage area, wherein each area comprises a plurality of storage nodes, and respectively transmitting the original remote sensing image data and the processing result generated based on the original data to different storage nodes for storage.
The method further comprises the steps of carrying out topography mapping auxiliary adjustment, reducing the resolution of the stereoscopic image, extracting the connection point of each fast view image pair, restoring the resolution of the fast view, extracting the conventional connection point of the original image by combining with the topography mapping elevation constraint of the spaceflight plane radar, carrying out original influence free network evaluation, extracting mountain top characteristic points and flat area characteristic points according to a certain interval, respectively projecting the two types of characteristic points onto the original influence, and extracting dense connection points.
The method further comprises the step of carrying out overlay product making module, and carrying out concurrent map tile drawing based on the time, the data type, the area information and the cloud cover information of the stereoscopic image data input to generate an overlay product.
As an alternative implementation manner, different remote sensing image data are transmitted to different processing nodes, and data processing is performed in parallel between the processing nodes.
As a further defined embodiment, before the remote sensing image data processing process is performed, determining the maximum parallel number and the computing power of each processing node, and determining N processing nodes with top ranked computing power to perform data processing tasks according to the maximum parallel number N.
As a further defined embodiment, remote sensing image data processing tasks are grouped, the tasks and the processing nodes are matched and allocated by using an optimal algorithm according to the computing capacity ranking of each processing node.
As an alternative implementation mode, before data storage, the data integrity is checked, and if the data integrity meets the requirement, the data is stored.
A remote sensing image data storage system comprising:
the data receiving module is configured to acquire remote sensing image data;
The digital surface model generation module is configured to perform preprocessing and texture analysis on the three-dimensional remote sensing image data, generate an image pyramid through wavelet transformation, transform a top layer image of the pyramid, perform matching based on message propagation, perform layer-by-layer matching and generate a digital surface model;
the digital orthophoto map generating module is configured to utilize a digital elevation model to carry out radiation correction, differential correction and mosaic on the three-dimensional remote sensing image data subjected to scanning treatment pixel by pixel, and cut according to a set map range to generate a digital orthophoto map;
the data ordering module is configured to store the data of the set time period in the local storage area according to the remote sensing image data acquisition time, and store the data outside the set time period to the external storage area;
and the data storage module is configured to respectively transmit the original remote sensing image data and the processing result generated based on the original data to different storage nodes for storage.
A computer readable storage medium having stored therein a plurality of instructions adapted to be loaded by a processor of a terminal device and to perform the steps of a remote sensing image data storage method.
A terminal device comprising a processor and a computer readable storage medium, the processor configured to implement instructions; the computer readable storage medium is for storing a plurality of instructions adapted to be loaded by a processor and to perform the steps of the one remote sensing image data storage method.
Compared with the prior art, the invention has the beneficial effects that:
the invention is that
In order to make the above objects, features and advantages of the present invention more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the invention.
FIG. 1 is a schematic diagram of data storage management;
FIG. 2 is a schematic diagram of a data management flow;
FIG. 3 is a schematic diagram of a data flow process of external data of the system;
FIG. 4 is a schematic diagram of a data flow process of internal data of the system;
FIG. 5 is a schematic diagram of a parallel scheduling framework;
FIG. 6 is a diagram of a physical architecture of a data storage system;
FIG. 7 is a diagram of a network relationship structure;
FIG. 8 is a schematic diagram of a data management flow based on modeling techniques;
FIG. 9 is a schematic diagram of a data parallel warehousing flow based on autonomous tasks;
FIG. 10 is a schematic diagram of a fast retrieval implementation of an image data storage system;
FIG. 11 (a) is a schematic diagram showing the comparison of the matching effect Gao Chengdian of the present system with the SGM algorithm;
FIG. 11 (b) is a schematic diagram showing a comparison of the DSM extracted by the present system with the DSM topography aspects extracted by other commercial software;
FIG. 12 is a specific process for automatically generating DSM by global optimization matching strategy;
FIG. 13 is an SRTM assisted adjustment flow;
FIG. 14 is a DSM rendering effect diagram;
FIG. 15 is a schematic view of an image obtained in the present embodiment;
FIG. 16 is a graph showing the result of matching control points for image data according to the prior art;
fig. 17 shows the matching result of the control points of the system on the image data.
The specific embodiment is as follows:
the invention will be further described with reference to the drawings and examples.
It should be noted that the following detailed description is illustrative and is intended to provide further explanation of the invention. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the present invention. As used herein, the singular is also intended to include the plural unless the context clearly indicates otherwise, and furthermore, it is to be understood that the terms "comprises" and/or "comprising" when used in this specification are taken to specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof.
A remote sensing image data storage system comprising:
the data receiving module is configured to acquire remote sensing image data;
the digital surface model generation module is configured to perform preprocessing and texture analysis on the three-dimensional remote sensing image data, generate an image pyramid through wavelet transformation, transform a top layer image of the pyramid, perform matching based on message propagation, perform layer-by-layer matching and generate a digital surface model;
the digital orthophoto map generating module is configured to utilize a digital elevation model to carry out radiation correction, differential correction and mosaic on the three-dimensional remote sensing image data subjected to scanning treatment pixel by pixel, and cut according to a set map range to generate a digital orthophoto map;
the data ordering module is configured to store the data of the set time period in the local storage area according to the remote sensing image data acquisition time, and store the data outside the set time period to the external storage area;
and the data storage module is configured to respectively transmit the original remote sensing image data and the processing result generated based on the original data to different storage nodes for storage.
The system is intended to realize effective management of massive multi-source heterogeneous remote sensing image data, improve the remote sensing image data processing and management efficiency, improve the remote sensing image data sharing and service level, and provide data guarantee for important projects such as basic geographic information resource construction, geographic condition-saving monitoring, emergency mapping and the like.
The data is first put in storage and stored. The data in warehouse include:
(1) Aerial remote sensing image original data
Mainly comprises ADS series, UC series, SWDC-4, SWDC-5, CS10000, DMCIII and the like, and the system also needs to support an expansion function to meet the data types of the follow-up novel aerial camera.
Frame-type original aerial photo
Frame data are divided into five types according to aerial photography: SWDC-4, SWDC-5, UC series, CS10000 and DMCIII. The data format is that a suffix corresponding to a plurality of aerial photo data is txt metadata file, and the metadata file records the corresponding spatial position information of each aerial photo according to rows, and mainly comprises the following specific steps: attribute information such as GPS Time(s), easting (degrees), northing (degrees), ellht (metres), omega (deg), phi (deg), kap (deg), lat (deg), lon (deg), etc., wherein the spatial location information is a point for locating the center point location of the web. The aerial photo data is in tiff format.
ADS push-broom type original data
The method comprises the steps of dividing L0 level data, L1 level data, L2 level data and the like, wherein L0 level data is original data, the L1 level data is generated by decompression through special software, the L1 level data and the L2 level data are large in data quantity, only the L0 level data are required to be backed up and managed, the shp is provided as a route connection table of a certain frame, and the space position information is multi-line.
(2) Original data of space remote sensing image
The data organization structure comprises: the tif image file, the xml metadata file, the rpc file and other auxiliary files, and the geolange folder store the spatial range of the image file.
(3) The achievement data includes: DOM, DEM, DLG and point cloud data.
The DOM data includes: satellite image overall view results and 1:500-1:25000 DOM results.
The result data organization structure comprises: a tif image file, a tfw file, an xml file, an ovr file, an xls metadata file, and a shp spatial extent join table file.
DEM outcome data, data organization structure includes: and (3) carrying out data organization according to a 1:10000 picture number specified in national basic scale topographic map framing and numbering (GB/T13989-2012) as a folder name, and storing metadata in an excel file form.
