Satellite-borne real-time registration method and system for multi-load large-width image data

Abstract

The application provides a kind of multi-load large width image data spaceborne real-time registration method and system, comprising: receiving multi-load transmission image data, carrying out data analysis work;Make ZYNQ receive multi-load image data and auxiliary data, carry out cache processing respectively;The head and tail data of the load cache carry out 4 corner point pixel rough positioning operation;Carry out latitude and longitude maximum and minimum value calculation;Extract corresponding DEM grid data at the same angle;Carry out frame by frame pixel rough positioning processing without control point, take out the elevation data of this pixel point in DDR, carry out elevation constraint precision positioning iterative solution;The precision positioning solution result of multi-load is carried out DEM grid projection processing respectively;Carry out traversal interpolation processing;Carry out subsequent multi-load multi-channel joint processing;Repeat until load data stops transmission or cache data solution is completed.The application proposes a kind of on-orbit multi-load data real-time gridding registration processing, solves the present situation of multi-load observation simultaneously in different domains.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of embedded image real-time preprocessing, in particular to a multi-load large-width image data satellite real-time registration method and system; more particularly, to a multi-load large-width image data satellite real-time registration method and system based on global DEM data. BACKGROUND

[0002] Remote sensing is an important application of spatial information network, and plays an important role in many fields such as environment, transportation, ocean, agriculture, water conservancy, surveying and mapping, geology, etc. With the development of business needs and the increasing demand for different detection spectral data, satellites are developing towards multi-load and multi-spectral direction. For example, the FY-3 series of satellites in China carry more than ten kinds of loads, and the number of spectral bands reaches thousands. However, with the continuous increase in the number of loads and spectral bands, the amount of data generated is also growing exponentially. The massive multi-spectral satellite remote sensing data not only brings higher quality data sources for remote sensing applications, but also reduces the timeliness of data application.

[0003] In actual application process, considering the real-time and application timeliness, the ground wants to quickly obtain the remote sensing product of the observation area, rather than waiting for the original data to be downloaded from the satellite and then carrying out ground post-processing, so the satellite real-time load data processing has become a problem that needs to be solved urgently. At present, large-width satellites mostly use line array CCD splicing or multi-element detector scanning to realize large-width demand. Considering that the imaging mechanisms of different loads are different, the multi-spectral data is not in the same grid reference during actual imaging detection. This multi-load imaging mechanism brings certain difficulties to the satellite real-time multi-spectral data fusion processing.

[0004] Firstly, there are many loads and spectral bands. When different spectral bands are observed at the same time, the observation points are not on the same grid reference, which brings difficulties to the joint use of multi-spectral bands. Secondly, satellite real-time positioning processing adopts uncontrolled point positioning method, which lacks height constraint, and the positioning accuracy cannot be effectively guaranteed. Thirdly, the massive observation data cannot be cached and slowly post-processed like ground processing. Limited by the storage resources on the satellite, real-time positioning solution and registration are required, which poses a challenge to the satellite processing strategy.

[0005] Aiming at the problem of real-time image data registration of multi-load wide swath imaging satellite, a multi-load wide swath image data real-time registration method based on global DEM data is proposed. The method is mainly based on the storage of global DEM grid data in the form of solid state disk external hanging on the satellite. Then, the latitude and longitude coordinate range of the observation data is quickly calculated within a certain time. The DEM grid data in the observation range is obtained and cached according to the appropriately expanded range of latitude and longitude coordinates. Finally, the multi-load multi-spectral registration grid data is obtained by carrying out the elevation-constrained per-pixel fine positioning calculation and the equal-angle projection and traversal interpolation processing of the cached DEM grid data, which is used for the multi-spectral joint processing in the back end. The problem of real-time registration of multi-load wide swath image data on satellite is solved. SUMMARY

[0006] In view of the defects in the prior art, the purpose of the present application is to provide a.

