WO2020206914A1 - 多源对地观测图像处理的可视化方法、装置及存储介质 - Google Patents

多源对地观测图像处理的可视化方法、装置及存储介质 Download PDF

Info

Publication number
WO2020206914A1
WO2020206914A1 PCT/CN2019/103584 CN2019103584W WO2020206914A1 WO 2020206914 A1 WO2020206914 A1 WO 2020206914A1 CN 2019103584 W CN2019103584 W CN 2019103584W WO 2020206914 A1 WO2020206914 A1 WO 2020206914A1
Authority
WO
WIPO (PCT)
Prior art keywords
processing
observation image
visualization
source ground
ground observation
Prior art date
Application number
PCT/CN2019/103584
Other languages
English (en)
French (fr)
Inventor
王俊
高鹏
谢国彤
Original Assignee
平安科技（深圳）有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 平安科技（深圳）有限公司 filed Critical 平安科技（深圳）有限公司
Priority to JP2021506308A priority Critical patent/JP7264989B2/ja
Publication of WO2020206914A1 publication Critical patent/WO2020206914A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/54—Browsing; Visualisation therefor

Definitions

  • This application relates to the field of big data technology, and in particular to a visualization method, device and storage medium for processing multi-source earth observation images.
  • VISUAL MODELING is a way of thinking about problems by organizing models around realistic ideas. It describes the process of the developed system in a graphical way. Visual modeling allows to ask the necessary details of a complex problem and filter unnecessary details. It also provides a mechanism to observe the developed system from a different perspective.
  • common visual modeling software includes Unified Modeling Language (UML), VISIO, Simulink, Model Maker, Model Builder, etc.
  • UML Unified Modeling Language
  • VISIO Simulink
  • Model Maker Model Maker
  • Model Builder etc.
  • the existing visualization platform can provide graphical modeling user interface through visual modeling software
  • the existing visualization platform cannot perform WYSIWYG on the data, algorithms, processes and hidden information flows involved in the processing and mining of earth observation data (including satellite images, structured data, etc.)
  • Visual processing integration is difficult to achieve serial or parallel processing of algorithms in the observation data processing process, and it is difficult to meet the needs of visual processing debugging and improving development efficiency.
  • the present application provides a visualization method, device, and storage medium for multi-source ground observation image processing, so as to solve the problem that the prior art is difficult to meet the problem of visualization processing integration in the process of ground observation data processing and mining.
  • one aspect of this application is to provide a visualization method for multi-source Earth observation image processing, including:
  • another aspect of the present application is to provide an electronic device, the electronic device comprising: a processor and a memory, the memory includes a multi-source ground observation image processing visualization program, the multi-source ground When the visualization program of observation image processing is executed by the processor, the steps of the above-mentioned multi-source ground observation image processing visualization method are realized.
  • another aspect of the present application is to provide a computer-readable storage medium, the computer-readable storage medium includes a multi-source earth observation image processing visualization program, the multi-source earth observation image processing
  • the visualization program is executed by the processor, the steps of the visualization method for multi-source ground observation image processing as described above are realized.
  • This application realizes the visualization of the processing process of the multi-source ground observation image through the visualization model generated according to the processing flow of the multi-source ground observation image, and intuitively represents the various steps of the image processing through the graphical object model in the visualization model.
  • the visibility of data processing is improved, and the need for visual integration of multi-source ground observation image processing is met in a WYSIWYG manner.
  • This application improves the editability of each sub-algorithm in the image processing process by associating the sub-algorithm with the graphic object, enhances the flexibility of each graphic object model combination in the visualization model, improves the development and debugging efficiency, and realizes the The design visualization, intermediate call visualization, and processing result visualization of processing research effectively meet the needs of users for high-level development and application.
  • FIG. 1 is a schematic flowchart of a visualization method for image processing of multi-source ground observations according to this application;
  • Fig. 2 is a schematic diagram of modules of a visualization program for image processing of multi-source earth observation in this application.
  • FIG. 1 is a schematic flow chart of the visualization method of multi-source ground observation image processing according to this application. As shown in FIG. 1, the visualization method of multi-source ground observation image processing according to this application includes the following steps:
  • Step S1 Acquire multiple sub-algorithms corresponding to the processing flow according to the processing flow of the multi-source ground observation image.
  • the processing flow of multi-source ground observation images can be generalized into a sequential instruction set formed by a series of modular continuous different image processing sets.
  • the processing flow of remote sensing images includes data input and output, image preprocessing (including geometric correction, fusion, mosaic, etc.), image information extraction (including manual interpretation, automatic classification, feature extraction, dynamic detection, etc.) , Thematic mapping/three-dimensional visualization analysis (including integrated geographic information system existing data, etc.) and results report (including geographic information system analysis and sharing, etc.) and other processing steps.
  • Each processing step in the processing flow corresponds to one or more sub-algorithms, and the corresponding processing and information extraction of remote sensing images are realized through the combination of sub-algorithms.
  • the sub-algorithm can be obtained through query in the open source image processing algorithm library.
  • Step S2 According to the processing flow, a graphic object corresponding to each sub-algorithm is generated, each graphic object represents a sub-algorithm, and the graphic objects can use the same or different shapes.
  • the Graphics View framework in the cross-platform C++ graphical user interface application framework (Qt) is used as the graphical user interface for generating graphical objects.
  • the Graphics View framework includes Diagram Scene, Diagram View, and Diagram Item.
  • Diagram Scene is the visualization work area
  • Diagram Item is a two-dimensional graphic frame that can be placed in the work area
  • Diagram View is used to complete the display of the contents in the Diagram Scene
  • the generated graphic objects are the two-dimensional graphics boxes added in the workspace, and the shape of the two-dimensional graphics boxes can be adjusted.
  • the space partitioning tree (Binary Space Partitioning tree, BSP tree) stores the generated graphic objects.
  • Step S3 Associate the sub-algorithm with the graphic object through a model generator to generate a graphic object model.
  • a model generator to generate a graphic object model.
  • the graphical object model can be run, edited or saved in the model library. It can be further edited by integrating different processing module sets. Among them, different processing modules include different sub-algorithms, which provide great value in research and testing and subsequent use. The convenience.
  • the graphical object model can also be printed out as a flowchart or displayed and explained in a research report.
  • the model generator provides a variety of operation operators such as data type conversion, image space and time domain basic processing, image transformation, projection calibration, feature extraction, and change monitoring. Raster data, vector data, and Sort data and other operations.
  • the visualization method of multi-source ground observation image processing is implemented based on the Qt cross-platform graphical interface, and the spatial graphical object model is generated by using a panel tool.
  • Step S4 Connect a plurality of graphical object models according to the processing flow to generate a visualization model of the multi-source ground observation image.
  • a visualization model is composed of a series of graphic object models.
  • Each graphic object model generated in step S3 is an independent individual and a spatial model element.
  • the connection between the various graphic object models is also orderly, and the connection is made according to the processing flow of the multi-source ground observation image .
  • the visualization model runs the sub-algorithms in each graphic object model in turn according to the processing flow, so as to complete the operation function of spatial geographic information and image processing through each graphic object model.
  • the connection can be realized by setting a connection line with arrows between each graphical object model, and the arrow indicates the direction of the data flow.
  • Step S5 Visualize the processing process of the multi-source ground observation image through the visualization model.
  • This application realizes the visualization of the processing process of the multi-source ground observation image through the visualization model generated according to the processing flow of the multi-source ground observation image, and intuitively represents the various steps of the image processing through the graphical object model in the visualization model.
  • the visibility of data processing is improved, and the need for visual integration of multi-source ground observation image processing is met in a WYSIWYG manner.
  • This application improves the editability of each sub-algorithm in the image processing process by associating the sub-algorithm with the graphic object, enhances the flexibility of each graphic object model combination in the visualization model, improves the development and debugging efficiency, and realizes the The design visualization, intermediate call visualization, and processing result visualization of processing research effectively meet the needs of users for high-level development and application.
  • the step of acquiring multiple sub-algorithms corresponding to the processing flow according to the processing flow of the multi-source ground observation image includes: determining the processing flow according to the processing purpose of the multi-source ground observation image And each sub-step in the processing flow; query the algorithm knowledge base according to the sub-steps and obtain the sub-algorithms corresponding to the sub-steps, where the clear processing purpose includes image preprocessing, image feature extraction, change monitoring, etc., The determined processing flow is the specific way to achieve the processing purpose and the required data support.
  • the processing flow includes: preprocessing the remote sensing image, performing edge detection on the preprocessed remote sensing image, extracting straight line information from the detected edge information, and extracting the rectangular building structure based on the straight line information to obtain the remote sensing image
  • preprocessing of remote sensing images, edge detection, extraction of straight line information, and extraction of rectangular structure in the processing flow are respectively a sub-step, and the sub-algorithm corresponding to the corresponding sub-step is queried through the query algorithm knowledge base. In order to realize the corresponding processing function through the sub-algorithm.
  • the step of associating the sub-algorithm with the graphic object includes: determining the filter type of the graphic object according to the sub-algorithm; setting the attribute parameter of the graphic object according to the filter type, and determining A filter corresponding to the graphic object; the sub-algorithm is associated with the graphic object through the filter.
  • the filter refers to the processing object that performs operations on the data, which may cause data changes or generate new data, and the independent operations of the sub-algorithms are encapsulated in the filter through the packaging technology to realize the functions of the sub-algorithms.