DLG outcome data, data organization structure comprising: the method comprises the steps of organizing data according to a picture number specified by national standard of national basic scale topographic map framing and numbering made in 2012 as a folder name, and storing metadata separately from a data entity, wherein the metadata is stored in an excel file form, and the organization of the metadata is also carried out according to the picture number as the folder name.
As shown in fig. 1, database storage is combined with file storage.
And the high-efficiency and safe storage of various data is carried out by comprehensively utilizing databases of different types and file storage systems. The method adopts a mode of combining database storage and file storage, wherein the database storage adopts a relational database (spatial database), and the file storage adopts shared file storage.
1) From the data quantity perspective, the remote sensing image data and the product result data are large in quantity, and the entity data are stored by adopting files.
2) From the data structure point of view, metadata relates to space data and attribute data, and is uniformly stored by adopting a relational database.
3) From the aspect of application requirements, the spatial layer data is mainly used for browsing, extracting and other services, a spatial database is used for storing, and file storage with high IO operation is needed for image files and the like.
4) From the viewpoints of data security migration and cost, for data with large access quantity, recently accessed data and latest achievements, the data are stored in a business intranet filing disk array, historical data entities are cleaned locally according to the needs, and only meta-information of the data is reserved.
The method takes aviation remote sensing data with different sources and different formats, various aerospace satellite remote sensing image data, produced product result data and the like which are collected by various production departments as main data sources, provides data service capability for production, migration and distribution on the basis of a comprehensive database of remote sensing images in Shandong province, comprises data archiving service, data query service and data extraction service, realizes unified data access service, supports data access to file and database formats, shields the difference of physical storage of various data, and transparentizes the physical storage of the data, thereby facilitating the use of various users.
And carrying out quick warehousing centralized management on the mass remote sensing image data and the product data acquired and generated every day, and realizing quick warehousing of the mass data by adopting task-based parallel archiving. The data parallel warehouse-in technical flow based on autonomous tasks is as shown in fig. 9, task information is acquired, writing and reading are performed, data file information to be archived is acquired, writing and reading are performed, metadata files, file list information and fast try files are archived, writing and reading are performed, data file information task allocation information is used for data to be copied, the data to be copied is allocated to a corresponding server in a load balancing mode, writing and reading are performed, and copying task result information is obtained.
The important objective of the image data storage system is to realize the integrated management of the aerospace remote sensing image data and the product result, so that the data circulation of various data in the system is necessary to be clarified. The overall flow of data management is shown in fig. 2, and specifically includes:
(1) For aerospace remote sensing data and product result data, a data integration module is required to perform standardized data integration work, and then data classification is required. The data integration module comprises working flows of data integrity check, availability check, data type identification and the like, wherein the data integrity check is to check an image data file, check whether a related reference file or an attribute file is absent, reject the data missing from the file and ensure the data integrity; the usability check is mainly to check whether the data can be decompressed or not, whether the file is damaged or not, and the problem data found by the check is also removed, so that the usability of the data is ensured.
(2) And extracting metadata and auxiliary information according to the belonged data category. According to different data types, different database tables are designed to ensure that metadata and auxiliary information data of each type of data are completely and effectively extracted and input.
(3) And (3) checking the data format and the integrity of metadata information, data constitution, auxiliary information and the like of the data, archiving and warehousing the data which accords with the checking, and marking the data with missing information so as to check the data later.
(4) And carrying out data creation, metadata registration and data body injection on the archive catalog according to the classified setting, and completing data archiving.
(5) And the data storage is stored online according to the storage strategy. The new warehouse-in data are all stored on line, and the historical data or other data with low use frequency required by the user are stored off line.
(6) And each service system performs data query through a query and search interface and extracts required data. Different business users have different authorities, after the common users submit data demands, super administrator users audit orders, and the data can be extracted through the rear users.
(7) And realizing the data synchronization work of the internal data and the data release server through data synchronization.
The system external data stream mainly describes how data enters the image data storage system and how data is exported to a user through the system. The external data flow of the system is mainly divided into: data archiving, data querying, data extraction and data synchronization, and specific system external data flow is shown in fig. 3. The system internal data flow mainly describes how to flow between each functional module after the data enter the image data storage system. The image data storage system classifies, inspects, files and stores data according to a data archiving task, is responsible for data resource and system maintenance management, provides inquiry and retrieval service of various archived data and statistics and analysis of related data information, provides reference for a system manager maintenance system, and completes the data synchronization function for the Internet and the electronic government intranet image data storage system. A specific system internal data flow is shown in fig. 4.
Of course, in order to achieve better management of data, specifications for various data storage space selection, naming, and other operations may be set. These are all flexible to the skilled person and are not described in detail here.
With the expected total management data reaching PB level, for data archiving, statistics and other services, the conventional single machine processing mode cannot meet the requirements of timely archiving and efficient statistics of data, and the system construction needs to adopt parallel architecture design. The system construction is realized based on an autonomous parallel computing framework, and by adopting the framework, the parallel processing capacity and the performance of the system are mainly limited by the number of parallel task processing nodes, the network throughput and the performance of disk array IO.
The parallel scheduling framework supports parallel resource scheduling, parallel task allocation and parallel computing execution, and a data grading subsystem, a data retrieval subsystem and a statistical analysis subsystem are developed based on the framework, so that multi-machine and multi-process parallel execution of archiving, retrieving and statistical services is realized. During execution, the data archiving, searching and counting tasks are decomposed into a plurality of subtasks, the subtasks are distributed to corresponding execution nodes for execution, and after the execution of each node is completed, the execution results are summarized and returned. By distributing and executing serial flows in parallel, the data archiving efficiency is improved, the data query speed is increased, and the data statistics time is shortened.
As shown in fig. 5, in the present system, parallel scheduling is employed. Registration and management of parallel computing node resources is supported. The maximum parallel number and the computing capacity of the nodes can be reasonably configured to fully utilize the computing resources. And a task parallel scheduling algorithm is realized. And grouping task items by adopting an optimal algorithm, and distributing the task items to the nodes for execution according to the resource occupation condition and the performance of each computing node. And task scheduling service is provided, a high-availability mechanism is realized, and a single-point problem is avoided. Scheduling and allocation of parallel tasks is performed based on the ideas of plug-ins and workflows.
As shown in fig. 6, in the system architecture, supported by the server, in this embodiment, it may include: calculation layer: the system comprises a parallel computing server, a task allocation scheduling server and an application server, wherein the application server is used for background parallel scheduling and computing of services such as parallel archiving, parallel data retrieval, parallel data cleaning, parallel statistics and the like.
Storage layer: the system comprises a database server, a database server and a database server, wherein the database server is used for storing contents such as space data, attribute data, quick view data, document data and the like so as to meet high transaction request response and concurrent access requests under the condition of large data volume; the storage device is used for storing the aerospace remote sensing image and the product result data, the mobile hard disk is used for data migration and backup, and the mobile hard disk is stored in an online mode as shared storage and in an offline mode as mobile hard disk.
Application layer: the system can be a workstation, a desktop, a notebook or a remote desktop, and is used for business operations such as data management, data browsing and inquiring of a client.
Network layer: it is desirable to concatenate the compute layer, the storage layer, and the application layer to provide an efficient data transmission path between each other.