[0007] According to the multi-load wide swath image data real-time registration method provided by the present application, the following steps are included:

[0008] Step S1: The interface module FPGA receives the multi-load transmitted image data, carries out data analysis, extracts the multi-load image data and auxiliary data, and sends them to ZYNQ after re-packeting;

[0009] Step S2: ZYNQ receives the multi-load image data and auxiliary data and carries out cache processing respectively;

[0010] Step S3: When the image data cache meets the preset amount, the 4-corner pixel coarse positioning operation is carried out on the cache head and tail data of any load;

[0011] Step S4: The maximum and minimum values of the 4-corner pixel coarse positioning results are calculated to obtain the geographical coordinate range of the observation data and to expand it;

[0012] Step S5: The corresponding DEM equal-angle grid data is extracted from the external storage module according to the geographical coordinates of the observation data and stored in the DDR cache;

[0013] Step S6: The observation data is processed by per-frame per-pixel coarse positioning without control points, and the elevation data of the pixel point is obtained in the DDR according to the latitude and longitude results of the coarse positioning, and the fine positioning iteration calculation is carried out with elevation constraint. The fine positioning calculation results of the multi-load are independently cached;

[0014] Step S7: The fine positioning calculation results of the multi-load are respectively subjected to DEM grid projection processing;

[0015] Step S8: respectively on the multi-load DEM grid projection data after the development of the traversal interpolation processing, form a complete DEM grid data;

[0016] Step S9: the data packet after interpolation processing to the processing backend, to carry out subsequent multi-load multi-channel joint processing;

[0017] Repeat steps S1 to S9, until the load data stops transmission or cache data solution is complete.

[0018] Preferably, in the step S2:

[0019] The cache processing adopts the way of circular queue for data storage, taking into account the stability of memory occupation in the on-board software and the convenience of development and use.

[0020] Preferably, in the step S4:

[0021] The geographic coordinate range calculation adopts the way of four-corner point coarse positioning to frame the latitude and longitude grid range of the observation data, reducing the operation resource overhead brought by the per-pixel positioning operation.

[0022] Preferably, in the step S5:

[0023] The DEM data extraction adopts the design of on-board external storage module, and the communication mode adopts the optical fiber interface, meeting the DEM data storage and regional DEM data extraction requirements;

[0024] The DEM data storage adopts the way of external solid state disk, and the DEM grid data resolution is 0.0005°x0.0005°. The design is on-orbit programmable, and the replacement operation of DEM data in any region is realized, meeting the demand of on-orbit dynamic update.

[0025] Preferably, in the step S6:

[0026] The pixel fine positioning operation based on DEM elevation information constraint considers the influence of elevation on positioning accuracy, and makes up for the defects of insufficient plane positioning accuracy under the condition of no control point;

[0027] The pixel fine positioning operation based on DEM elevation information constraint adopts the way of pre-coarse positioning to extract DEM grid data in the observation area and store it to the processing board DDR, avoiding the resource overhead brought by repeated reading of solid state disk, and improving the speed of DEM elevation data acquisition.

[0028] Preferably, in the step S7:

[0029] The multi-load fine positioning solution result projection DEM grid processing carries out the traversal interpolation processing again on the grid data after coarse projection, to obtain the complete multi-load grid data based on DEM registration;

[0030] In-orbit real-time grid registration processing is carried out on different types of satellite-carrying load data, and the current situation of multi-load observation at the same time in different domains is solved, and the processed multi-load grid data can be sent to the backend to carry out different types of band combination interpretation processing.

[0031] Preferably, a satellite-borne embedded architecture of FPGA, ZYNQ and solid state disk storage combination is provided, a preset high-throughput FPGA and a preset high-performance ARM processor are adopted, and a preset large-capacity high-throughput solid state disk combination is adopted, so as to meet the demand of large real-time image throughput and high computing resource, and at the same time, the solid state disk storage mode is utilized, real-time DEM grid data of an observation area is acquired, and multi-load per-pixel geometric accurate positioning processing is carried out.