  • the filter types include one or more of basic filters such as image denoising, image transformation, image analysis, image segmentation, image compression, image enhancement, image blurring, and image registration or composite filters of image processing.
  • the step of connecting a plurality of graphical object models according to the processing flow includes: constructing a filter connection channel according to the processing flow; connecting the filters in the graphical object model according to the filter connection channel, so that Each filter is connected according to the processing flow of the multi-source ground observation image to facilitate the processing of the multi-source ground observation image.
  • the filter connection channel connects the filters that perform various processing steps into a process that performs a series of processing.
  • OTB remote sensor image processing library
  • each processing step of the processing flow is constructed to form a filter connection channel, then the filter connection channel includes input filter, The first processing filter, the second processing filter, and the output filter.
  • the filter connection channel includes input filter, The first processing filter, the second processing filter, and the output filter.
  • the method further includes: setting the parameters of the filter and setting the parameters of the sub-algorithm.
  • the method further includes: modifying the parameters of the filter and modifying the parameters of the sub-algorithm.
  • the visualization method can realize the visualization of the simple processing process of the multi-source ground observation image, and can also realize the visualization of the complex processing process of the multi-source ground observation image.
  • the visualization model includes an input graphic object model, a processing graphic object model, and an output graphic object model, so as to realize the visualization of a simple image processing process.
  • the visualization model includes one or more processing graphic object models, and each processing graphic object model is connected to one or more input graphic object models and one or more output graphic object models to realize the complex processing process Visualization, for example, can be used to integrate and analyze the feature information of multiple multi-source ground observation images, and output multiple analysis reports from different angles.
  • the step of visualizing the processing process of the multi-source ground observation image through the visualization model includes: inputting the multi-source ground observation image into a corresponding graphical object model according to the processing flow; running the visualization model ; View the corresponding processing steps through the graphical object model, and output the processing results of the multi-source ground observation image.
  • the method further includes:
  • the processing requirements correspond to the above-mentioned processing purposes.
  • the processing requirements include: All the building target features in the processed remote sensing image are extracted, and the extracted building target is accurate.
  • the parameters of one or more graphical object models in the visualization model are modified (for example, the combination of filter types in the visualization model can be modified), and if the processing requirements are met, the processing results and The visualization model.
  • the visualization method for multi-source ground observation image processing further includes: before acquiring multiple sub-algorithms corresponding to the processing flow according to the processing flow of the multi-source ground observation image, acquiring the read instruction of the visualization model; determining whether to store There is a visualization model corresponding to the read instruction; when the visualization model corresponding to the read instruction is stored, the stored visualization model is read, and the step of generating the visualization model is no longer performed; the visualization model is visualized through the read visualization model Multi-source ground observation image processing process.
  • the visualization model corresponding to the read instruction When the visualization model corresponding to the read instruction is not stored, the corresponding multiple sub-algorithms are obtained according to the processing flow, and the graphic object model and the visualization model are generated, and the processing process of the multi-source ground observation image is visualized through the generated visualization model .
  • reading the stored visualization model After reading the stored visualization model, it also includes: correcting the read visualization model according to the processing flow of the multi-source ground observation image, and visualizing the processing process of the multi-source ground observation image through the modified visualization model, thereby realizing multiple Visualization of source-to-earth observation image processing.
  • Reading the stored visual model file will automatically execute the process of adding graphical arrows and defining each attribute, which is a process of reusing the visual model of image processing.
  • the step of reading the stored visualization model includes two steps: reading the shape and reading the arrow, wherein the shape refers to the shape of the graphical object model in the visualization model, and the arrow refers to the visualization Connecting arrows between multiple image object models in the model.
  • the step of reading the shape includes: obtaining the position coordinates of the graphic frame of the graphic object and the corresponding filter type; establishing the graphic frame under the corresponding position coordinates in the graphic scene section, and setting the filter type; using "attribute-value"
  • the attribute value is read out in the form of; according to the attribute value, the filter algorithm function of the set attribute interface defined in the algorithm knowledge base is sequentially invoked, and the corresponding attribute item is set to the corresponding attribute value.
  • the steps of reading the arrow include: obtaining the coordinates of the starting point of the arrow; searching for the shape of the position of the arrow according to the coordinates of the starting point of the arrow, and then establishing an arrow connecting the starting point in the same way as the process of hand animation arrow.
  • the graphic object model stored in the model library can be read and edited, and further editing and development can be done by integrating different graphic object models.
  • the visual model thus formed can encapsulate the verified processing flow by recording the input and output interfaces between a series of processing procedures, and the parameter settings of different processing procedures or filter operators.
  • the visual model can be used for yourself or Cited by others, with the continuous formation of modules that realize different processing procedures, the module group is further integrated to form a processing process with a certain scale to complete complex image processing functions while removing the time cost and labor cost of human-computer interaction.
  • the corresponding adjustable parameter items are displayed in the attribute box, and the value of the parameter item can be set by clicking in the displayed parameter item grid to adjust the required parameters for each OTB filter;
  • the visualization method of this application is suitable for the processing and visualization analysis of dozens of large data such as satellite images and spatial geographic information.
  • the multi-source ground observation image processing visualization method described in this application is applied to an electronic device, and the electronic device may be a terminal device such as a television, a smart phone, a tablet computer, and a computer.
  • the electronic device includes: a processor; a memory for storing a visualization program for multi-source ground observation image processing, and the processor executes the multi-source ground observation image processing visualization program to realize the following multi-source ground observation image Steps of the processing visualization method:
  • the electronic device also includes a network interface, a communication bus, and the like.
  • the network interface may include a standard wired interface and a wireless interface
  • the communication bus is used to realize the connection and communication between various components.
  • the memory includes at least one type of readable storage medium, which can be a non-volatile storage medium such as a flash memory, a hard disk, an optical disk, or a plug-in hard disk, etc., and is not limited to this, and can be stored in a non-transitory manner Any device that provides instructions or software and any associated data files to the processor so that the processor can execute the instructions or software program.
  • the software program stored in the memory includes a visualization program for multi-source ground observation image processing, and can provide the processor with the multi-source ground observation image processing visualization program, so that the processor can execute the multi-source ground observation image processing visualization program.
  • Observation image processing visualization program realize the steps of multi-source earth observation image processing visualization method.
  • the processor may be a central processing unit, a microprocessor, or other data processing chips, etc., and may run a program stored in the memory, for example, a visualization program for processing multi-source ground observation images in this application.
  • the electronic device may also include a display, and the display may also be called a display screen or a display unit.
  • the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an organic light-emitting diode (OLED) touch device, and the like.
  • the display is used to display the information processed in the electronic device and to display the visual work interface.
  • the electronic device may also include a user interface, and the user interface may include an input unit (such as a keyboard), a voice output device (such as a stereo, earphone), and the like.
  • the user interface may include an input unit (such as a keyboard), a voice output device (such as a stereo, earphone), and the like.
  • the visualization program for multi-source ground observation image processing can also be divided into one or more modules, and the one or more modules are stored in the memory and executed by the processor to complete the application.
  • the module referred to in this application refers to a series of computer program instruction segments that can complete specific functions.
  • Fig. 2 is a schematic diagram of modules of a visualization program for multi-source ground observation image processing in this application. As shown in Fig. 2, the visualization program for multi-source ground observation image processing can be divided into: acquisition module 1, graphic object generation Module 2, Association Module 3, Model Generation Module 4 and Visualization Module 5. The functions or operation steps implemented by the above modules are all similar to the above, and will not be described in detail here. For example, for example:
  • the acquisition module 1 acquires multiple sub-algorithms corresponding to the processing flow according to the processing flow of the multi-source ground observation image;
  • the graphic object generating module 2 generates graphic objects corresponding to each sub-algorithm according to the processing flow;
  • the association module 3 associates the sub-algorithm with the graphic object through a model generator to generate a graphic object model
  • the model generation module 4 connects multiple graphical object models according to the processing flow to generate a visualization model of the multi-source ground observation image
  • the visualization module 5 visualizes the processing process of the multi-source ground observation image through the visualization model.
  • the processing flow of multi-source ground observation images can be generalized into a sequential instruction set formed by a series of modular continuous different image processing sets.
  • the processing flow of remote sensing images includes data input and output, image preprocessing (including geometric correction, fusion, mosaic, etc.), image information extraction (including manual interpretation, automatic classification, feature extraction, dynamic detection, etc.) , Thematic mapping/three-dimensional visualization analysis (including integrated geographic information system existing data, etc.) and results report (including geographic information system analysis and sharing, etc.) and other processing steps.