Meanwhile, in order to support the user to access and review the related data, as shown in fig. 7, the system provides the intranet and extranet services. As shown in fig. 7, the system is provided with 3 sets of networks, wherein the 3 sets of networks are respectively deployed, namely a service intranet, an e-government intranet and the Internet, the service intranet is an internal private network of a unit established by adopting the Internet technology, and the system uses a TCP/IP protocol as a basis and Web as a core application to form a unified and convenient information exchange platform; the electronic government affairs intranet is a secret-related party government agency office business network and is physically isolated from the Internet, and the electronic government affairs intranet is interconnected with the national electronic government affairs intranet in the upper part of the enterprise under the premise of meeting the working requirements. The public service and the internal service are connected by the gateway, and the gateway is used for exchanging information only in a data ferry mode (the general protocols such as HTTP, FTP, SMTP of the gateway are all closed or do not provide support for the protocols) instead of the traditional firewall; the internet also becomes an external network, and is connected with all open information and resources outside, so that the information storage capacity is large, the information transmission is rapid and efficient, the oriented user range is wide, and the method is not limited by regions.
The data storage system adopts a modularized design, the whole system covers the functions of all internal networks, the Internet and government affair networks, different networks are distinguished through configuration files, different functional modules are loaded, and finally, the adaptation of the different networks is realized:
business intranet system: the user object is an internal user, and according to the authority of the login user, the modules such as data archiving and warehousing, space retrieval, data query statistics and the like can be automatically loaded, so that the daily work requirements of the internal user of the unit are met, and all functions of the system are covered. The system needs to manage all original data images and product result images, all data storage needs to be realized through an intranet system, and data catalogs used by an internet system and an e-government intranet system need to be realized through a data synchronization function of the intranet system, and the intranet system has a synchronous data export function.
The Internet system: the user object is a public user, a small amount of public version data inquiry and downloading functions are provided, the data source of the Internet system is realized by performing a data synchronization function with the intranet system, and the Internet system has a synchronous data importing function.
E-government intranet system: the user object is a government user, and provides a management and query function of part of original data and part of confidential data, the data source of the electronic government intranet system is realized by performing a data synchronization function with the intranet system, and the electronic government intranet system has a synchronous data importing function.
And 3 sets of networks are physically isolated from each other due to different security levels among the business intranet system, the internet system and the e-government intranet system, the data synchronization function cannot realize cross-system automation, and synchronous data files are required to be manually copied to the internet system and the e-government intranet system.
Of course, some data confidentiality and security protection measures, such as encryption, authority setting, account management, data consistency check, etc., are also required in the communication process, and all of these may be adopted in the prior art, which is not described herein.
Because the data is multi-source, multi-standard and multi-format, and unified standard storage of the data is needed, the system adopts a data management mode based on modeling technology, as shown in fig. 8, the data modeling process can be divided into three layers, namely catalogue planning, data modeling and physical storage. The resource cataloging system is divided into three layers, namely cataloging nodes, abstract nodes and data nodes. The cataloging node and the data node have only one level, and the abstract node has multiple levels. The cataloging node is the root node of the cataloging of the database. The abstract nodes are logical classification nodes of data, and the data nodes are data nodes of each data product. For each data node, the image data storage system manages in a modeled manner, rather than directly defining each data product.
The data modeling is a process of modeling and defining one type of data, the data comprises two parts of descriptive information and entities, wherein the descriptive information comprises data metadata and spatialization information, the entities comprise image fast-try and image entities, the data modeling is an instantiation process of the data metadata, the spatialization information, the data types and the storage positions, and the spatialization information and the storage positions are expansion of the data and are not necessary conditions.
The image data storage system adopts a mode of combining catalogs and metadata to manage data, and manages space data in a space data set mode, a set of unified database management base platform is used as a base to construct a resource library which accords with respective characteristics, and simultaneously, the resource catalogs of the two are registered in the resource catalog storage system to support the release and sharing of resource information.
The system comprises an engineering establishment subsystem, a control point, a check point acquisition subsystem, a DSM editing subsystem, a control point, a connection point extraction and area network adjustment subsystem, a DSM production subsystem and a DOM production subsystem.
The engineering establishment subsystem is used for automatically searching the SC-level satellite images according to the set data catalogue, and matching and grouping the searched images according to indexes such as acquisition time, position, overlapping degree and the like, namely stereo image pairs, multispectral images, full-color images and multispectral images with the same visual angle; checking the integrity of the images and RPC, displaying all the images as a block diagram, and storing the images as shp files.
And the control point and check point acquisition subsystem is used for calling out the SC-level image according to the manual input ground coordinates, calling out the control base map simultaneously under the condition that the control base map is set, and obtaining the same-name point of the overlapped image through matching after puncturing the point on one image, so as to call out the image at the selected position.
The DSM editing subsystem is used for simultaneously opening the DSM to be edited and the replacement DSM (such as SRTM), checking the vulnerability area of the DSM to be edited, realizing vulnerability repair, simultaneously opening the DSM to be edited and the DOM, checking the water area on the DOM, realizing regular cutting of the water boundary on the DSM, and realizing deletion of the building or water surface placement of the selected area.
The control point and connection point extraction and adjustment subsystem is used for automatically matching control points (stereo and non-stereo) in a reference base map and a reference DEM in a cluster environment, acquiring SRTM control points (stereo) facing stereoscopic image processing in the cluster environment, automatically acquiring connection points (stereo and non-stereo) in the cluster environment and adjusting regional network in the cluster environment (stereo and non-stereo).
The DSM production subsystem is used for automatically producing the scenery-dividing DSM in the cluster environment, automatically splicing the DSM in the cluster environment, framing according to the degree, dividing 1*1 pixels and 2 x 2 pixels by the DSM matching distance, adjusting the resolution of the DSM and supporting the matching of the two-line array and the three-line array stereoscopic images.
The data storage system supports a data query function and mainly comprises three modules of data query conditions, query results and coverage rate results, wherein the query conditions support spatial range query modes such as manual drawing, administrative area query, picture frame query, text import and shp import, and attribute query modes such as time, resolution, cloud cover, sensors and picture frame numbers. In order to facilitate the inquiry of different demands of users, the system supports two display modes of a vector base map and an image base map, and additionally has an image annotation base map and a vector annotation base map, thereby facilitating the positioning inquiry of users.
And the DOM production subsystem is used for correcting the 8-bit and 16-bit orthographic images in the cluster environment, fusing the 8-bit and 16-bit panchromatic and multispectral orthographic images in the cluster environment, and integrally homogenizing the 8-bit RGB orthographic images in the cluster environment.
In the data processing process, the key technical scheme mainly comprises the following steps:
(1) Pixel-by-pixel dense matching plus global optimization
The stereoscopic image global optimization matching algorithm aims at global energy minimization, and overall matching is achieved through global optimization based on an energy equation. The energy equation generally includes a data item and a smoothing item. The data item refers to the similarity of a corresponding point of a certain parallax corresponding to the stereoscopic image, and can be represented by one or a combination of a plurality of gray scale distance, mutual information, correlation coefficient, CENSUS distance and the like; the smoothing term refers to a penalty parameter applied when the parallax of two adjacent pixels changes, and the smoothing term and the corresponding global optimization method are key steps different from a local matching algorithm. The global optimization matching algorithm is essentially that the matching degree of points and the parallax smoothness are balanced in the whole matching area, and a result superior to that of the local matching algorithm can be obtained under most conditions. However, the global optimization matching algorithm has the defects of large calculation amount, large memory consumption and the like, and the global optimization can form the aggregation or extension effect of parallax, and the aggregation or extension direction is not necessarily carried out along the ideal direction, so that the incorrect matching result can be caused.
Common global optimization matching algorithms are confidence propagation algorithms (Belief Propgation, BP), graph cut algorithms (GC), total Variation (TV), and generalized Total Variation (Total Generalized Variation, TGV). The traditional global optimization algorithm implies a kind of "front parallel"
The (front-Parallel) effect, i.e. better "like" or "good" match the same plane of parallax, may occur in areas of parallax variation, especially in areas of weak texture where parallax varies, as a step-like parallax caused by the "front-Parallel" effect, rather than a smooth parallax plane. The improved algorithm in recent years initializes the parallax of each point to a random plane through an over-parameterization method, and obtains a better matching effect on the inclined plane through the combined use of local optimization and global optimization. But the calculation amount is significantly increased compared with the conventional method.