[0032] According to the application, a multi-load large-width image data satellite-borne real-time registration system is provided, which comprises:

[0033] Module M1: the interface module FPGA receives multi-load transmitted image data, carries out data analysis, extracts multi-load image data and auxiliary data, and sends the re-packed data to ZYNQ;

[0034] Module M2: ZYNQ receives multi-load image data and auxiliary data, and carries out cache processing respectively;

[0035] Module M3: when the image data cache meets the preset amount, the 4-corner pixel rough positioning operation is carried out on the cache head and tail data of any load;

[0036] Module M4: the maximum and minimum value calculation processing is carried out on the 4-corner pixel rough positioning result, the geographic coordinate range of the observation data is acquired and is expanded outwardly;

[0037] Module M5: according to the observation data geographic coordinates, the corresponding DEM equal-angle grid data is extracted from the external storage module and is stored in the DDR cache;

[0038] Module M6: the frame-by-frame per-pixel uncontrolled point rough positioning processing is carried out on the observation data, and according to the latitude and longitude result of the rough positioning, the elevation data of the pixel point is taken out from the DDR, the elevation-constrained accurate positioning iteration solution is carried out, and the accurate positioning solution results of the multi-load are independently cached;

[0039] Module M7: the DEM grid projection processing is carried out on the accurate positioning solution results of the multi-load respectively;

[0040] Module M8: the traversal interpolation processing is carried out on the DEM grid projection data of the multi-load respectively, and the complete DEM grid data is formed;

[0041] Module M9: The data set after interpolation processing is packaged and sent to the processing backend to carry out subsequent multi-load multi-channel joint processing;

[0042] Repeat modules M1 to M9 until the load data stops transmission or the cache data is calculated.

[0043] Preferably, in the module M2:

[0044] The cache processing adopts a circular queue mode for data storage, taking into account the stability of memory occupation in the on-board software and the convenience of development and use;

[0045] In the module M4:

[0046] The geographic coordinate range calculation adopts a four-corner point coarse positioning method to frame the latitude and longitude grid range of the observation data, reducing the computational resource overhead caused by the per-pixel positioning operation;

[0047] In the module M5:

[0048] The DEM data extraction adopts a design of on-board external storage module, and the communication mode adopts a fiber interface, meeting the DEM data storage and regional DEM data extraction requirements;

[0049] The DEM data storage adopts a solid state disk external hanging method, and the DEM grid data resolution is 0.0005°x0.0005°. The design is on-orbit programmable, and the DEM data of any region can be replaced, meeting the demand of on-orbit dynamic update;

[0050] In the module M6:

[0051] The pixel fine positioning operation based on DEM elevation information constraint considers the influence of elevation on positioning accuracy, and makes up for the defects of insufficient plane positioning accuracy without control points;

[0052] The pixel fine positioning operation based on DEM elevation information constraint adopts a method of pre-coarse positioning to extract DEM grid data within the observation area range and store it to the processing board DDR, avoiding the resource overhead caused by repeated reading of solid state disk, and improving the speed of DEM elevation data acquisition.

[0053] Preferably, in the module M7:

[0054] The multi-load fine positioning result projection DEM grid processing carries out a traversal interpolation processing again on the coarse projected grid data to obtain complete multi-load grid data based on DEM registration;

[0055] Real-time grid registration processing is carried out on different types of satellite-carrying load data in orbit in real time, the current situation of multi-load observation at the same time in different domains is solved, and the processed multi-load grid data can be sent to the back end to carry out different types of band combination interpretation processing.

[0056] The application provides a satellite-borne embedded architecture of an FPGA, a ZYNQ and a solid state disk storage combination, adopts a preset high-throughput FPGA and a preset high-performance ARM processor, and a preset large-capacity high-throughput solid state disk combination, meets the requirements of large real-time image throughput and high computing resources, and simultaneously uses the solid state disk storage mode to acquire DEM grid data of an observation area in real time and carry out multi-load per-pixel geometric accurate positioning processing.

[0057] Compared with the prior art, the application has the following beneficial effects:

[0058] 1. The application provides a satellite-borne embedded architecture of an FPGA+ZYNQ+solid state disk storage combination, adopts a high-throughput FPGA and a high-performance ARM processor, and a large-capacity high-throughput solid state disk combination, meets the requirements of large real-time image throughput and high computing resources, and simultaneously meets the large-capacity DEM data storage requirement.

[0059] 2. The application provides a method for quickly acquiring the longitude and latitude range of an observation area through coarse positioning of an angle point, and effectively reduces the operation resource consumption caused by per-pixel coarse positioning operation when the observation range is acquired.