  • Each processing step in the processing flow corresponds to one or more sub-algorithms, and the corresponding processing and information extraction of remote sensing images are realized through the combination of sub-algorithms.
  • the sub-algorithm can be obtained through query in the open source image processing algorithm library.
  • the graphical object model can be run, edited or saved in the model library. It can be further edited by integrating different processing module sets. Among them, different processing modules include different sub-algorithms, which provide great value in research and testing and subsequent use. The convenience.
  • the graphical object model can also be printed out as a flowchart or displayed and explained in a research report.
  • the model generator provides a variety of operation operators such as data type conversion, image space and time domain basic processing, image transformation, projection calibration, feature extraction, and change monitoring. Raster data, vector data, and Sort data and other operations.
  • This application realizes the visualization of the processing process of the multi-source ground observation image through the visualization model generated according to the processing flow of the multi-source ground observation image, and intuitively represents the various steps of the image processing through the graphical object model in the visualization model.
  • the visibility of data processing is improved, and the need for visual integration of multi-source ground observation image processing is met in a WYSIWYG manner.
  • the acquiring module 1 includes: a first determining unit, which determines a processing flow and each sub-step in the processing flow according to the processing purpose of the multi-source ground observation image; a query unit, which queries the algorithm knowledge base according to the sub-steps and obtains the The sub-algorithms corresponding to the sub-steps, for example, require a clear processing purpose including image preprocessing, image feature extraction and change monitoring, etc.
  • the determined processing flow is the specific way to achieve the processing purpose and the required data support.
  • the association module 3 includes: a second determining unit, which determines the filter type of the graphic object according to the sub-algorithm; a setting unit, which sets the attribute parameters of the graphic object according to the filter type; and a filter determining unit , Determine the filter corresponding to the graphic object; an association unit, associate the sub-algorithm with the graphic object through the filter.
  • the filter refers to the processing object that performs operations on the data, which may cause data changes or generate new data, and the independent operations of the sub-algorithms are encapsulated in the filter through the packaging technology to realize the functions of the sub-algorithms.
  • the filter types include one or more of basic filters such as image denoising, image transformation, image analysis, image segmentation, image compression, image enhancement, image blurring, and image registration or composite filters of image processing.
  • the model generation module includes: a channel construction unit, which constructs a filter connection channel according to the processing flow; and a connection unit, which connects the filters in the graphic object model according to the filter connection channel.
  • the filter connection channel connects the filters that perform various processing steps into a process that performs a series of processing.
  • each processing step of the processing flow is constructed to form a filter connection channel
  • the filter connection channel includes input filter, The first processing filter, the second processing filter, and the output filter.
  • the first processing filter performs the first processing, and the first processing filter
  • the processing result is transmitted to the second processing filter, the second processing is performed, and then the second processing result is transmitted to the output filter, and the final processing result is output.
  • the correlation module 3 also includes a parameter setting unit, which performs parameter setting on the filter and sets the parameters of the sub-algorithm.
  • the correlation module 3 also includes a parameter modification unit, which modifies the parameters of the filter and modifies the parameters of the sub-algorithm. Continuously modify the parameters of the sub-algorithm and review the filtering results to achieve the required processing results. If the requirements are not met, return to re-modify the parameters of the sub-algorithms or adjust the filter type combination until the obtained processing results meet the requirements.
  • a parameter modification unit which modifies the parameters of the filter and modifies the parameters of the sub-algorithm. Continuously modify the parameters of the sub-algorithm and review the filtering results to achieve the required processing results. If the requirements are not met, return to re-modify the parameters of the sub-algorithms or adjust the filter type combination until the obtained processing results meet the requirements.
  • the electronic device can realize the visualization of the simple processing process of the multi-source ground observation image, and can also realize the visualization of the complex processing process of the multi-source ground observation image.
  • the visualization model includes an input graphic object model, a processing graphic object model, and an output graphic object model, so as to realize the visualization of a simple image processing process.
  • the visualization model includes a plurality of processing graphic object models, and each processing graphic object model connects one or more input graphic object models and one or more output graphic object models to realize the visualization of complex processing processes, For example, it can be used to integrate and analyze the feature information of multiple multi-source ground observation images, and output multiple analysis reports from different angles.
  • the visualization module 5 includes: an input unit, which inputs the multi-source ground observation image into a corresponding graphic object model according to the processing flow; an operation unit, which runs the visualization model; and an output unit, which views the corresponding image through the graphic object model.
  • the visualization module 5 also includes: a demand judgment unit that judges whether the processing result meets the processing demand of the multi-source ground observation image, and if the processing demand does not meet the processing demand, corrects one or more graphical objects in the visualization model
  • the parameters of the model for example, the combination of filter types in the visualization model can be modified), and if the processing requirements are met, the processing result and the visualization model are stored.
  • the electronic device further includes: a judgment module to obtain a read instruction of the visualization model; determine whether a visualization model corresponding to the read instruction is stored; when a visualization model corresponding to the read instruction is stored, read the stored Visualization model; visualize the processing process of the multi-source ground observation image through the read visualization model.
  • reading the stored visualization model After reading the stored visualization model, it also includes: correcting the read visualization model according to the processing flow of the multi-source ground observation image, and realizing the visualization of the multi-source ground observation image processing through the modified visualization model. Reading the stored visual model file will automatically execute the process of adding graphical arrows and defining each attribute, which is a process of reusing the visual model of image processing. Among them, the steps of reading the stored visualization model include: reading the shape and reading the arrow.
  • the graphic object model stored in the model library can be read and edited, and further editing and development can be done by integrating different graphic object models.
  • the visual model thus formed can encapsulate the verified processing flow by recording the input and output interfaces between a series of processing procedures, and the parameter settings of different processing procedures or filter operators.
  • the visual model can be used for yourself or Cited by others, with the continuous formation of modules that realize different processing procedures, the module group is further integrated to form a processing process with a certain scale to complete complex image processing functions while removing the time cost and labor cost of human-computer interaction.
  • the electronic device further includes a display module, the display module includes a graphic scene block, an attribute frame block, and a result display block, wherein the graphic scene block is used to place one or more graphic frames (for example, it may be a rectangular frame). , Triangle frame, circular frame, etc.), each graphic frame represents the corresponding processing data and steps, and multiple graphic frames are connected by connecting lines, and the connecting lines are used to indicate the data flow direction (for example, set on the connecting line Arrow) to reflect the sequence of data processing.
  • the display module includes a graphic scene block, an attribute frame block, and a result display block, wherein the graphic scene block is used to place one or more graphic frames (for example, it may be a rectangular frame). , Triangle frame, circular frame, etc.), each graphic frame represents the corresponding processing data and steps, and multiple graphic frames are connected by connecting lines, and the connecting lines are used to indicate the data flow direction (for example, set on the connecting line Arrow) to reflect the sequence of data processing.
  • Both the graphic frame and the connecting line can be dragged instantly, and the connecting line can be dragged to modify the start and end endpoints;
  • the attribute frame section is used to display the currently selected graphic
  • Each attribute of the processing data or step represented by the box, and the attribute value corresponding to the graphic object can be modified through the attribute box section, for example, by clicking the corresponding square item in the attribute box, it is convenient to directly modify each attribute value;
  • the result display section is used to display running results.
  • the result display section includes one or more sub-windows through which intermediate processing results obtained from the processing flow running to the corresponding graphic object model are displayed, and in each sub-window, The sub-window that is the main program window is always displayed on top of other sub-windows.
  • Each link in the image processing process is encapsulated into a visual graphical object model to intuitively represent the data processing process, and the adjustment of parameters and each link can be easily realized in the graphical interface, and the processing results of each link can be passed
  • the sub-window display facilitates the comparison of each processing result.
  • the entire processing flow method in the graphics scene can be stored as a "*.mdl” format file, and unfinished experiments can be continued after reading the last stored "*.mdl” format file. Store files to improve the processing flow.
  • the computer-readable storage medium may be any tangible medium that contains or stores a program or instruction.
  • the program can be executed, and the stored program instructs related hardware to realize the corresponding function.
  • the computer-readable storage medium may be a computer disk, hard disk, random access memory, read-only memory, and so on.
  • the present application is not limited to this, and can be any device that stores instructions or software and any related data files or data structures in a non-transitory manner and can be provided to the processor to enable the processor to execute the programs or instructions therein.
  • the computer-readable storage medium includes a multi-source ground observation image processing visualization program. When the multi-source ground observation image processing visualization program is executed by the processor, the following multi-source ground observation image processing visualization is realized method:
  • the specific implementation of the computer-readable storage medium of the present application is substantially the same as the specific implementation of the above-mentioned multi-source earth observation image processing visualization method and electronic device, and will not be repeated here.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Processing Or Creating Images (AREA)
  • Image Analysis (AREA)

