A data rendering method applied to satellite tile data

Abstract

The application discloses a kind of data rendering methods applied to satellite tile data, it is related to image processing technical field, effectively improve the accuracy of image rendering.The application labels the scene image data collected by unmanned aerial vehicle as ground scene image data, and labels the scene image data collected by satellite as ground-air scene image data, and then divides the ground-air scene image data into a plurality of tile region image data, generates a plurality of scene object three-dimensional images of different time nodes according to each ground scene image data, and then matches and splices the scene object three-dimensional images of each different time node and tile region image data according to the collection time, labels dynamic elements or static elements in each scene object three-dimensional image, sets a plurality of ground-air scene image models on tile region image data, matches and maps dynamic elements and static elements with each tile region image data, and then obtains complete scene visual space image.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a data rendering method applied to satellite tile data. Background Art

[0002] Satellite image rendering refers to the process of processing and optimizing raw data obtained from satellites to produce clearer and more informative images. Satellites acquire image data of the Earth's surface through various sensors. These data are usually digital and include information from different spectral bands.

[0003] During satellite image rendering, the raw data is processed and optimized through various digital image processing techniques, such as color enhancement, contrast adjustment, resolution enhancement, mosaicking, and geometric correction, resulting in higher image quality and clearer visibility of terrain features. These processed satellite images can be used in a variety of fields, such as map production, natural resource management, environmental monitoring, and urban planning.

[0004] Existing satellite image rendering has the following defects:

[0005] Color enhancement: Improves image quality by enhancing contrast and color saturation, making ground features clearer. However, satellite images may have geometric correction errors during transmission and processing, resulting in image distortion or position offset.

[0006] Specific target extraction: Through target detection and recognition algorithms, specific targets such as buildings and roads are extracted for automated analysis. The amount of raw satellite data is large, and processing and rendering require a lot of computing resources and time, and the processing complexity is relatively high.

[0007] Therefore, how to improve the accuracy of data processing while ensuring the accuracy of image data is a difficulty in the existing technology. To this end, a data rendering method applied to satellite tile data is provided. Summary of the Invention

[0008] In order to solve the above technical problems, the purpose of the present invention is to provide a data rendering method applied to satellite tile data.

[0009] In order to achieve the above object, the present invention provides the following technical solutions:

[0010] A data rendering method applied to satellite tile data comprises the following steps:

[0011] Step S1: several drones are set up in the target scene, and each drone is connected to a satellite for communication, so that each drone and the satellite simultaneously collect scene image data of the target scene;

[0012] Step S2: Label the scene image data collected by the drone as ground scene image data, label the scene image data collected by the satellite as ground-air scene image data, and then divide the ground-air scene image data into a number of tile area image data;

[0013] Step S3: Establishing a three-dimensional spatial coordinate system, generating a number of three-dimensional images of scene objects at different time nodes based on each ground scene image data, and then matching and splicing the three-dimensional images of scene objects at different time nodes with the tile area image data according to the acquisition time;

[0014] Step S4: Annotate the three-dimensional image of each scene object with dynamic elements or static elements, adopt the process of step S3 to generate the three-dimensional image of the scene object, set several ground-air scene image models on the tile area image data, and annotate the motion trajectories of the ground-air scene image models at different time nodes, match and map the dynamic elements and static elements with the image data of each tile area, and then obtain a complete scene visualization spatial image.

[0015] Furthermore, the scene image data collection process includes:

[0016] Num drones are set up in a target scene, each of which is equipped with a wireless signal transmission device, a GPS positioning device, and multiple cameras. The target scene can be a mountainous area, a forest, a city, or other location.

[0017] Each drone is connected to a satellite through a wireless signal transmission device, and the satellite is equipped with a data processing unit and a variety of remote sensing devices;

[0018] The satellite then assigns a number to each drone based on the communication results and obtains the initial geographic location of each drone through the GPS positioning device;

[0019] The satellite generates a data collection command and sends it to each drone simultaneously. After receiving the data collection command, each drone sends a reception confirmation prompt to the satellite. Then the drone flies randomly in the target scene and collects scene image data of the target scene through the camera. The satellite collects scene image data of the target scene through various remote sensing devices. At the same time, the initial geographical location of each drone is used as the starting point, and the flight path of each drone is obtained through the GPS positioning device.

[0020] The satellite sets a corresponding drone number for each scene image data and flight route, and integrates the scene image data with the same number to obtain a scene dataset.

[0021] Furthermore, the processing of the ground scene image data includes:

[0022] The data processing device establishes a three-dimensional spatial coordinate system, maps the ground scene image data and the flight path in each scene data set onto the three-dimensional coordinate system, sets a number of time marks on the flight path according to the generation time of the flight path, matches the time marks on the ground scene image data with the time marks on the flight path, and arranges the ground scene image data on the flight path in sequence according to the matching results;

[0023] Each ground scene image data is gray-scaled, and the regional pixel values ​​of each gray-scale pixel in the ground scene image data are averaged.