[0060] 3. The application provides a real-time grid registration processing of on-orbit multi-load data, and solves the current situation of multi-load observation at the same time in different domains. BRIEF DESCRIPTION OF DRAWINGS

[0061] Other features, objects and advantages of the application will become more apparent from the following detailed description of non-limiting embodiments with reference to the following drawings:

[0062] Fig. 1 FIG. 1 is a flow chart of a satellite-borne real-time registration method for multi-load large-width image data based on global DEM data;

[0063] Fig. 2 FIG. 5 is an example diagram of a circular queue data buffer. DETAILED DESCRIPTION

[0064] The application will be described in detail below with reference to specific embodiments. The following embodiments will help those skilled in the art to further understand the application, but do not limit the application in any form. It should be pointed out that, for those skilled in the art, without departing from the concept of the application, a number of changes and improvements can be made. These all belong to the protection scope of the application.

[0065] Embodiment 1

[0066] The application relates to a multi-load wide image data satellite real-time registration method based on global DEM data, which provides a solution for real-time registration of multi-load data, reduces the consumption of embedded resources, and solves the problem that multi-channel image data cannot be jointly applied in real time due to simultaneous observation in different domains; meanwhile, a satellite embedded architecture of FPGA+ZYNQ+solid state disk storage combination is provided, which not only meets the real-time processing problem of satellite large data, but also meets the demand of satellite large-capacity data storage.

[0067] According to the multi-load wide image data satellite real-time registration method provided by the application, as shown in Figs. 1-2 , the method comprises the following steps:

[0068] Step S1: the interface module FPGA receives multi-load transmitted image data, carries out data analysis, extracts multi-load image data and auxiliary data, and sends the re-packed data to the ZYNQ;

[0069] Step S2: the ZYNQ receives multi-load image data and auxiliary data, and carries out cache processing respectively;

[0070] Specifically, in the step S2, the cache processing adopts a circular queue mode for data storage, and the stability of memory occupation in the satellite software and the convenience of development and use are considered.

[0071] The cache processing adopts a circular queue mode for data storage, and the stability of memory occupation in the satellite software and the convenience of development and use are considered.

[0072] Step S3: when the image data cache meets the preset amount, the cache head and tail data of any load are subjected to 4-corner point pixel rough positioning operation;

[0073] Step S4: the 4-corner point pixel rough positioning result is subjected to maximum and minimum value calculation processing, the geographical coordinate range of the observation data is obtained, and the geographical coordinate range is expanded outward;

[0074] Specifically, in the step S4, the geographical coordinate range calculation adopts a four-corner point rough positioning mode to frame the latitude and longitude grid range of the observation data, so that the operation resource consumption caused by the pixel-by-pixel positioning operation is reduced.

[0075] The geographical coordinate range calculation adopts a four-corner point rough positioning mode to frame the latitude and longitude grid range of the observation data, so that the operation resource consumption caused by the pixel-by-pixel positioning operation is reduced.

[0076] Step S5: according to the geographical coordinates of the observation data, corresponding DEM equal-angle grid data is extracted in the external storage module and stored in the DDR cache;

[0077] Specifically, in the step S5, the DEM data extraction adopts a satellite external storage module design, and the communication mode adopts a fiber interface, so that the DEM data storage and regional DEM data extraction requirements are met.

[0078] The DEM data extraction adopts a satellite external storage module design, and the communication mode adopts a fiber interface, so that the DEM data storage and regional DEM data extraction requirements are met.

[0079] The DEM data storage is realized in a way of solid state disk external connection, the DEM grid data resolution is 0.0005*0.0005, and the on-orbit programmable design is realized, the DEM data of any region is replaced, and the demand of on-orbit dynamic updating is met.

[0080] Step S6: carrying out frame-by-frame and pixel-by-pixel uncontrolled point rough positioning processing on the observation data, simultaneously taking out the elevation data of the pixel point in the DDR according to the longitude and latitude results of rough positioning, carrying out elevation-constrained precise positioning iterative solution, and independently caching the precise positioning solution results of the multiple loads;

[0081] Specifically, in the step S6:

[0082] The pixel precise positioning operation based on DEM elevation information constraint comprehensively considers the influence of elevation on positioning accuracy, and makes up for the defects of insufficient plane positioning accuracy under uncontrolled points.