Abstract

一种多源对地观测图像处理的可视化方法、电子装置和存储介质。该方法包括：根据多源对地观测图像的处理流程获取与所述处理流程相对应的多个子算法（S1）；根据所述处理流程生成与每个子算法相对应的图形对象（S2）；通过模型生成器将所述子算法与所述图形对象相关联，生成图形对象模型（S3）；按照所述处理流程连接多个图形对象模型，生成所述多源对地观测图像的可视化模型（S4）；通过所述可视化模型可视化所述多源对地观测图像的处理过程（S5）。该方法实现对多源对地观测图像的处理过程的可视化，直观地显示各个处理步骤，以所见即所得的方式满足对多源对地观测图像处理过程进行可视化整合的需求。

Description

多源对地观测图像处理的可视化方法、装置及存储介质
本申请要求于2019年04月11日提交的中国专利申请号201910290454.X的优先权益，上述案件全部内容以引用的方式并入本文中。
技术领域
本申请涉及大数据技术领域，尤其涉及一种多源对地观测图像处理的可视化方法、装置及存储介质。
背景技术
可视化建模(VISUAL MODELING)是利用围绕现实想法组织模型的一种思考问题的方法，是以图形的方式描述所开发的系统的过程。可视化建模允许提出一个复杂问题的必要细节，过滤不必要的细节。它也提供了一种从不同的视角观察被开发系统的机制。
目前，常见的可视化建模软件有统一建模语言(Unified Modeling Language，UML)、VISIO、Simulink、Model Maker以及Model Builder等，现有的可视化平台可以通过可视化建模软件提供图形化建模用户界面，但是，现有的可视化平台不能将对地观测数据(包括卫星图像、结构化数据等)处理挖掘过程中的涉及到的数据、算法、流程及其幕后隐藏的信息流进行所见即所得的可视化处理整合，难以实现对观测数据处理过程中的算法进行串行或并行处理，难以满足可视化处理调试、提高开发效率的需求。
发明内容
本申请提供一种多源对地观测图像处理的可视化方法、装置及存储介质，以解决现有技术难以满足对地观测数据处理挖掘的过程进行可视化处理整合的问题。
为了实现上述目的，本申请的一个方面是提供一种多源对地观测图像处理的可视化方法，包括：
根据多源对地观测图像的处理流程获取与所述处理流程相对应的多个子 算法；根据所述处理流程生成与每个子算法相对应的图形对象；通过模型生成器将所述子算法与所述图形对象相关联，生成图形对象模型；按照所述处理流程连接多个图形对象模型，生成所述多源对地观测图像的可视化模型；通过所述可视化模型可视化所述多源对地观测图像的处理过程。
为了实现上述目的，本申请的另一个方面是提供一种电子装置，该电子装置包括：处理器和存储器，所述存储器中包括多源对地观测图像处理的可视化程序，所述多源对地观测图像处理的可视化程序被所述处理器执行时实现如上所述的多源对地观测图像处理的可视化方法的步骤。
为了实现上述目的，本申请的再一个方面是提供一种计算机可读存储介质，所述计算机可读存储介质中包括多源对地观测图像处理的可视化程序，所述多源对地观测图像处理的可视化程序被处理器执行时，实现如上所述的多源对地观测图像处理的可视化方法的步骤。
相对于现有技术，本申请具有以下优点和有益效果：
本申请通过根据多源对地观测图像的处理流程生成的可视化模型实现对多源对地观测图像的处理过程的可视化，通过可视化模型中的图形对象模型直观地表示对图像进行处理的各个步骤，提高了数据处理的可视性，以所见即所得的方式满足对多源对地观测图像处理过程进行可视化整合的需求。
本申请通过将子算法与图形对象相关联提高了对图像处理过程中各个子算法的可编辑性，增强了可视化模型中各个图形对象模型组合的灵活性，提高了开发调试效率，实现了对图像处理研究的设计可视化、中间调用可视化和处理结果可视化，有效满足了用户进行高层次开发应用的需要。
附图说明
图1为本申请所述多源对地观测图像处理的可视化方法的流程示意图；
图2为本申请中多源对地观测图像处理的可视化程序的模块示意图。
本申请目的的实现、功能特点及优点将结合实施例，参照附图做进一步说明。
具体实施方式
下面将参考附图来描述本申请所述的实施例。本领域的普通技术人员可 以认识到，在不偏离本申请的精神和范围的情况下，可以用各种不同的方式或其组合对所描述的实施例进行修正。因此，附图和描述在本质上是说明性的，仅仅用以解释本申请，而不是用于限制权利要求的保护范围。此外，在本说明书中，附图未按比例画出，并且相同的附图标记表示相同的部分。
图1为本申请所述多源对地观测图像处理的可视化方法的流程示意图，如图1所示，本申请所述多源对地观测图像处理的可视化方法，包括以下步骤：
步骤S1、根据多源对地观测图像的处理流程获取与所述处理流程相对应的多个子算法。
多源对地观测图像的处理流程可以概化成一系列模块化的连续的不同图像处理集形成的顺序指令集合。以遥感图像为例，遥感图像的处理流程包括数据输入输出、图像的预处理(包括几何纠正、融合、镶嵌等)、影像信息提取(包括人工解译、自动分类、特征提取、动态检测等)、专题制图/三维可视化分析(包括集成地理信息系统的现有数据等)和成果报告(包括地理信息系统的分析和共享等)等处理步骤。处理流程中的每个处理步骤都对应一个或多个子算法，通过子算法的组合实现对遥感图像的相应处理和信息提取。子算法可以通过开源图像处理算法库中查询获取。
步骤S2、根据所述处理流程生成与每个子算法相对应的图形对象，每个图形对象表征一个子算法，图形对象可以使用相同或不同的形状。具体地，使用跨平台的C++图形用户界面应用程序框架(Qt)中的Graphics View框架作为图形用户界面，用于生成图形对象。Graphics View框架包括Diagram Scene、Diagram View和Diagram Item，其中，Diagram Scene是可视化工作区，Diagram Item是可以放置于工作区的二维图形框，Diagram View用于完成Diagram Scene中内容的显示，在Diagram Scene工作区中可以添加形状、文本以及创建连接方向线，生成的图形对象即为在工作区中添加的二维图形框，并可以对二维图形框的形状进行调整，通过使用空间场景管理的空间分割树(Binary Space Partitioning tree，BSP树)存储生成的图形对象。
步骤S3、通过模型生成器将所述子算法与所述图形对象相关联，生成图形对象模型。在面向目标的模型语言环境中，可以直观地在一个页面场景上绘制流程图，将流程图中的各个图形对象与表征输入数据、操作函数、运算 规则和输出数据等子算法关联，由此生成多个图像处理步骤的空间对象模型。
图形对象模型可以运行、编辑或保存在模型库中，通过集成不同的处理模块集合可以做进一步的编辑，其中，不同的处理模块包括不同的子算法，在研究测试与后续投入使用中提供很大的便利。图形对象模型也可以作为流程图打印输出，或在研究报告中展示讲解。模型生成器中提供了数据类型转换、图像空间域时间域的基本处理、图像变换、投影校准、特征提取和变化监测等多种操作算子，可以通过图形对象模型进行栅格数据、矢量数据和分类数据等操作。在一个实施例中，多源对地观测图像处理的可视化方法基于Qt跨平台图形界面实现，通过使用面板工具来产生空间图形对象模型。
步骤S4、按照所述处理流程连接多个图形对象模型，生成所述多源对地观测图像的可视化模型。一个可视化模型是由一系列的图形对象模型构成，步骤S3中生成的各个图形对象模型是相互独立的个体，是一个一个的空间模型要素，需要将各个图形对象模型有机连接起来，才能形成一个完整的可视化模型，由于生成的可视化模型用于完成对多源对地观测图像的处理，因此，各个图形对象模型之间的连接也是有序的，按照对多源对地观测图像的处理流程进行连接，使得可视化模型在运行时，按照处理流程依次运行各个图形对象模型中的子算法，从而通过各个图形对象模型完成空间地理信息和图像处理的操作功能。其中，连接可以通过在各个图形对象模型之间设置带箭头的连接线实现，通过箭头表示数据流的方向。
步骤S5、通过所述可视化模型可视化所述多源对地观测图像的处理过程。
本申请通过根据多源对地观测图像的处理流程生成的可视化模型实现对多源对地观测图像的处理过程的可视化，通过可视化模型中的图形对象模型直观地表示对图像进行处理的各个步骤，提高了数据处理的可视性，以所见即所得的方式满足对多源对地观测图像处理过程进行可视化整合的需求。
本申请通过将子算法与图形对象相关联提高了对图像处理过程中各个子算法的可编辑性，增强了可视化模型中各个图形对象模型组合的灵活性，提高了开发调试效率，实现了对图像处理研究的设计可视化、中间调用可视化和处理结果可视化，有效满足了用户进行高层次开发应用的需要。
本申请的一个可选实施例中，根据多源对地观测图像的处理流程获取与所述处理流程相对应的多个子算法的步骤包括：根据对多源对地观测图像的 处理目的确定处理流程以及处理流程中的各个子步骤；根据所述子步骤查询算法知识库并获取与所述子步骤对应的子算法，其中，需要明确的处理目的包括图像预处理、图像特征提取和变化监测等，确定的处理流程即为达到处理目的的具体途径和所需要的数据支持。例如，若对多源对地观测图像的处理目的是提取图像特征(道路、建筑物、水体等)，则根据此处理目的，得到提取图像特征的处理流程，以提取遥感图像中的建筑物特征为例，处理流程包括：对遥感图像进行预处理，对经过预处理得到的遥感图像进行边缘检测，从检测到的边缘信息中提取直线信息，根据直线信息提取矩形建筑物结构，从而得到遥感图像中的建筑物特征，其中，处理流程中的对遥感图像进行预处理、边缘检测、提取直线信息、提取矩形结构分别为一个子步骤，通过查询算法知识库查询与相应子步骤对应的子算法，以便于通过子算法实现相应的处理功能。
优选地，将所述子算法与所述图形对象相关联的步骤包括：根据所述子算法确定所述图形对象的滤波器类型；根据所述滤波器类型设置所述图形对象的属性参数，确定所述图形对象对应的滤波器；通过所述滤波器将所述子算法与所述图形对象相关联。其中，滤波器指对数据执行操作，可能引起数据变化或产生新数据的处理对象，将子算法的独立操作通过封装技术封装在滤波器中实现子算法的功能。