The global optimization matching method aims at global energy minimization and generally adopts the following energy equation:
where E (D) represents the global energy level, D represents the "disparity map" of the entire image, q is the neighboring pixels of pixel p, and Np represents the collection of neighboring pixels of pixel p. Dp and Dq represent the parallax of these two pixels. C represents the data item of the pixel p when the parallax is Dp, the similarity degree of the pixels of the left and right slices, S represents the smooth item, and the penalty imposed by Dp and Dq is not the same.
In this embodiment, a message propagation method is used as a basic strategy of global optimization, a pyramid matching method is used, and the parallax obtained by matching the pyramid image of the previous layer is used as the parallax initial value of the current layer and the basic data for estimating the local parallax gradient. It is assumed that the disparity map D0 of the upper layer has been obtained. And obtaining the parallax initial value D of the current layer through linear interpolation.
As can be seen in fig. 11 (a) and (b), the matching efficiency Gao Chengdian of the system is compared with the SGM algorithm, and the DSM extracted by the system is compared with the DSM extracted by other methods in terms of topography, and it is obvious that the optimization method of the storage system is better.
Control point and connection point automatic matching:
besides supporting the traditional control point acquisition mode, the high-density control points can be automatically matched from the reference base map, and the high-density connection points can be automatically matched. As shown in fig. 15-17, a sufficient number of control points and connection points can be matched even where the cloud is very large.
Fig. 16 shows google bottom map matching to 18994 control points, while the same area of the present application can be matched to 31519 connection points. Is obviously superior to the prior art.
In addition, in the full-automatic orthographic image production process, the traditional adjustment calculation is generally called a stereoscopic image condition as a strong intersection condition, and the intersection of a plurality of images imaged by non-side-sway is a weak intersection condition. In order to acquire images of a specific area in actual operation of a satellite, the observation angles are often random, so that the sizes of intersection angles are different. The system can process the images of different sensors and any intersection angle in a joint way. In terms of precision, taking national land utilization survey data of a certain province as an example, the edge connecting precision is generally better than 0.5 pixels and is not more than 2 pixels at maximum. The experiment uses a resource No. three uncontrolled DOM as a reference base map, the total nesting precision of the DOM and the base map is better than 1 pixel, and the maximum is not more than 3 pixels.
The uncontrolled precision is improved by adding generalized control such as SRTM, GLAS and the like in the model splicing/free network adjustment, and the DSM is automatically generated by adopting a global optimization matching strategy based on message propagation, as shown in fig. 12, and the method comprises the following steps:
1) Stereoscopic image and RPC import: the front view, the rear view images and the corresponding RPCs are distinguished according to file names, and the RPCs are corrected by adjustment.
2) Image preprocessing and texture analysis: one key step in image preprocessing is to determine the minimum effective value of the image, and uniformly set DN values lower than the minimum effective value to the minimum effective value, so that smooth transition of the low-value areas can be realized, otherwise, false texture influence matching occurs. The satellite images are easy to generate large-area non-texture or pseudo-texture areas in areas with low gray scale, such as mountain area backlight areas with large fluctuation, the areas are detected through calculation based on local window variance, the matching cost of the areas is uniformly set to be a fixed value, and smooth transition can be realized through the following matching cost propagation steps.
3) Generating an image pyramid: an image pyramid is generated by wavelet transformation and stored in a diagonal region of the data volume. The key point here is whether the number of layers of the pyramid and the set top search range correspond to the height difference in the region.
4) CENSUS-based top-level matching: and (3) performing CENSUS transformation on the top-layer image and matching based on message propagation, detecting suspicious regions, if suspicious regions caused by large-area non-texture deletion occur, and the heights of the regions are close to the maximum or minimum heights, setting the searching range to be too small, enlarging the searching range and matching the error regions again, and if the error regions are no longer represented as large-area regions, reducing the top-layer matching nodes if the regions are reduced and the error regions are also provided with large-area error region bundles, and enlarging the searching range until suspicious regions caused by the searching range do not exist (the front-back matching results are not obviously different). The result of the matching is the elevation of each image point of the front image.
5) Layer-by-layer matching, suspicious region detection and refinement matching based on CENSUS and mutual information (Mutual Information, hereinafter abbreviated as MI): and adopting block matching, wherein two adjacent images have a certain overlapping area, performing CENSUS conversion on each layer of images below the top layer by layer, calculating front and rear vision correction images by using the elevation obtained by each front vision image point of the upper layer, and calculating an MI lookup table. The initial elevation of each point of the current layer is obtained by the matched elevation of the previous layer, and the searching range of each point is determined by the elevation in the neighborhood (the neighborhood size depends on the local elevation difference) of the point, and a certain margin is reserved. And carrying out message propagation optimization matching based on MI and CENSUS, and determining suspicious regions according to the setting of smooth parameters, the difference between multipath independent matching results and aggregated matching results and other methods. Classifying the mismatching areas according to the texture features of the front-view image and the front-view and rear-view correction image in the mismatching areas, the elevations around the mismatching areas, the matching measure of the mismatching areas and the like, performing refined matching on suspicious areas of different types, and calculating the images without using the suspicious areas by using the MI lookup table during the refined matching.
6) Generating a DSM: after bottom layer matching is completed, calculating ground points according to the main image points, the corresponding elevation of the image points and RPC, and generating DSM according to the set DSM grid distance by a distance reciprocal weighting method.
7) DSM generation and concatenation: and generating DSMs according to the set DSM grid intervals by a distance reciprocal weighting method, and splicing the single model DSMs into the whole DSMs. And recording a single model serial number corresponding to each grid point in the splicing process, and providing the single model serial number for a DOM extraction link.
In case of suspicious region matching, the following method may be selected:
1) Generating a epipolar stereoscopic image pair of the stereoscopic image, wherein the epipolar stereoscopic image pair comprises a left epipolar image and a right epipolar image;
and generating a epipolar line stereo image pair of the stereo image by adopting a projection track method. The epipolar stereo image pair for generating stereo images by projection trajectory method is a well-known technique in the art, and will not be described here.
2) Reading the epipolar line stereo image pair, and respectively performing wavelet transformation on a left epipolar line image and a right epipolar line image of the epipolar line stereo image pair to establish respective image pyramids of the left epipolar line image and the right epipolar line image, wherein the image pyramids comprise a plurality of layers of pyramid images, the left epipolar line image and the right epipolar line image which are originally read are first layers of pyramid images, and a second layer of pyramid images and pyramid images above the second layer of pyramid images which are obtained after wavelet transformation comprise an image area and three texture areas;
The image pyramid refers to a reduced image obtained by downsampling an image, that is, the image pyramid is a plurality of wavelet transformed images (and texture images) with lengths and widths decreasing by one half in sequence.
2) The wavelet transformation process in (2) specifically comprises the following steps: and performing wavelet transformation on the i-th layer pyramid image to obtain an i+1-th layer epipolar line image, wherein i is an integer greater than 1, the i+1-th layer epipolar line image comprises images of four areas of upper left, upper right, lower left and lower right, the image of the upper left area is an image area and comprises a reduced image with the length and width being half of those of the epipolar line image, and the three areas of upper right, lower left and lower right are texture areas and comprise texture information of the epipolar line image. The three texture areas after wavelet transformation have the same length and width as the image area.
After the second-layer pyramid image of the left epipolar line image and the right epipolar line image is obtained, the second-layer pyramid image is used as a new original image, and a third-layer pyramid image (comprising an image area and three texture areas) is obtained after wavelet transformation. And by analogy, the pyramid images of the fourth layer and the fifth layer can be obtained. According to a preferred embodiment of the invention, the number of layers of the pyramid image is 5.