[0083] The pixel precise positioning operation based on DEM elevation information constraint adopts the mode of pre-rough positioning extraction of DEM grid data in the observation region and storage to the DDR of the processing board, avoids the resource consumption caused by repeated reading of the solid state disk, and improves the speed of DEM elevation data acquisition.

[0084] Step S7: carrying out DEM grid projection processing on the precise positioning solution results of the multiple loads respectively;

[0085] Specifically, in the step S7:

[0086] The DEM grid projection processing of the precise positioning solution results of the multiple loads carries out once again the traversal interpolation processing on the grid data after rough projection, and obtains complete multiple load grid data based on DEM registration;

[0087] Real-time grid registration processing is carried out on different types of load data carried by the satellite in real time, the current situation of multiple load observation at the same time in different domains is solved, and the processed multiple load grid data can be sent to the backend to carry out different types of band combination interpretation processing.

[0088] Step S8: carrying out traversal interpolation processing on the DEM grid projection data of the multiple loads respectively, and forming complete DEM grid data;

[0089] Step S9: sending the data after the interpolation processing to the processing backend in a packet, and carrying out subsequent multiple load multi-channel joint processing;

[0090] Repeat steps S1 to S9 until the load data stops transmission or the cache data solution is completed.

[0091] Specifically, a satellite-borne embedded architecture of FPGA, ZYNQ and solid state disk storage combination is provided, a preset high-throughput FPGA and a preset high-performance ARM processor are adopted, and a preset large-capacity high-throughput solid state disk combination is adopted, so that the demand for large real-time image throughput and high computing resource is met, and at the same time, the solid state disk storage mode is used to obtain DEM grid data of an observation area in real time and carry out per-pixel geometric accurate positioning processing of multiple loads.

[0092] Embodiment 2:

[0093] Embodiment 2 is a preferred example of Embodiment 1, and is used to more specifically illustrate the present application.

[0094] The present application also provides a satellite-borne real-time registration system for multi-load large-width image data, which can be realized by executing the flow steps of the satellite-borne real-time registration method for multi-load large-width image data, that is, the satellite-borne real-time registration method for multi-load large-width image data can be understood by those skilled in the art as a preferred embodiment of the satellite-borne real-time registration system for multi-load large-width image data.

[0095] According to the present application, a satellite-borne real-time registration system for multi-load large-width image data is provided, which comprises:

[0096] Module M1: The interface module FPGA receives multi-load transmitted image data, carries out data analysis, extracts multi-load image data and auxiliary data, and sends the re-packaged data to ZYNQ;

[0097] Module M2: ZYNQ receives multi-load image data and auxiliary data and carries out cache processing respectively;

[0098] Specifically, in the module M2:

[0099] The cache processing adopts a circular queue mode for data storage, taking into account the stability of memory occupation in the satellite-borne software and the convenience of development and use;

[0100] Module M3: When the image data cache meets the preset amount, the 4-corner pixel rough positioning operation is carried out on the head and tail data of any load;

[0101] Module M4: The maximum and minimum value calculation processing of the 4-corner pixel rough positioning result is carried out to obtain the geographical coordinate range of the observation data and to expand it outwardly;

[0102] In the module M4:

[0103] The geographical coordinate range calculation adopts a four-corner rough positioning mode to frame the latitude and longitude grid range of the observation data, so as to reduce the operation resource consumption caused by per-pixel positioning operation;

[0104] Module M5: Extract corresponding DEM gnomonic grid data from the external storage module according to the observed data geographic coordinates and store it in the DDR cache;

[0105] In the module M5:

[0106] The DEM data extraction adopts a satellite-borne external storage module design, and the communication mode adopts a fiber interface to meet the DEM data storage and regional DEM data extraction requirements;

[0107] The DEM data storage is realized in the form of a solid state disk external module, the DEM grid data resolution is 0.0005°x0.0005°, and the design is on-orbit programmable, enabling replacement operations for DEM data in any region to meet the demand for on-orbit dynamic updates;

[0108] Module M6: Perform frame-by-frame and pixel-by-pixel uncontrolled point coarse positioning processing on the observed data, and simultaneously extract the elevation data of the pixel point from the DDR according to the latitude and longitude results of coarse positioning to carry out elevation-constrained fine positioning iterative calculation, and independently cache the fine positioning results of multiple loads;

[0109] In the module M6:

[0110] The pixel fine positioning operation based on DEM elevation information constraint considers the influence of elevation on positioning accuracy, and makes up for the deficiency of insufficient plane positioning accuracy under uncontrolled points;

[0111] The pixel fine positioning operation based on DEM elevation information constraint adopts the method of pre-coarse positioning to extract DEM grid data within the observed region and store it to the DDR of the processing board, avoiding the resource overhead caused by repeated reading of the solid state disk, and at the same time improving the speed of DEM elevation data acquisition.