所述滤波器类型包括图像处理的图像去噪、图像变换、图像分析、图像分割、图像压缩、图像增强、图像模糊、图像配准等基础滤波器或者复合滤波器中的一种或多种。
进一步地，按照所述处理流程连接多个图形对象模型的步骤包括：根据所述处理流程构建滤波器连接通道；将所述图形对象模型中的滤波器根据所述滤波器连接通道连接，从而使得各个滤波器按照对多源对地观测图像的处理流程连接起来，以便于对多源对地观测图像的处理。
其中，滤波器连接通道将执行各种处理步骤的滤波器连接成一条执行一系列处理的流程，例如，在远程传感图像处理库(ORFEO Tool Box，OTB)中可以通过SetInput()和GetOutput()函数实现将图形对象模型中的滤波器根据所述滤波器连接通道连接的步骤，其中，ORFEO是Optical and Radar Federated Earth Observation的简称。在执行滤波器连接通道的处理过程时，将图形对象 模型中的滤波器根据所述滤波器连接通道连接的步骤之后，还包括：在滤波器连接通道的某一环节调用执行函数Update()，更新图形对象模型中的滤波器的数据，会从连接通道的起始位置开始，以最新数据依次执行通道内各个滤波器操作，到连接通道的终点位置为止。以一个处理流程包括输入、第一次处理、第二次处理和输出为例进行说明，将该处理流程的各个处理步骤构建形成一个滤波器连接通道，则该滤波器连接通道包括输入滤波器、第一处理滤波器、第二处理滤波器和输出滤波器，工作时，先将图像文件通过所述输入滤波器获取图像数据，由第一处理滤波器进行第一次处理，将第一次的处理结果传输至第二处理滤波器，进行第二次处理，然后将第二次的处理结果传至输出滤波器，输出最终的处理结果。
进一步地，确定所述图形对象对应的滤波器的步骤之后，还包括：对所述滤波器进行参数设置，设置所述子算法的参数。
进一步地，确定所述图形对象对应的滤波器的步骤之后，还包括：对所述滤波器进行参数修改，修改所述子算法的参数。通过不断地修正子算法的参数并检视滤波结果，以达到需要的处理结果，若滤波结果不满足需求，则返回重新修改子算法的参数或者调整滤波器类型组合，直至滤波结果满足需求。
所述可视化方法可以实现对所述多源对地观测图像的简单处理过程的可视化，也可以实现对所述多源对地观测图像的复杂处理过程的可视化。优选地，所述可视化模型包括输入图形对象模型、处理图形对象模型和输出图形对象模型，以实现对图像简单处理过程的可视化。优选地，所述可视化模型包括一个或多个处理图形对象模型，每个处理图形对象模型均连接一个或多个输入图形对象模型与一个或多个输出图形对象模型，以实现对复杂处理过程的可视化，例如，可以用于对多个多源对地观测图像的特征信息进行整合，分析，输出多个不同角度的分析报告等。
优选地，通过所述可视化模型可视化所述多源对地观测图像的处理过程的步骤包括：根据所述处理流程将所述多源对地观测图像输入相应的图形对象模型；运行所述可视化模型；通过图形对象模型查看对应的处理步骤，输出所述多源对地观测图像的处理结果。
进一步地，输出所述多源对地观测图像的处理结果的步骤之后，还包括：
判断所述处理结果是否满足多源对地观测图像的处理需求，其中，处理需求与上述的处理目的相对应，例如，若处理目的是提取遥感图像中的建筑物目标特征，则处理需求包括：处理后的遥感图像中的所有建筑物目标特征均被提取，且提取的建筑物目标准确等。若不满足处理需求，则修正所述可视化模型中的一个或多个图形对象模型的参数(例如，可以修正可视化模型中滤波器类型的组合)，若满足处理需求，则存储所述处理结果和所述可视化模型。通过不断修正参数，使得最终输出的处理结果满足对多源对地观测图像的处理目的，并且，可以通过不断的修正参数，获取最优的处理结果，最终存储较优的可视化模型。
所述多源对地观测图像处理的可视化方法还包括：根据多源对地观测图像的处理流程获取与所述处理流程相对应的多个子算法之前，获取可视化模型的读取指令；确定是否存储有与读取指令相对应的可视化模型；当存储有与读取指令相对应的可视化模型时，读取存储的可视化模型，不再进行生成可视化模型的步骤；通过读取的可视化模型可视化所述多源对地观测图像的处理过程。当未存储有与读取指令相对应的可视化模型时，则根据处理流程获取相对应的多个子算法，生成图形对象模型和可视化模型，通过生成的可视化模型可视化多源对地观测图像的处理过程。
读取存储的可视化模型之后，还包括：根据多源对地观测图像的处理流程修正读取的可视化模型，通过修正得到的可视化模型可视化所述多源对地观测图像的处理过程，从而实现多源对地观测图像处理的可视化。读取存储的可视化模型文件会自动执行添加图形箭头并定义各个属性的过程，是复用图像处理的可视化模型的过程。其中，读取存储的可视化模型的步骤包括：读形状与读箭头两个步骤，其中，所述形状指的是所述可视化模型中的图形对象模型的形状，所述箭头指的是所述可视化模型中的多个图像对象模型之间的连接箭头。优选地，读形状的步骤包括：获取图形对象的图形框位置坐标与相应的过滤器类型；在图形场景版块中相应的位置坐标下建立图形框，并设置过滤器类型；以“属性—值”的形式读出属性值；根据所述属性值依次调用预先在算法知识库中定义的设置属性接口滤波算法函数，将对应的属性项设置成相应的属性值。读箭头的步骤包括：获取箭头起始点坐标；按照箭头起始点坐标搜索所述箭头所在位置的形状，再与手动画箭头过程一样， 建立连接起始点的箭头。通过上述的读取步骤可读取和编辑保存在模型库中的图形对象模型，通过集成不同的图形对象模型可以做进一步的编辑和研发。由此形成的可视化模型通过记录一系列的处理过程之间的输入输出接口，以及不同的处理过程或滤波算子的参数设置便可以将经过验证的处理流程封装起来，该可视化模型可以供自己或他人引用，随着实现不同处理流程的模块的不断形成，进一步整合模块群形成具有一定规模的处理过程，完成复杂的图像处理功能的同时，去除了中间的人机交互的时间成本和人力成本。
下面结合界面部分及对应后台处理部分对可视化方法的流程进一步说明：
首先在图形场景区域通过创建代表的远程传感图像处理库(ORFEO Tool Box，OTB)中过滤器的形状并用箭头连接，或者读取“*.mdl”格式文件的方式布置出完整的处理流程，此时后台将对应的OTB过滤器创建完毕，同时将被选中过滤器形状对应的OTB过滤器含有的可调整参数项显示在属性框中；
然后通过选中图形场景中不同的过滤器形状在属性框显示相应的可调整参数项，在显示的参数项方格中单击可设置参数项的值，对各个OTB过滤器调整所需的参数；
接下来选中处理流程中某一处理步骤对应的过滤器，可选择执行或执行并显示执行结果，完成一次操作；
然后通过修改各个图形对象模型的位置改变处理流程，或者修改参数项调整实验参数，分别查看对比实验结果。
通过实验结果的对比最终确定可视化模型中各个图形对象的参数，使得保存的可视化模型的精度更高。
本申请的可视化方法适用于数十种卫星图像、空间地理信息等大数据的处理和可视化分析。
本申请所述多源对地观测图像处理的可视化方法应用于电子装置，所述电子装置可以是电视机、智能手机、平板电脑、计算机等终端设备。
所述电子装置包括：处理器；存储器，用于存储多源对地观测图像处理的可视化程序，处理器执行所述多源对地观测图像处理的可视化程序，实现以下的多源对地观测图像处理的可视化方法的步骤：
根据多源对地观测图像的处理流程获取与所述处理流程相对应的多个子算法；根据所述处理流程生成与每个子算法相对应的图形对象；通过模型生 成器将所述子算法与所述图形对象相关联，生成图形对象模型；按照所述处理流程连接多个图形对象模型，生成所述多源对地观测图像的可视化模型；通过所述可视化模型可视化所述多源对地观测图像的处理过程。
所述电子装置还包括网络接口和通信总线等。其中，网络接口可以包括标准的有线接口、无线接口，通信总线用于实现各个组件之间的连接通信。
存储器包括至少一种类型的可读存储介质，可以是闪存、硬盘、光盘等非易失性存储介质，也可以是插接式硬盘等，且并不限于此，可以是以非暂时性方式存储指令或软件以及任何相关联的数据文件并向处理器提供指令或软件程序以使该处理器能够执行指令或软件程序的任何装置。本申请中，存储器存储的软件程序包括多源对地观测图像处理的可视化程序，并可以向处理器提供该多源对地观测图像处理的可视化程序，以使得处理器可以执行该多源对地观测图像处理的可视化程序，实现多源对地观测图像处理的可视化方法的步骤。
处理器可以是中央处理器、微处理器或其他数据处理芯片等，可以运行存储器中的存储程序，例如，本申请中多源对地观测图像处理的可视化程序。
所述电子装置还可以包括显示器，显示器也可以称为显示屏或显示单元。在一些实施例中显示器可以是LED显示器、液晶显示器、触控式液晶显示器以及有机发光二极管(Organic Light-Emitting Diode，OLED)触摸器等。显示器用于显示在电子装置中处理的信息以及用于显示可视化的工作界面。
所述电子装置还可以包括用户接口，用户接口可以包括输入单元(比如键盘)、语音输出装置(比如音响、耳机)等。
在其他实施例中，多源对地观测图像处理的可视化程序还可以被分割为一个或者多个模块，一个或者多个模块被存储于存储器中，并由处理器执行，以完成本申请。本申请所称的模块是指能够完成特定功能的一系列计算机程序指令段。图2为本申请中多源对地观测图像处理的可视化程序的模块示意图，如图2所示，所述多源对地观测图像处理的可视化程序可以被分割为：获取模块1、图形对象生成模块2、关联模块3、模型生成模块4和可视化模块5。上述模块所实现的功能或操作步骤均与上文类似，此处不再详述，示例性地，例如其中：
获取模块1，根据多源对地观测图像的处理流程获取与所述处理流程相对 应的多个子算法；
图形对象生成模块2，根据所述处理流程生成与每个子算法相对应的图形对象；
关联模块3，通过模型生成器将所述子算法与所述图形对象相关联，生成图形对象模型；
模型生成模块4，按照所述处理流程连接多个图形对象模型，生成所述多源对地观测图像的可视化模型；
可视化模块5，通过所述可视化模型可视化所述多源对地观测图像的处理过程。