The gray values of an image region and three texture regions after wavelet transformation are the result of weighted accumulation summation according to filter coefficients within a 4*4 window. Specifically, the gray G0 (j, k) of the kth line and j-th line of the wavelet-transformed image region, the gray G1 (W/2+j, k) of the kth line, W/2+j-th line, the gray G2 (j, H/2+k) of the j-th line, the H/2+k-th line, the H/2+k-th line, and the gray G3 (W/2+j, H/2+k) of the W/2+j-th line of the left-lower texture region are calculated according to the following expression.
3) For a left epipolar line image and a right epipolar line image of each layer of image pyramid below the top layer, determining suspicious areas in the left epipolar line image and the right epipolar line image layer by layer;
if the number of pyramid layers is n, the top layer is the n-1 th layer, and the following layers are n-2 layers, n-3 layers, the number of the pyramid layers is equal to the number of the pyramid layers, and the number of the pyramid layers is 0, and the number of the pyramid layers is also 0, namely the initial left epipolar line image layer and the initial right epipolar line image layer.
The parallax search range of each image point of the left and right epipolar images is set as h, and the search starting parallax is set as-h/2. Generally, left and right epipolar line image parallax of the top layer of the pyramid is obtained through an SGM matching method. The top layer does not detect suspicious regions.
4) Performing global optimization on the left epipolar line image-right epipolar line image matching cost and the right epipolar line image-left epipolar line image matching cost to acquire a plurality of pairs of parallax images based on different path combinations and corresponding smoothing models;
Only one of the plurality of candidate disparities within the corresponding search range for each image point P is correct, which requires determining the disparity for each image point by global optimization of step 3.4.
After the left-right matching cost and the right-left matching cost are calculated, parallax of image points can be obtained through two methods (a local algorithm and a global optimization algorithm), the global optimization algorithm has stronger robustness than the local algorithm, and cloud areas and cloud-free areas are more obvious in comparison in matching stability, so that global optimization is carried out on the left epipolar line image-right epipolar line image matching cost and the right epipolar line image-left epipolar line image matching cost.
And adopting an optimization strategy based on message propagation to globally optimize the left epipolar line image-right epipolar line image matching cost and the right epipolar line image-left epipolar line image matching cost. The 16 directional center points of message propagation represent the pixels receiving the message, and the surrounding 16 pixels represent the 16 closest pixels to propagate upward. Four message propagation path combinations are adopted for each image pair matching direction (two directions of left epipolar line image-right epipolar line image direction and right epipolar line image-left epipolar line image direction), and the four message propagation path combinations are respectively as follows:
path combination 1: 0. 12, 1, 13;
Path combination 2: 4. 14, 5, 15;
path combination 3: 2. 9, 3 and 8;
path combination 4: 6. 11, 7, 10.
A dual-parameter model is used for path combinations 1 and 2, but the sizes of parameters P1 and P2 are not consistent, and a Huber model is used for path combinations 3 and 4, but the sizes of specific parameters a are also not consistent.
The global optimization will then result in a matching cost L after each pixel optimization in a form similar to the global data item C, being an array of length equal to the pixel search range Dmax. The minimum number Dmin in the array (array subscripts 0,1, …, dmax-1) plus the minimum disparity Ds for that pixel, i.e., the disparity denod for that pixel. The pixel column coordinates col and the parallax Dend are added to obtain pixel column coordinates col' corresponding to the right epipolar line image, and the two pixel row coordinates row of the left and right epipolar line images are identical. The corresponding ground coordinates (longitude L, latitude B and elevation H) can be obtained by the front intersection of the image point coordinates of the left and right quasi-epipolar images and the respective RPCs of the left and right quasi-epipolar images.
5) And (3) preliminarily determining suspicious regions in the images based on the determined pairs of disparity maps obtained based on different path combinations and corresponding smoothing models in the step 4).
The SRTM auxiliary adjustment flow adopted by the system is shown in fig. 13, and comprises the following steps:
1) Connection point extraction based on fast view
The resolution of the obtained stereoscopic image (set as a resource III) is reduced in a mode of taking an average value through n x n pixel gray scale, the resolution of the resource III 01 star front view image is reduced by 8 times, the resolution of the front and back view images is reduced by 5 times, and the resolution of the 02 star three view images is reduced by 8 times. The corresponding RPC parameters LINE_OFF, LINE_SCALE, SAMP_OFF and SAMP_SCALE conversion methods are as follows:
LINE_SCALE＝LINE_SCALE/n，SAMP_SCALE＝SAMP_SCALE/n
LINE_OFF＝(LINE_OFF-n/2+0.5)/n，SAMP_OFF＝(SAMP_OFF-n/2+0.5)/n
firstly, extracting homonymy points of each quick view image pair (2 or 3 views) through SIFT matching, and extracting homonymy points between the front images of any two image pairs. And determining the point pairs which are common to both pairs by a coordinate comparison method, and combining the point pairs into a multi-view homonymy point.
And (3) circularly executing 'intersection in front of the same name point-obtaining image points and ground point pairs of each image-RPC correction of each image', wherein the first several (e.g. 5) iterations only correct the molecular constant term of the RPC, and the later iterations correct the molecular constant term and the first-order term corresponding to the normalized channel and dimension.
And calculating the offset with the maximum correlation between the object space elevation of the same name point and the corresponding elevation of the SRTM by using an equal step search method. The object coordinates of the same name point are added with the offset, and RPC of all images is corrected.
2) Converting the quick view RPC into the original resolution RPC combined with SRTM elevation constraint to extract the conventional connection point of the original image
And replacing the molecular constant term and the primary term corresponding to longitude and latitude of the rational polynomial parameter of the original image RPC with the corrected quick view RPC corresponding parameter. The relative accuracy of the RPC parameter is mainly limited by two factors, namely resolution of quick vision and accuracy of quick trying connection points, and the accuracy of the predicted initial position of the same name point of the original image is also limited by the deviation of the local elevation of the SRTM and the corresponding elevation of the same name point, wherein the deviation is obviously related to the topography. For co-track stereoscopic imaging, the image of the terrain is predominantly represented in the row direction of the image. Therefore, the search range in the row direction is set to a large value, the search range in the column direction is set to 5, the search range in the row direction is set to 9, and the matching window is set to 15×15. Firstly, extracting a harris characteristic point of a main image, obtaining a matching point with pixel-level precision through a correlation coefficient method, and obtaining a sub-pixel position through least square matching, wherein the least square matching only considers the deformation of the image in the row direction. And preliminarily eliminating the homonymous points which do not meet the conditions through the correlation coefficient threshold and reverse matching.
And (3) after the conventional connection point extraction is completed, performing free net adjustment by adopting a method similar to the step (1), wherein the difference is that after each iteration, RPC correction of each scene image is completed, and recording the largest points of the scene image residual difference larger than a certain threshold a. After all large residual points are ordered, the largest part of points (e.g., 20%) is deleted. The threshold a will progressively decrease as the number of iterations increases.
All RPCs are restored (looking at the RPC transform results quickly) and iterated again to steady (iterating a certain number of times).
3) SRTM feature point extraction
The mountain top characteristic points are extracted by adopting a local difference detection operator, flat characteristic points are detected according to a certain step distance, the SRTM row and column numbers corresponding to the current coordinates are assumed to be (r, c), points with the height difference between the (r, c) point and the 8 neighborhood direction are searched by taking the (r, c) point as the center, and if points with the height difference exceeding the limit are not found in 8 directions in a certain searching range, the points are flat characteristic points.