[0112] Module M7: Perform DEM grid projection processing on the fine positioning results of multiple loads respectively;

[0113] Specifically, in the module M7:

[0114] The DEM grid projection processing of the fine positioning results of multiple loads performs a traversal interpolation processing again on the grid data after coarse projection to obtain complete multi-load grid data based on DEM registration;

[0115] Real-time grid registration processing is carried out on different types of load data carried by the satellite in real time to solve the current situation of multi-load observation at the same time in different domains, and the processed multi-load grid data can be sent to the backend for different types of band combination interpretation processing;

[0116] Module M8: Perform traversal interpolation processing on the DEM grid projection data of multiple loads respectively to form complete DEM grid data;

[0117] Module M9: Sends the interpolated data packets to the processing backend for subsequent multi-load, multi-channel joint processing;

[0118] Repeatedly trigger modules M1 through M9 until the load data stops being transmitted or the cached data is fully processed.

[0119] A spaceborne embedded architecture combining FPGA, ZYNQ, and solid-state disk storage is provided. It adopts a combination of a pre-set high-throughput FPGA, a pre-set high-performance ARM processor, and a pre-set large-capacity high-throughput solid-state disk to meet the requirements of high real-time image throughput and high computing resources. At the same time, it utilizes solid-state disk storage to acquire DEM grid data of the observation area in real time and carry out pixel-by-pixel geometric precision positioning processing for multiple payloads.

[0120] Example 3:

[0121] Example 3 is a preferred example of Example 1, and is used to illustrate the present invention in more detail.

[0122] The purpose of this invention is to address the shortcomings of existing technologies by providing a practical and feasible real-time satellite-borne registration method for multi-payload wide-swath image data based on global DEM data. This method uses a corner coarse positioning approach to quickly obtain the latitude and longitude coordinate range of the observation data. Subsequently, real-time DEM grid projection processing is performed on the pixel-by-pixel positioning results under elevation constraints. This can minimize the massive data caching and processing pressure caused by multiple payloads and multiple spectral bands, and meet the requirements for real-time and rapid multi-payload image registration on satellite.

[0123] This embodiment provides a real-time satellite registration method for multi-payload wide-swath image data based on global DEM data, including the following steps:

[0124] (1) The interface module FPGA receives the wide-width image data transmitted by multiple payloads, performs data parsing, extracts the image data and auxiliary data of multiple payloads, reassembles the data and sends it to ZYNQ (a scalable processing platform of Xilinx, one of the chip series is called ZYNQ, which is a common abbreviation in the embedded processing industry); In order to better illustrate the implementation method and operation process of this invention, the number of payloads is simplified to N here;

[0125] (2) ZYNQ receives multi-payload image data and auxiliary data, and performs buffering processing on each; ZYNQ uses a circular queue to perform buffering operations on various types of data, which can meet the stability of memory usage and the convenience of development and use in spaceborne software. Fig. 2 This can better help explain the storage method used in this step;

[0126] (3) When the image data buffer of ZYNQ meets a certain amount, the 4-corner pixel rough positioning operation is carried out on the head and tail data of any load in the buffer; in order to better illustrate the embodiments and operation process of the application, the certain amount of the image data buffer meeting a certain amount is set as Period, any load in the head and tail data of any load is assumed as load 1, the 4 corners are the first and last observation points of the first frame and the Period frame of the image buffer of load 1, and are counted as pt1-pt4; the pixel rough positioning operation is carried out on pt1-pt4, and (lon1, lat1), (lon2, lat2), (lon3, lat3) and (lon4, lat4) are obtained.