多源对地观测图像的处理流程可以概化成一系列模块化的连续的不同图像处理集形成的顺序指令集合。以遥感图像为例，遥感图像的处理流程包括数据输入输出、图像的预处理(包括几何纠正、融合、镶嵌等)、影像信息提取(包括人工解译、自动分类、特征提取、动态检测等)、专题制图/三维可视化分析(包括集成地理信息系统的现有数据等)和成果报告(包括地理信息系统的分析和共享等)等处理步骤。处理流程中的每个处理步骤都对应一个或多个子算法，通过子算法的组合实现对遥感图像的相应处理和信息提取。子算法可以通过开源图像处理算法库中查询获取。
图形对象模型可以运行、编辑或保存在模型库中，通过集成不同的处理模块集合可以做进一步的编辑，其中，不同的处理模块包括不同的子算法，在研究测试与后续投入使用中提供很大的便利。图形对象模型也可以作为流程图打印输出，或在研究报告中展示讲解。模型生成器中提供了数据类型转换、图像空间域时间域的基本处理、图像变换、投影校准、特征提取和变化监测等多种操作算子，可以通过图形对象模型进行栅格数据、矢量数据和分类数据等操作。
本申请通过根据多源对地观测图像的处理流程生成的可视化模型实现对多源对地观测图像的处理过程的可视化，通过可视化模型中的图形对象模型直观地表示对图像进行处理的各个步骤，提高了数据处理的可视性，以所见即所得的方式满足对多源对地观测图像处理过程进行可视化整合的需求。
获取模块1包括：第一确定单元，根据对多源对地观测图像的处理目的确定处理流程以及处理流程中的各个子步骤；查询单元，根据所述子步骤查 询算法知识库并获取与所述子步骤对应的子算法，例如，需要明确的处理目的包括图像预处理、图像特征提取和变化监测等，确定的处理流程即为达到处理目的的具体途径和所需要的数据支持。
优选地，关联模块3包括：第二确定单元，根据所述子算法确定所述图形对象的滤波器类型；设置单元，根据所述滤波器类型设置所述图形对象的属性参数；滤波器确定单元，确定所述图形对象对应的滤波器；关联单元，通过所述滤波器将所述子算法与所述图形对象相关联。其中，滤波器指对数据执行操作，可能引起数据变化或产生新数据的处理对象，将子算法的独立操作通过封装技术封装在滤波器中实现子算法的功能。
所述滤波器类型包括图像处理的图像去噪、图像变换、图像分析、图像分割、图像压缩、图像增强、图像模糊、图像配准等基础滤波器或者复合滤波器中的一种或多种。
进一步地，模型生成模块包括：通道构建单元，根据所述处理流程构建滤波器连接通道；连接单元，将所述图形对象模型中的滤波器按照所述滤波器连接通道连接。
其中，滤波器连接通道将执行各种处理步骤的滤波器连接成一条执行一系列处理的流程，在执行滤波器连接通道的处理过程时，从连接通道的起始位置开始，以最新数据依次执行通道内各个滤波器操作，到连接通道的终点位置为止。以一个处理流程包括输入、第一次处理、第二次处理和输出为例进行说明，将该处理流程的各个处理步骤构建形成一个滤波器连接通道，则该滤波器连接通道包括输入滤波器、第一处理滤波器、第二处理滤波器和输出滤波器，工作时，先将图像文件通过所述输入滤波器获取图像数据，由第一处理滤波器进行第一次处理，将第一次的处理结果传输至第二处理滤波器，进行第二次处理，然后将第二次的处理结果传至输出滤波器，输出最终的处理结果。
进一步地，关联模块3还包括参数设置单元，对所述滤波器进行参数设置，设置所述子算法的参数。
进一步地，关联模块3还包括参数修正单元，对所述滤波器进行参数修改，修改所述子算法的参数。通过不断地修正子算法的参数并检视滤波结果，以达到需要的处理结果，若不满足需求，则返回重新修改子算法的参数或者 调整滤波器类型组合，直至得到的处理结果满足需求。
所述电子装置可以实现对所述多源对地观测图像的简单处理过程的可视化，也可以实现对所述多源对地观测图像的复杂处理过程的可视化。优选地，所述可视化模型包括输入图形对象模型、处理图形对象模型和输出图形对象模型，以实现对图像简单处理过程的可视化。优选地，所述可视化模型包括多个处理图形对象模型，每个处理图形对象模型均连接一个或多个输入图形对象模型与一个或多个输出图形对象模型，以实现对复杂处理过程的可视化，例如，可以用于对多个多源对地观测图像的特征信息进行整合，分析，输出多个不同角度的分析报告等。
优选地，可视化模块5包括：输入单元，根据所述处理流程将所述多源对地观测图像输入相应的图形对象模型；运行单元，运行所述可视化模型；输出单元，通过图形对象模型查看对应的处理步骤，输出所述多源对地观测图像的处理结果。
进一步地，可视化模块5还包括：需求判断单元，判断所述处理结果是否满足多源对地观测图像的处理需求，若不满足处理需求，则修正所述可视化模型中的一个或多个图形对象模型的参数(例如，可以修正可视化模型中滤波器类型的组合)，若满足处理需求，则存储所述处理结果和所述可视化模型。
所述电子装置还包括：判断模块，获取可视化模型的读取指令；确定是否存储有与读取指令相对应的可视化模型；当存储有与读取指令相对应的可视化模型时，读取存储的可视化模型；通过读取的可视化模型可视化所述多源对地观测图像的处理过程。
读取存储的可视化模型之后，还包括：根据多源对地观测图像的处理流程修正读取的可视化模型，通过修正得到的可视化模型实现多源对地观测图像处理的可视化。读取存储的可视化模型文件会自动执行添加图形箭头并定义各个属性的过程，是复用图像处理的可视化模型的过程。其中，读取存储的可视化模型的步骤包括：读形状与读箭头两个步骤，读形状时，先获取图形对象的图形框位置坐标与相应的过滤器类型，在图形场景版块中相应的位置坐标下建立图形框，并设置过滤器类型，再以“属性—值”的形式读出属性值，并根据所述属性值依次调用预先在算法知识库中定义的设置属性接口 滤波算法函数，将对应的属性项设置成相应的属性值。读箭头时，先按照箭头起始点坐标搜索该位置的形状，再与手动画箭头过程一样，建立连接起始点形状的箭头。通过上述的读取步骤可读取和编辑保存在模型库中的图形对象模型，通过集成不同的图形对象模型可以做进一步的编辑和研发。由此形成的可视化模型通过记录一系列的处理过程之间的输入输出接口，以及不同的处理过程或滤波算子的参数设置便可以将经过验证的处理流程封装起来，该可视化模型可以供自己或他人引用，随着实现不同处理流程的模块的不断形成，进一步整合模块群形成具有一定规模的处理过程，完成复杂的图像处理功能的同时，去除了中间的人机交互的时间成本和人力成本。
所述电子装置还包括：显示模块，所述显示模块包括图形场景版块、属性框版块和结果显示版块，其中，所述图形场景版块用于放置一个或多个图形框(例如，可以是矩形框、三角形框、圆形框等)，每个图形框表示对应的处理数据和步骤，多个图形框之间通过连接线连接，所述连接线用于表示数据流向(例如，在连接线上设置箭头)，以体现对数据所做处理的先后流程，图形框和连接线均可以即时拖动，连接线可以通过拖动修改起始和终止端点；所述属性框版块用于显示当前选中的图形框所表示的处理数据或者步骤的各个属性，并且，通过所述属性框版块可以修改图形对象所对应的属性值，例如，通过点击属性框中相应方格项方便地直接修改各个属性值；所述结果显示版块用于显示运行结果，所述结果显示版块包括一个或多个子窗口，通过所述子窗口显示处理流程运行至相应的图形对象模型得到的中间处理结果，并且，各个子窗口中，作为主程序窗口的子窗口始终显示于其他子窗口的上层。
将对图像的处理过程中的各个环节均封装成可视化的图形对象模型，以直观地表示数据的处理过程，并且在图形界面方便实现参数与各个环节的调整，将各个环节的处理结果均可通过子窗口显示，便于各个处理结果的比较。
另外，可以将图形场景中的整个处理流程方法存储为“*.mdl”格式文件，未完成的实验可以在读取上次存储的“*.mdl”格式文件后继续进行，可根据读取的存储文件进行处理流程的完善。
本申请的一个实施例中，计算机可读存储介质可以是任何包含或存储程序或指令的有形介质，其中的程序可以被执行，通过存储的程序指令相关的 硬件实现相应的功能。例如，计算机可读存储介质可以是计算机磁盘、硬盘、随机存取存储器、只读存储器等。本申请并不限于此，可以是以非暂时性方式存储指令或软件以及任何相关数据文件或数据结构并且可提供给处理器以使处理器执行其中的程序或指令的任何装置。所述计算机可读存储介质中包括多源对地观测图像处理的可视化程序，所述多源对地观测图像处理的可视化程序被处理器执行时，实现如下的多源对地观测图像处理的可视化方法：
根据多源对地观测图像的处理流程获取与所述处理流程相对应的多个子算法；根据所述处理流程生成与每个子算法相对应的图形对象；通过模型生成器将所述子算法与所述图形对象相关联，生成图形对象模型；按照所述处理流程连接多个图形对象模型，生成所述多源对地观测图像的可视化模型；通过所述可视化模型可视化所述多源对地观测图像的处理过程。
本申请之计算机可读存储介质的具体实施方式与上述多源对地观测图像处理的可视化方法、电子装置的具体实施方式大致相同，在此不再赘述。
需要说明的是，在本文中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、装置、物品或者方法不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、装置、物品或者方法所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括该要素的过程、装置、物品或者方法中还存在另外的相同要素。
上述本申请实施例序号仅仅为了描述，不代表实施例的优劣。通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品存储在如上所述的一个存储介质(如ROM/RAM、磁碟、光盘)中，包括若干指令用以使得一台终端设备(可以是手机，计算机，服务器，或者网络设备等)执行本申请各个实施例所述的方法。
以上仅为本申请的优选实施例，并非因此限制本申请的专利范围，凡是利用本申请说明书及附图内容所作的等效结构或等效流程变换，或直接或间接运用在其他相关的技术领域，均同理包括在本申请的专利保护范围内。