4) Dense connection point acquisition corresponding to SRTM feature points
The two types of characteristic points are projected onto an original image to obtain a main image and a secondary image with certain sizes, the SGM object space method is used for carrying out pixel-by-pixel dense matching to obtain homonymous image points, and the homonymous image points are thinned according to 45 meters by one point (about half of SRTM resolution).
Unlike mountain top connection points, flat dense connection points first filter out corresponding points of buildings and the like that may exist by morphological filtering, because SRTM does not contain such information.
5) Regional net adjustment with SRTM constraint
The two steps of conventional junction front intersection and each scene image RPC correction are alternately performed, and the current RPC is used for calculating the ground point clusters corresponding to the dense junction. And solving the maximum correlation point with the SRTM mountain peak to obtain the horizontal offset and the vertical offset. The flat region dense connection points are vertically offset only from the corresponding SRTM. The conventional connection point corresponds to the ground point plus horizontal and vertical offsets according to a reciprocal distance weighting method.
Meanwhile, the system adopts a method for post-processing of the digital surface model and information extraction, and the urban building area is automatically acquired through DSM. Firstly, identifying non-ground points by a space hierarchy decomposition and information statistics method. And (5) after the non-ground points are removed, using surrounding elevation fitting to obtain DTM (digital terrine model), subtracting the DTM height from the DSM height to obtain building area information, and removing noise through morphological transformation.
1) Setting initial parameters, including a window parameter initial value cr0=50, an effective point threshold initial value minmx0=800, an elevation hierarchical dividing parameter hoff=2.0 meters, and iteration times xhnum=3.
2) Identifying non-ground points through multiple iterations
The basic idea of identifying non-ground points is that in a space cube, ground points are majority, non-ground points are minority, possible non-ground points are separated by recording effective points in a hierarchical space, and non-ground point information is obtained through comprehensive analysis of gradient, neighborhood support and other information.
3) Non-ground point elevation fitting
And detecting effective points from each ineffective point to the 8 neighborhood direction, recording the corresponding detection distance, and fitting the elevation of the non-ground point by a distance reciprocal weighting method. The DSM rendering effect is shown in fig. 14.
Of course, the parameters in the above embodiments are merely examples, and in other embodiments, the parameters may be adjusted according to the specific situation.
In addition, in view of the characteristics of complex queries such as a large number of spatial data queries, combined data queries, associated data queries and the like contained in the data management, the image storage system adopts a mode of combining the spatial data queries and the database to realize quick retrieval of data.
The image storage system stores the space information and the metadata information to be queried in a database in a partition table mode. For data containing spatial information, the image storage system manages in a spatial data set manner, and a grid index is established through a spatial data engine. And managing the metadata information by adopting a database partition table, and establishing a database index. When the data retrieval operation is performed, the image storage system performs the retrieval operation by using different indexes based on different query conditions.
When data query is carried out, the image storage system comprehensively utilizes a multithreading query mechanism, a paging query mechanism, a real-time dynamic query mechanism and an index management mechanism, so that the query efficiency is improved, and the user experience is improved. As shown in fig. 10.
Meanwhile, the system also supports a data browsing function, namely browsing the queried image results, and supports image block diagram browsing and quick view browsing. And automatically positioning the map area to the image graphic area by checking a check box recorded in the query result.
And data archiving statistics are also provided, all archived data are classified and counted according to conditions, all data types are selected by default, and the data are displayed in a query result bar graph. The current inventory data can also be respectively classified and counted on line and off line.
And when the space of the disk is insufficient, providing a data migration function, migrating the data which is not used for the longest time out of the disk, transferring the data to an externally-hung disk, vacating the space of the corresponding disk, writing the record into a database, and then inquiring the specific storage position of the image in the system to find out the corresponding data if the data is needed.
The storage system adopts a background periodical automatic product manufacturing mode to realize the requirement of quick drawing of the coverage product. The image storage system can automatically cover the production of the product in the background based on time, data type, region information and cloud cover information defined by a user, and the system can store the product after the storage of the covered product is completed. When the user uses the system, only the data query downloading operation is needed.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
While the foregoing description of the embodiments of the present invention has been presented in conjunction with the drawings, it should be understood that it is not intended to limit the scope of the invention, but rather, it is intended to cover all modifications or variations within the scope of the invention as defined by the claims of the present invention.
Claims (10)
1. A remote sensing image data storage method is characterized in that: the method comprises the following steps:
acquiring remote sensing image data;
preprocessing and texture analysis are carried out on the three-dimensional remote sensing image data, an image pyramid is generated through wavelet transformation, top-layer images of the pyramid are transformed and matched based on message transmission, layer-by-layer matching is carried out, and a digital surface model is generated;
specifically, a message propagation method is used as a basic strategy of global optimization, a pyramid matching method is adopted, and the parallax obtained by matching the pyramid image of the upper layer is used as the parallax initial value of the current layer and the basic data for estimating the local parallax gradient;
1) Stereoscopic image and RPC importing;
2) Image preprocessing and texture analysis;
3) Generating an image pyramid;
4) CENSUS-based top-level matching: performing CENSUS transformation on the top-layer image and performing matching based on message propagation, detecting suspicious regions, if suspicious regions caused by large-area non-texture deletion occur, and the heights of the regions are close to the maximum or minimum heights, setting the searching range to be too small, enlarging the searching range and matching the error regions again, if the error regions are no longer represented as large-area regions, reducing the top-layer matching nodes if the regions are reduced and there are large-area error region bundles, and enlarging the searching range until suspicious regions caused by the searching range do not exist, namely, the front-back matching results have no obvious difference; the matching result is the elevation of each image point of the front view image;
5) Based on CENSUS and mutual information Mutual Information, layer-by-layer matching of MI, hereinafter referred to as MI, suspicious region detection and refinement matching: adopting block matching, wherein two adjacent images have a certain overlapping area, performing CENSUS conversion on each layer of images below the top layer by layer, calculating front and rear vision correction images by using the elevation obtained by each front vision image point of the upper layer, and calculating an MI lookup table; the initial elevation of each point of the current layer is obtained by the matched elevation of the previous layer, the search range of each point is determined by the neighborhood of the point, the size of the neighborhood depends on the local elevation difference, and a certain margin is reserved; performing MI and CENSUS based message propagation optimization matching, and determining suspicious regions according to the setting of smooth parameters and the difference between multipath independent matching results and aggregated matching results; classifying the mismatching areas according to the texture features of the front-view image and the front-view and rear-view correction image in the mismatching areas, the elevations around the mismatching areas and the matching measures of the mismatching areas, carrying out refined matching on suspicious areas of different types, and calculating images without using suspicious areas by using an MI lookup table during the refined matching;
6) Generating a DSM;
7) DSM generation and splicing;
when suspicious regions are matched, the following method is selected:
1) Generating a epipolar stereoscopic image pair of the stereoscopic image, wherein the epipolar stereoscopic image pair comprises a left epipolar image and a right epipolar image;
generating a epipolar line stereo image pair of the stereo image by adopting a projection track method;
2) Reading the epipolar line stereo image pair, and respectively performing wavelet transformation on a left epipolar line image and a right epipolar line image of the epipolar line stereo image pair to establish respective image pyramids of the left epipolar line image and the right epipolar line image, wherein the image pyramids comprise a plurality of layers of pyramid images, the left epipolar line image and the right epipolar line image which are originally read are first layers of pyramid images, and a second layer of pyramid images and pyramid images above the second layer of pyramid images which are obtained after wavelet transformation comprise an image area and three texture areas;