[0127] (4) ZYNQ carries out the longitude and latitude maximum and minimum value calculation processing on the rough positioning results of pt1-pt4, that is, the lon maximum value and minimum value and the lat maximum value and minimum value of (lon1, lat1), (lon2, lat2), (lon3, lat3) and (lon4, lat4) are calculated, and are counted as lon min , lon max , lat min , lat max , and the outer extension Δt (angle, parameter configurable) is obtained, and lon1, lon2, lat1 and lat2 are obtained.

[0128] (5) ZYNQ extracts the corresponding DEM isometric grid data according to the observation data geographic coordinates in the external storage module, and stores it in the DDR buffer; in order to better illustrate the embodiments and operation process of the application, the global DEM grid data resolution is set as 0.0005° (parameter configurable), that is, the global DEM grid data size is 720000*360000; according to lon1, lon2, lat1 and lat2, the local DEM grid data of the observation area is framed in the global DEM grid data, and the size of the local DEM grid data is calculated as: (lon2-lon1) / 0.005, (lat2-lat1) / 0.005, respectively, and is recorded as Width and Height.

[0129] (6) ZYNQ carries out frame-by-frame and pixel-by-pixel rough positioning processing on the observation data without control points. In order to better illustrate the embodiments and operation processes of the present application, taking one observation pixel ptt of the payload 1 as an example, the longitude and latitude results (lont, latt) of the rough positioning of ptt are calculated, the elevation data of the position of the observation pixel ptt in the DEM grid (indi, indj) of the observation area is obtained according to the mode of indi=(lont-lon1) / 0.005, indj=(latt-lat1) / 0.005, and the elevation-constrained precise positioning iteration is carried out. The precise positioning of all pixels in the observation area of the payload 1 to the payload N is sequentially completed, and the results are independently cached, which are buf1 to bufN;

[0130] (7) ZYNQ carries out DEM grid projection processing on the precise positioning results of the multiple payloads. In order to better illustrate the embodiments and operation processes of the present application, taking one pixel in buf1 as an example, assuming that the precise positioning result is (lonj, latj), i=(lonj-lon1) / 0.005, j=(latj-lat1) / 0.005 are calculated, and the pixel data is filled into the index position (i, j) of the DEM grid. All pixels in buf1 are sequentially processed in the same way, and the DEM gridding of the payload 1 is completed. buf2 to bufN are sequentially processed in the same way as buf1, and the DEM gridding of the payload 2 to the payload N is completed. The gridded results are independently cached, which are grid1 to gridN.

[0131] (8) ZYNQ carries out traversal interpolation processing on grid1 to gridN respectively, and carries out interpolation processing in the mode of bilinear interpolation, forming complete grid data grid1 to gridN.

[0132] (9) ZYNQ sends the registered grid1 to gridN data group to the processing back end, carries out subsequent multi-payload and multi-channel joint processing, and repeats the above steps until the payload data stops transmission or the cached data is solved.

[0133] Those skilled in the art know that, in addition to implementing the system provided by the present application and each device, module and unit thereof in the form of pure computer readable program code, the system provided by the present application and each device, module and unit thereof can also be implemented in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers and embedded microcontrollers, etc. by logically programming the method steps to achieve the same functions. Therefore, the system provided by the present application and each device, module and unit thereof can be considered as a hardware component, and the devices, modules and units included therein for achieving various functions can also be considered as structures within the hardware component; the devices, modules and units for achieving various functions can also be considered as both software modules implementing methods and structures within hardware components.

[0134] The specific embodiments of the present application are described above. It needs to be understood that the present application is not limited to the specific embodiments described above, and various changes or modifications can be made by those skilled in the art within the scope of the claims, which does not affect the essential content of the present application. The embodiments of the present application and the features in the embodiments can be combined with each other in any manner without conflict.