Claims (20)

  1. 一种多源对地观测图像处理的可视化方法，应用于电子装置，其特征在于，包括：
    根据多源对地观测图像的处理流程获取与所述处理流程相对应的多个子算法；
    根据所述处理流程生成与每个子算法相对应的图形对象；
    通过模型生成器将所述子算法与所述图形对象相关联，生成图形对象模型；
    按照所述处理流程连接多个图形对象模型，生成所述多源对地观测图像的可视化模型；
    通过所述可视化模型可视化所述多源对地观测图像的处理过程。
  2. 根据权利要求1所述的多源对地观测图像处理的可视化方法，其特征在于，根据多源对地观测图像的处理流程获取与所述处理流程相对应的多个子算法的步骤包括：
    根据对多源对地观测图像的处理目的确定处理流程以及处理流程中的各个子步骤；
    根据所述子步骤查询算法知识库并获取与所述子步骤对应的子算法。
  3. 根据权利要求1所述的多源对地观测图像处理的可视化方法，其特征在于，将所述子算法与所述图形对象相关联的步骤包括：
    根据所述子算法确定所述图形对象的滤波器类型；
    根据所述滤波器类型设置所述图形对象的属性参数，确定所述图形对象对应的滤波器；
    通过所述滤波器将所述子算法与所述图形对象相关联。
  4. 根据权利要求3所述的多源对地观测图像处理的可视化方法，其特征在于，确定所述图形对象对应的滤波器的步骤之后，还包括：对所述滤波器进行参数设置，设置所述子算法的参数。
  5. 根据权利要求3所述的多源对地观测图像处理的可视化方法，其特征在于，确定所述图形对象对应的滤波器的步骤之后，还包括：对所述滤波器进行参数修改，修改所述子算法的参数，并判断滤波结果是否满足需求，若 不满足需求，则返回重新修改子算法的参数或者调整滤波器类型的组合，直至滤波结果满足需求。
  6. 根据权利要求3所述的多源对地观测图像处理的可视化方法，其特征在于，所述按照所述处理流程连接多个图形对象模型的步骤包括：
    根据所述处理流程构建滤波器连接通道；
    将所述图形对象模型中的滤波器根据所述滤波器连接通道连接。
  7. 根据权利要求6所述的多源对地观测图像处理的可视化方法，其特征在于，所述图形对象模型中的滤波器根据所述滤波器连接通道连接，通过在远程传感图像处理库中的SetInput()和GetOutput()函数实现。
  8. 根据权利要求6所述的多源对地观测图像处理的可视化方法，其特征在于，将所述图形对象模型中的滤波器根据所述滤波器连接通道连接的步骤之后，还包括：调用执行函数Update()，更新图形对象模型中的滤波器的数据，从所述滤波器连接通道的起点位置开始，以更新后的数据依次执行所述滤波器连接通道内的各个滤波器操作，到所述滤波器连接通道的终点位置为止。
  9. 根据权利要求1所述的多源对地观测图像处理的可视化方法，其特征在于，通过所述可视化模型可视化所述多源对地观测图像的处理过程的步骤包括：
    根据所述处理流程将所述多源对地观测图像输入相应的图形对象模型；
    运行所述可视化模型；
    通过图形对象模型查看对应的处理步骤，输出所述多源对地观测图像的处理结果。
  10. 根据权利要求9所述的多源对地观测图像处理的可视化方法，其特征在于，输出所述多源对地观测图像的处理结果的步骤之后，还包括：
    判断所述处理结果是否满足多源对地观测图像的处理需求，若不满足处理需求，则修正所述可视化模型中的一个或多个图形对象模型的参数，若满足处理需求，则存储所述处理结果和所述可视化模型。
  11. 根据权利要求1所述的多源对地观测图像处理的可视化方法，其特征在于，根据多源对地观测图像的处理流程获取与所述处理流程相对应的多个子算法之前，还包括：
    获取可视化模型的读取指令；
    确定是否存储有与读取指令相对应的可视化模型；
    当存储有与读取指令相对应的可视化模型时，读取存储的可视化模型；
    通过读取的可视化模型可视化所述多源对地观测图像的处理过程。
  12. 根据权利要求11所述的多源对地观测图像处理的可视化方法，其特征在于，读取存储的可视化模型之后，还包括：根据多源对地观测图像的处理流程修正读取的可视化模型，通过修正后的可视化模型可视化所述多源对地观测图像的处理过程。
  13. 根据权利要求11所述的多源对地观测图像处理的可视化方法，其特征在于，读取存储的可视化模型的步骤包括：读形状与读箭头两个步骤，其中，所述形状指的是所述可视化模型中的图形对象模型的形状，所述箭头指的是所述可视化模型中的多个图像对象模型之间的连接箭头。
  14. 根据权利要求13所述的多源对地观测图像处理的可视化方法，其特征在于，读形状的步骤包括：
    获取图形对象的图形框位置坐标与相应的过滤器类型；
    在图形场景版块中相应的位置坐标下建立图形框，并设置过滤器类型；
    以“属性—值”的形式读出属性值；
    根据所述属性值依次调用预先在算法知识库中定义的设置属性接口滤波算法函数，将对应的属性项设置成相应的属性值。
  15. 根据权利要求13所述的多源对地观测图像处理的可视化方法，其特征在于，读箭头的步骤包括：
    获取箭头起始点坐标；
    按照所述箭头起始点坐标搜索所述箭头所在位置的形状；
    建立连接起始点的箭头。
  16. 根据权利要求1所述的多源对地观测图像处理的可视化方法，其特征在于，所述可视化模型包括输入图形对象模型、处理图形对象模型和输出图形对象模型，其中，所述处理图形对象模型有一个或多个，且每个处理图形对象模型均连接一个或多个输入图形对象模型与一个或多个输出图形对象模型。
  17. 根据权利要求1所述的多源对地观测图像处理的可视化方法，其特 征在于，根据所述处理流程生成与每个子算法相对应的图形对象，通过将跨平台的C++图形用户界面应用程序框架中的Graphics View框架作为图形用户界面实现。
  18. 一种电子装置，其特征在于，该电子装置包括：处理器和存储器，所述存储器中包括多源对地观测图像处理的可视化程序，所述多源对地观测图像处理的可视化程序被所述处理器执行时实现如权利要求1至17中任一项所述的多源对地观测图像处理的可视化方法的步骤。
  19. 根据权利要求18所述的电子装置，其特征在于，所述电子装置还包括：显示模块，所述显示模块包括图形场景版块、属性框版块和结果显示版块，其中，所述图形场景版块用于放置一个或多个图形框，每个图形框表示对应的处理数据和步骤，多个图形框之间通过连接线连接，所述连接线用于表示数据流向；所述属性框版块用于显示当前选中的图形框所表示的处理数据或者步骤的各个属性；所述结果显示版块用于显示运行结果，所述结果显示版块包括一个或多个子窗口，通过所述子窗口显示处理流程运行至相应的图形对象模型得到的中间处理结果。
  20. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质中包括多源对地观测图像处理的可视化程序，所述多源对地观测图像处理的可视化程序被处理器执行时，实现如权利要求1至17中任一项所述的多源对地观测图像处理的可视化方法的步骤。
PCT/CN2019/103584 2019-04-11 2019-08-30 多源对地观测图像处理的可视化方法、装置及存储介质 WO2020206914A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2021506308A JP7264989B2 (ja) 2019-04-11 2019-08-30 マルチソース地球観測画像処理の可視化方法、機器及び記録媒体