the image pyramid refers to a reduced image obtained by downsampling an image, namely, the image pyramid is a plurality of wavelet transformation images and texture images, the length and the width of which are sequentially reduced by one half;
2) The wavelet transformation process in (2) specifically comprises the following steps: taking an ith layer pyramid image as a prokaryotic line image, carrying out wavelet transformation on the ith layer pyramid image to obtain an ith layer epipolar line image, wherein i is an integer larger than 1, the ith layer epipolar line image comprises images of four areas of upper left, upper right, lower left and lower right, the image of the upper left area is an image area and comprises a reduced image with the length and width being half of those of the prokaryotic line image, and the three areas of upper right, lower left and lower right are texture areas and comprise texture information of the prokaryotic line image; the three texture areas after wavelet transformation have the same length and width as the image area;
After a second-layer pyramid image of the left epipolar line image and the right epipolar line image is obtained, taking the second-layer pyramid image as a new original image, and carrying out wavelet transformation on the new original image to obtain a third-layer pyramid image; comprises an image area and three texture areas; and by analogy, the fourth and fifth pyramid images can be obtained; the number of layers of the pyramid image is 5;
the gray values of an image area and three texture areas after wavelet transformation are the result of weighted accumulation summation according to the filter coefficients in a 4*4 window; specifically, the gray scales G0 j, k of the kth row and j column, the kth row and k of the upper right texture region, the gray scales G1W/2+j, k of the W/2+j column, the H/2+k row and j column of the lower left texture region, the gray scales G2 j, H/2+k of the j column, the H/2+k row and the gray scales G3W/2+j, H/2+k of the W/2+j column of the image region after wavelet transformation are calculated according to the following formula;
3) For a left epipolar line image and a right epipolar line image of each layer of image pyramid below the top layer, determining suspicious areas in the left epipolar line image and the right epipolar line image layer by layer;
if the number of pyramid layers is n, the top layer is an n-1 layer, and the following layers are n-2 layers, n-3 layers, the number of the pyramid layers is equal to the number of the pyramid layers, and the number of the pyramid layers is 0, wherein the number of the pyramid layers is also 0, namely an initial left epipolar line image layer and an initial right epipolar line image layer;
Setting the parallax search range of each image point of the left epipolar line image and the right epipolar line image as h, and setting the search initial parallax as-h/2; normally, obtaining left and right epipolar line image parallax of the top layer of the pyramid by an SGM matching method; the top layer does not detect suspicious areas;
4) Performing global optimization on the left epipolar line image-right epipolar line image matching cost and the right epipolar line image-left epipolar line image matching cost to acquire a plurality of pairs of parallax images based on different path combinations and corresponding smoothing models;
only one of the plurality of candidate disparities within the corresponding search range for each image point P is correct, which requires determining the disparities for each image point by global optimization;
after the left-right matching cost and the right-left matching cost are calculated, two methods are adopted: the local algorithm and the global optimization algorithm acquire the parallax of the image points, and the global optimization algorithm has stronger robustness than the local algorithm, and the cloud area and the cloud-free area are more obviously compared in the matching stability, so that the left epipolar line image-right epipolar line image matching cost and the right epipolar line image-left epipolar line image matching cost are subjected to global optimization;
adopting an optimization strategy based on message propagation to globally optimize the left epipolar line image-right epipolar line image matching cost and the right epipolar line image-left epipolar line image matching cost; the 16 directional center points of message propagation represent the pixels receiving the message, and the surrounding 16 pixels represent 16 pixels propagating to the nearest upward; matching directions for each pair: the two directions of the left epipolar line image-right epipolar line image direction and the right epipolar line image-left epipolar line image direction adopt four message propagation path combinations, which are respectively:
Path combination 1: 0. 12, 1, 13;
path combination 2: 4. 14, 5, 15;
path combination 3: 2. 9, 3 and 8;
path combination 4: 6. 11, 7, 10;
a double-parameter model is adopted for the path combinations 1 and 2, but the sizes of the parameters P1 and P2 are inconsistent, a Huber model is adopted for the path combinations 3 and 4, but the sizes of the specific parameters a are also inconsistent;
the form of the matching cost L after the optimization of each image point is similar to the form of the global data item C and is an array with the length equal to the image point searching range Dmax; the minimum value of the sequence Dmin in the array, the array subscripts 0,1, …, dmax-1, and the minimum parallax Ds of the image point, namely the parallax Dend of the image point; adding the pixel column coordinates col and the parallax Dend to obtain pixel column coordinates col' corresponding to the right epipolar line image, wherein the row coordinates row of the two pixels of the left and right quasi-epipolar line images are identical; the corresponding ground coordinates are obtained by front intersection of the image point coordinates of the left and right quasi-epipolar line images and the respective RPCs of the left and right quasi-epipolar line images: longitude L, latitude B, and elevation H;
5) Preliminarily determining suspicious regions in the images based on the multiple pairs of disparity maps which are determined by the step 4) and are acquired based on different path combinations and corresponding smoothing models;
carrying out radiation correction, differential correction and mosaic on the three-dimensional remote sensing image data subjected to scanning treatment by utilizing a digital elevation model, and cutting according to a set picture range to generate a digital orthographic image;
According to the remote sensing image data acquisition time, storing the data in a set time period in a local storage area, storing the data outside the set time period in an external storage area, wherein each area comprises a plurality of storage nodes, and respectively transmitting the original remote sensing image data and the processing result generated based on the original data to different storage nodes for storage.
2. The remote sensing image data storage method as claimed in claim 1, wherein: the method comprises the steps of carrying out terrain surveying and mapping auxiliary adjustment, reducing the resolution of a stereoscopic image, extracting the connection point of each fast view image pair, restoring the resolution of the fast view, extracting the conventional connection point of an original image by combining the altitude constraint of the terrain surveying and mapping of a spaceflight radar, carrying out free network evaluation of the original influence, extracting mountain top characteristic points and flat area characteristic points according to a certain interval, respectively projecting the two types of characteristic points onto the original influence, and extracting dense connection points.
3. The remote sensing image data storage method as claimed in claim 1, wherein: the method also comprises a coverage map product making module which is used for carrying out concurrent map tile drawing based on the time, the data type, the area information and the cloud amount information of the stereoscopic image data input to generate a coverage map product.
4. The remote sensing image data storage method as claimed in claim 1, wherein: and transmitting the data of different remote sensing images to different processing nodes, and processing the data in parallel among the processing nodes.
5. The method for storing remote sensing image data according to claim 4, wherein: before the remote sensing image data processing process is carried out, determining the maximum parallel number and computing capacity of each processing node, and determining N processing nodes with the top ranking computing capacity to execute data processing tasks according to the maximum parallel number N.
6. The method for storing remote sensing image data according to claim 4, wherein: grouping remote sensing image data processing tasks, ranking according to the computing capacity of each processing node, and matching and distributing the tasks and the processing nodes by utilizing an optimal algorithm.
7. The remote sensing image data storage method as claimed in claim 1, wherein: before data storage, checking the data integrity, and if the data integrity meets the requirement, storing.
8. A remote sensing image data storage system implemented by the remote sensing image data storage method as claimed in claim 1, characterized in that: comprising the following steps:
The data receiving module is configured to acquire remote sensing image data;
the digital surface model generation module is configured to perform preprocessing and texture analysis on the three-dimensional remote sensing image data, generate an image pyramid through wavelet transformation, transform a top layer image of the pyramid, perform matching based on message propagation, perform layer-by-layer matching and generate a digital surface model;
the digital orthophoto map generating module is configured to utilize a digital elevation model to carry out radiation correction, differential correction and mosaic on the three-dimensional remote sensing image data subjected to scanning treatment pixel by pixel, and cut according to a set map range to generate a digital orthophoto map;
the data ordering module is configured to store the data of the set time period in the local storage area according to the remote sensing image data acquisition time, and store the data outside the set time period to the external storage area;
and the data storage module is configured to respectively transmit the original remote sensing image data and the processing result generated based on the original data to different storage nodes for storage.