Claims

1. A multi-load large-width image data spaceborne real-time registration method, characterized in that, Comprise: Step S1: the interface module FPGA receives the multi-load transmission image data, carries out data analysis work, extracts multi-load image data and auxiliary data, and sends to ZYNQ after repackaging; Step S2: let ZYNQ receive multi-load image data and auxiliary data, respectively carry out cache processing; Step S3: when the image data cache meets the preset amount, the 4 corner point pixel rough positioning operation is carried out on the cache head and tail data of any one load; Step S4: the maximum and minimum value calculation processing is carried out on the 4 corner point pixel rough positioning result, the geographical coordinate range of the observation data is obtained and is expanded outwardly; Step S5: according to the observation data geographical coordinates, the corresponding DEM grid data is extracted in the external storage module, and is stored in the DDR cache; Step S6: the frame-by-frame and pixel-by-pixel uncontrolled point rough positioning processing is carried out on the observation data, and the elevation data of the corresponding pixel points is taken out in the DDR according to the longitude and latitude results of rough positioning, the elevation constrained fine positioning iterative solution is carried out, and the fine positioning solution results of multi-load are independently cached; Step S7: the DEM grid projection processing is carried out on the fine positioning solution results of multi-load respectively; Step S8: the traversal interpolation processing is carried out on the data after DEM grid projection of multi-load respectively, and the complete DEM grid data is formed; Step S9: the data after interpolation processing is packaged and sent to the processing backend, and subsequent multi-load multi-channel joint processing is carried out; Repeat steps S1 to S9 until the load data stops transmission or the cache data solution is completed.

2. The method of claim 1, wherein, In the step S2: The cache processing adopts the way of circular queue for data storage, and the stability of memory occupation in satellite-borne software and the convenience of development and use are considered.

3. The method of claim 1, wherein, In the step S5: The DEM data extraction adopts the design of satellite-borne external storage module, and the communication mode adopts optical fiber interface, which meets the DEM data storage and regional DEM data extraction requirements; The DEM data storage is realized by adopting the way of solid state disk external hanging, the DEM grid data resolution is 0.0005°x0.0005°, the design is on-orbit programmable, the replacement operation of DEM data of any region is realized, and the demand of on-orbit dynamic updating is met.

4. The method of claim 1, wherein, In the step S9: The processed multi-load gridding data is sent to the backend to carry out different types of band combination interpretation processing.

5. A multi-load large swath image data on-board real-time registration system, characterized in that, Comprise: Module M1: the interface module FPGA receives the multi-load transmission image data, carries out data analysis work, extracts multi-load image data and auxiliary data, and sends to ZYNQ after repackaging; Module M2: let ZYNQ receive multi-load image data and auxiliary data, respectively carry out cache processing; Module M3: when the image data cache meets the preset amount, the 4 corner point pixel rough positioning operation is carried out on the cache head and tail data of any one load; Module M4: the maximum and minimum value calculation processing is carried out on the 4 corner point pixel rough positioning result, the geographical coordinate range of the observation data is obtained and is expanded outwardly; Module M5: according to the observation data geographical coordinates, the corresponding DEM grid data is extracted in the external storage module, and is stored in the DDR cache; Module M6: Performs frame-by-frame, pixel-by-pixel coarse localization processing on the observation data. Simultaneously, based on the latitude and longitude results of the coarse localization, it retrieves the elevation data of the corresponding pixels in DDR, performs fine localization iterative calculation with elevation constraints, and independently caches the fine localization calculation results of multiple loads. Module M7: Performs DEM mesh projection processing on the fine positioning solution results of multiple loads respectively; Module M8: Performs traversal interpolation processing on the projected data of the DEM grid under multiple loads to form complete DEM grid data; Module M9: Sends the interpolated data packets to the processing backend for subsequent multi-load, multi-channel joint processing; Repeatedly trigger modules M1 through M9 until the load data stops being transmitted or the cached data is fully processed.

6. The multi-payload wide-swath image data spaceborne real-time registration system according to claim 5, characterized in that: In module M2: The cache processing uses a circular queue for data storage, balancing the stability of memory usage in the onboard software with the convenience of development and use; In module M5: DEM data extraction adopts a spaceborne external storage module design and uses a fiber optic interface for communication, which meets the requirements for DEM data storage and regional DEM data extraction. The DEM data storage is implemented using an external solid-state drive. The DEM grid data resolution is 0.0005°×0.0005°. It is designed to be programmable in orbit, enabling the replacement of DEM data in any area and meeting the requirements for dynamic updates in orbit.

7. The multi-payload wide-swath image data spaceborne real-time registration system according to claim 5, characterized in that: In module M9: The processed multi-load gridded data is sent to the backend for different types of band combination interpretation processing.

Images