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910290454.X 2019-04-11
CN201910290454.XA CN110110114B (zh) 2019-04-11 2019-04-11 多源对地观测图像处理的可视化方法、装置及存储介质

Publications (1)

Publication Number Publication Date
WO2020206914A1 true WO2020206914A1 (zh) 2020-10-15

Family

ID=67483832

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/103584 WO2020206914A1 (zh) 2019-04-11 2019-08-30 多源对地观测图像处理的可视化方法、装置及存储介质

Country Status (3)

Country Link
JP (1) JP7264989B2 (zh)
CN (1) CN110110114B (zh)
WO (1) WO2020206914A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112766146A (zh) * 2021-01-14 2021-05-07 黄河勘测规划设计研究院有限公司 一种基于多源数据的水库动态监控系统及方法
CN112950619A (zh) * 2021-03-25 2021-06-11 征图智能科技(江苏)有限公司 基于可视化仿真的视觉检测方法

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110110114B (zh) * 2019-04-11 2024-05-03 平安科技（深圳）有限公司 多源对地观测图像处理的可视化方法、装置及存储介质
CN112132284B (zh) * 2020-08-24 2022-04-15 中国科学院空天信息创新研究院 一种对地观测知识枢纽系统
CN112036548A (zh) * 2020-08-28 2020-12-04 济南浪潮高新科技投资发展有限公司 一种深度学习装置及深度学习应用方法
CN118642970A (zh) * 2024-08-14 2024-09-13 浙江大华技术股份有限公司 可视化验证方法、电子设备和计算机可读存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090112666A1 (en) * 2007-10-31 2009-04-30 Chang Jie Guo Apparatus and method for generating a monitoring view of an executable business process
CN101960400A (zh) * 2008-02-26 2011-01-26 Keba股份公司 机器流程的配置
CN106874183A (zh) * 2016-07-05 2017-06-20 阿里巴巴集团控股有限公司 业务异常检测方法及装置
CN107590070A (zh) * 2017-08-18 2018-01-16 深圳怡化电脑股份有限公司 一种业务流程调试方法及装置
CN110110114A (zh) * 2019-04-11 2019-08-09 平安科技（深圳）有限公司 多源对地观测图像处理的可视化方法、装置及存储介质

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102006043910B4 (de) * 2006-09-19 2010-02-25 Siemens Ag Ergebnisfilter und Verfahren zur Selektion der Ergebnisdaten einer Applikation zur automatischen Mustererkennung
CN1959717B (zh) * 2006-10-09 2011-09-28 北京道达天际软件技术有限公司 订单驱动的海量遥感数据集群化预处理系统及其方法
JP4358262B2 (ja) 2007-07-02 2009-11-04 ザイオソフト株式会社 医療画像処理装置およびプログラム
CN101387958B (zh) * 2008-10-20 2011-06-15 东软集团股份有限公司 一种图像数据的处理方法及装置
KR20110014864A (ko) * 2009-08-06 2011-02-14 주식회사 제니텀 데이터 플로우 프레임워크에 기반한 비디오 데이터의 3차원 재구성 시스템
US20160203178A1 (en) * 2015-01-12 2016-07-14 International Business Machines Corporation Image search result navigation with ontology tree
CN105389589B (zh) * 2015-11-06 2018-09-18 北京航空航天大学 一种基于随机森林回归的胸腔x光片肋骨检测方法
CN107423053B (zh) * 2017-06-15 2020-08-21 东莞理工学院 一种遥感图像处理的web化模型封装与分布式处理方法
CN107292841A (zh) * 2017-06-16 2017-10-24 石福荣 模式化图像处理技术
CN109165272B (zh) 2018-08-21 2021-02-19 东北大学 面向高分遥感影像的地理信息可视化分析系统

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090112666A1 (en) * 2007-10-31 2009-04-30 Chang Jie Guo Apparatus and method for generating a monitoring view of an executable business process
CN101960400A (zh) * 2008-02-26 2011-01-26 Keba股份公司 机器流程的配置
CN106874183A (zh) * 2016-07-05 2017-06-20 阿里巴巴集团控股有限公司 业务异常检测方法及装置
CN107590070A (zh) * 2017-08-18 2018-01-16 深圳怡化电脑股份有限公司 一种业务流程调试方法及装置
CN110110114A (zh) * 2019-04-11 2019-08-09 平安科技（深圳）有限公司 多源对地观测图像处理的可视化方法、装置及存储介质

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112766146A (zh) * 2021-01-14 2021-05-07 黄河勘测规划设计研究院有限公司 一种基于多源数据的水库动态监控系统及方法
CN112766146B (zh) * 2021-01-14 2022-11-29 黄河勘测规划设计研究院有限公司 一种基于多源数据的水库动态监控系统及方法
CN112950619A (zh) * 2021-03-25 2021-06-11 征图智能科技(江苏)有限公司 基于可视化仿真的视觉检测方法

Also Published As

Publication number Publication date
JP7264989B2 (ja) 2023-04-25
CN110110114A (zh) 2019-08-09
CN110110114B (zh) 2024-05-03
JP2021532497A (ja) 2021-11-25

Similar Documents

Publication Publication Date Title
WO2020206914A1 (zh) 多源对地观测图像处理的可视化方法、装置及存储介质
US8896595B2 (en) System, apparatus, and method of modifying 2.5D GIS data for a 2D GIS system
US7587685B2 (en) Data exploration system
US9892531B2 (en) Chart data-binding design time user experience with dynamic sample generation
US8751969B2 (en) Information processor, processing method and program for displaying a virtual image
US9787752B2 (en) Hotspot editor for a user interface
US20150161822A1 (en) Location-Specific Digital Artwork Using Augmented Reality
CN108628608A (zh) 一种数据可视化的方法及装置
CN104850388B (zh) 网页绘制方法及装置
US9105094B2 (en) Image layers navigation
KR20080107444A (ko) 그래프형 다이어그램을 편집하기 위한 2차원 트리들의 동적레이아웃 방법, 영구적 변경 보존 방법, 및 컴퓨터 프로그램 제품
EP3259679B1 (en) An automatically invoked unified visualization interface
US20140244219A1 (en) Method of creating a pipe route line from a point cloud in three-dimensional modeling software
CN104484163B (zh) 一种基于统一建模环境的异构模型转换方法
CN104504171B (zh) 一种基于统一建模环境的建模方法
CN110955803A (zh) 基于图数据库的关系图谱的绘制方法、装置、设备及介质
CN114138269A (zh) 业务数据的可视化应用构建方法和装置
CN115294308A (zh) 基于深度学习的增强现实辅助装配作业引导系统
US8576223B1 (en) Multiple label display for 3D objects
KR102282284B1 (ko) 오브젝트 집합의 시각적 표현을 제공하는 방법, 장치, 기기 및 저장 매체
US20110161846A1 (en) Highlighting changes on computer diagrams
KR20120075626A (ko) 웹 기반 서비스에서의 전자해도 처리 장치 및 방법
CN115794980A (zh) 一种数据展示方法、设备、装置及介质
JP4960188B2 (ja) 画面遷移図の表示方法およびシステム
CN117634605B (zh) 一种基于知识图谱的岭南文化可视化展示方法及系统

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19924094

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2021506308

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19924094

Country of ref document: EP

Kind code of ref document: A1