9. A computer-readable storage medium, characterized by: in which a plurality of instructions are stored, said instructions being adapted to be loaded by a processor of a terminal device and to carry out the steps of a remote sensing image data storage method according to any one of claims 1-7.
10. A terminal device, characterized by: comprising a processor and a computer-readable storage medium, the processor configured to implement instructions; a computer readable storage medium for storing a plurality of instructions adapted to be loaded by a processor and to perform the steps of a remote sensing image data storage method as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011551192.7A CN112561832B (en) | 2020-12-24 | 2020-12-24 | Remote sensing image data storage method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011551192.7A CN112561832B (en) | 2020-12-24 | 2020-12-24 | Remote sensing image data storage method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112561832A CN112561832A (en) | 2021-03-26 |
| CN112561832B true CN112561832B (en) | 2023-12-01 |
Family
ID=75033483
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011551192.7A Active CN112561832B (en) | 2020-12-24 | 2020-12-24 | Remote sensing image data storage method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112561832B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113066181A (en) * | 2021-04-08 | 2021-07-02 | 中铁十八局集团有限公司 | Terrain simulation method based on satellite images and digital elevation data |
| CN113836095A (en) * | 2021-09-26 | 2021-12-24 | 广州极飞科技股份有限公司 | Point cloud data storage method and device, storage medium and electronic equipment |
| CN114241125B (en) * | 2021-11-30 | 2022-06-21 | 感知天下(北京)信息科技有限公司 | Multi-view satellite image-based fine three-dimensional modeling method and system |
| CN116303809A (en) * | 2022-11-29 | 2023-06-23 | 自然资源部国土卫星遥感应用中心 | Satellite image data management method and management system |
| CN116091881B (en) * | 2023-02-14 | 2023-07-11 | 安徽星太宇科技有限公司 | Remote sensing information management system based on multisource data fusion |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104537125A (en) * | 2015-01-28 | 2015-04-22 | 中国人民解放军国防科学技术大学 | Remote-sensing image pyramid parallel building method based on message passing interface |
| CN105045856A (en) * | 2015-07-09 | 2015-11-11 | 中国资源卫星应用中心 | Hadoop-based data processing system for big-data remote sensing satellite |
| CN105303535A (en) * | 2015-11-15 | 2016-02-03 | 中国人民解放军空军航空大学 | Global subdivision pyramid model based on wavelet transformation |
| CN108846436A (en) * | 2018-06-13 | 2018-11-20 | 武汉朗视软件有限公司 | A kind of more view stereoscopic matching process |
| CN109427043A (en) * | 2017-08-25 | 2019-03-05 | 国家测绘地理信息局卫星测绘应用中心 | A kind of matched smooth item calculation method of parameters of stereopsis global optimization and equipment |
| CN110046211A (en) * | 2019-04-16 | 2019-07-23 | 重庆市地理信息中心 | Surveying and mapping result catalogue based on semantic integrity automaticly inspects and storage dissemination method |
| CN110763205A (en) * | 2019-11-05 | 2020-02-07 | 新疆维吾尔自治区测绘科学研究院 | Method for generating orthophoto map of border narrow and long area by digital photogrammetric system |
| CN111723221A (en) * | 2020-06-19 | 2020-09-29 | 珠江水利委员会珠江水利科学研究院 | Mass remote sensing data processing method and system based on distributed architecture |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109783665B (en) * | 2018-12-29 | 2022-10-14 | 武汉大学 | Design method for realizing Hbase database remote sensing big data storage model based on Google S2 |
-
2020
- 2020-12-24 CN CN202011551192.7A patent/CN112561832B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104537125A (en) * | 2015-01-28 | 2015-04-22 | 中国人民解放军国防科学技术大学 | Remote-sensing image pyramid parallel building method based on message passing interface |
| CN105045856A (en) * | 2015-07-09 | 2015-11-11 | 中国资源卫星应用中心 | Hadoop-based data processing system for big-data remote sensing satellite |
| CN105303535A (en) * | 2015-11-15 | 2016-02-03 | 中国人民解放军空军航空大学 | Global subdivision pyramid model based on wavelet transformation |
| CN109427043A (en) * | 2017-08-25 | 2019-03-05 | 国家测绘地理信息局卫星测绘应用中心 | A kind of matched smooth item calculation method of parameters of stereopsis global optimization and equipment |
| CN108846436A (en) * | 2018-06-13 | 2018-11-20 | 武汉朗视软件有限公司 | A kind of more view stereoscopic matching process |
| CN110046211A (en) * | 2019-04-16 | 2019-07-23 | 重庆市地理信息中心 | Surveying and mapping result catalogue based on semantic integrity automaticly inspects and storage dissemination method |
| CN110763205A (en) * | 2019-11-05 | 2020-02-07 | 新疆维吾尔自治区测绘科学研究院 | Method for generating orthophoto map of border narrow and long area by digital photogrammetric system |
| CN111723221A (en) * | 2020-06-19 | 2020-09-29 | 珠江水利委员会珠江水利科学研究院 | Mass remote sensing data processing method and system based on distributed architecture |
Non-Patent Citations (2)
| Title |
|---|
| 基于半全局优化的资源三号卫星影像DSM提取方法;岳庆兴等;《武汉大学学报(信息科学版)》;20161005(第10期);全文 * |
| 高分辨率遥感影像DSM的改进半全局匹配生成方法;杨幸彬等;《测绘学报》;20181015(第10期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112561832A (en) | 2021-03-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112561832B (en) | Remote sensing image data storage method and system | |
| CN112559534B (en) | Remote sensing image data filing management system and method | |
| CN112765095B (en) | Method and system for filing image data of stereo mapping satellite | |
| Rupnik et al. | MicMac–a free, open-source solution for photogrammetry | |
| US11222465B2 (en) | Embedded urban design scene emulation method and system | |
| US11682170B2 (en) | Generating three-dimensional geo-registered maps from image data | |
| CN110659369B (en) | On-orbit high-precision lightweight global image control point database construction method and system | |
| CN112113544B (en) | Remote sensing data processing method and system based on unmanned aerial vehicle image | |
| CN109102563A (en) | A kind of outdoor scene three-dimensional modeling method | |
| CN110660125B (en) | Three-dimensional modeling device for power distribution network system | |
| Voulodimos et al. | Four-dimensional reconstruction of cultural heritage sites based on photogrammetry and clustering | |
| CN112100297A (en) | Three-dimensional visualization method for data management in urban updating | |
| CN117315146B (en) | Reconstruction method and storage method of three-dimensional model based on trans-scale multi-source data | |
| CN113808269A (en) | Map generation method, positioning method, system and computer readable storage medium | |
| CN113268533A (en) | Method for making two-label data in two-label four-real data platform | |
| Maltezos et al. | LIDAR vs dense image matching point clouds in complex urban scenes | |
| Collins et al. | Scalable construction monitoring for an as-performed progress documentation across time | |
| Lou et al. | Cave feature extraction and classification from rockery point clouds acquired with handheld laser scanners | |
| CN113282695B (en) | Vector geographic information acquisition method and device based on remote sensing image | |
| Brenner et al. | Continuous generalization for small mobile displays | |
| Ahmed et al. | High-quality building information models (BIMs) using geospatial datasets | |
| CN112561981A (en) | Photogrammetry point cloud filtering method fusing image information | |
| Yang et al. | Three-dimensional panoramic terrain reconstruction from aerial imagery | |
| Wu et al. | Building Facade Reconstruction Using Crowd-Sourced Photos and Two-Dimensional Maps | |
| Wei et al. | Indoor and outdoor multi-source 3D data fusion method for ancient buildings |
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 |