[0024] Furthermore, the process of averaging the regional pixel values ​​includes:

[0025] A square pixel frame is set to select several grayscale pixels in the ground scene image data in turn, and the pixel values ​​of each grayscale pixel in the square pixel frame are accumulated and averaged, and the average value is assigned to each grayscale pixel in the square pixel frame as a new pixel value, and so on, until the square pixel frame selects the last grayscale pixel in the ground scene image data.

[0026] Furthermore, the process of establishing the three-dimensional image of the scene object includes:

[0027] Setting a pixel value threshold, comparing the pixel value of the grayscale pixel in each ground scene image data with the pixel value threshold, and setting a label for the grayscale pixel if the pixel value of the grayscale pixel is greater than or equal to the pixel value threshold;

[0028] If the pixel value of the grayscale pixel is less than the pixel value threshold, no operation is performed;

[0029] Connect adjacent labeled grayscale pixels in sequence, and then divide several scene elements in the ground scene image data;

[0030] Match the ground scene image data with the same time stamp in scene datasets with different numbers. If the two have the same scene elements, copy the corresponding scene elements in the two and classify them for storage. If the two do not have the same scene elements, do nothing.

[0031] When the ground scene image data with the same time mark is matched, the classified and stored scene elements will be overlapped and mapped to obtain the corresponding three-dimensional image of the scene object, and the corresponding time mark will be set;

[0032] Repeat the above operation to obtain the three-dimensional images of the same scene object at different time nodes.

[0033] Furthermore, the process of matching and splicing the three-dimensional image of the scene object and the tile region image data according to the acquisition time includes:

[0034] The flight paths and tile image data of each drone are simultaneously mapped onto the 3D spatial model. The positions on each flight path are matched according to the time stamps of each tile image data, and the real-time positions of each drone are then marked on each tile image data.

[0035] Based on the ground scene image data captured by each drone at different time points, the direction and distance of each scene element in the ground scene image data relative to the drone are determined, and then the corresponding scene element position is marked in the corresponding drone's surrounding position in each tile area image data according to the determination result;

[0036] Match the scene target 3D image with the same time tag with the tile area image data, and then map the scene target 3D image to the scene element position. Repeat the above operation until the scene target 3D images at all time nodes are mapped to the tile area image data.

[0037] Furthermore, the process of marking dynamic elements or static elements includes:

[0038] Arranging tile region image data with the same number according to time sequence, and overlapping and mapping the same three-dimensional images of scene objects on the tile region image data in adjacent sequences;

[0039] If all pixels on the three-dimensional images of the two scene objects completely overlap, it is determined that the corresponding scene objects have not moved within the corresponding time interval;

[0040] If more than x pixels in the three-dimensional images of two scene objects do not overlap, then the corresponding scene objects are determined to have moved within the corresponding time interval, where x is 1 / 10 of the total number of pixels in the three-dimensional images of the corresponding scene objects;

[0041] It is determined whether a three-dimensional image of a scene object in the tile area image data has displacement within three consecutive time intervals, and then the three-dimensional image of the scene object is marked as a dynamic element or a static element.

[0042] Furthermore, the process of establishing a complete scene visualization spatial image includes:

[0043] Setting a number of ground-air scene image models on tile area image data at different time nodes, overlapping and mapping the same ground-air scene image models on tile area image data with the same number but different time tags, and then obtaining the motion trajectory of each ground-air scene image model;

[0044] Match and map the dynamic and static elements with the motion trajectories of each ground-air scene image model. If the dynamic and static elements match the motion trajectories of each ground-air scene image model by more than 80%, overlap and map the corresponding three-dimensional image of the scene object with the ground-air scene image model.

[0045] Otherwise, the scene image data collected by the UAV and satellite are re-acquired simultaneously until each ground-air scene image model has an overlapping mapped three-dimensional image of the scene object;

[0046] The tile area image data are spliced ​​in sequence according to the splitting order to obtain a complete scene visualization spatial image.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] The present invention generates a plurality of three-dimensional images of scene objects at different time nodes based on various ground scene image data, and then matches and splices the three-dimensional images of scene objects at different time nodes with the tile area image data according to the acquisition time, annotates the three-dimensional images of each scene object with dynamic elements or static elements, sets a plurality of ground-air scene image models on the tile area image data, matches and maps the dynamic elements and static elements with the various tile area image data, and thus obtains a complete scene visualization spatial image, which effectively improves the accuracy of data processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention.

[0050] Figure 1 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0051] To make the objectives, technical solutions, and advantages of the present invention more apparent, the technical solutions of the present invention will be described in detail below. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other implementations obtained by those of ordinary skill in the art without inventive effort are within the scope of protection of the present invention.

[0052] Example 1

[0053] like Figure 1 As shown, a data rendering method applied to satellite tile data includes the following steps:

[0054] Step S1: several drones are set up in the target scene, and each drone is connected to a satellite for communication, so that each drone and the satellite simultaneously collect scene image data of the target scene;

[0055] Step S2: Label the scene image data collected by the drone as ground scene image data, label the scene image data collected by the satellite as ground-air scene image data, and then divide the ground-air scene image data into a number of tile area image data;

[0056] Step S3: Establishing a three-dimensional spatial coordinate system, generating a number of three-dimensional images of scene objects at different time nodes based on each ground scene image data, and then matching and splicing the three-dimensional images of scene objects at different time nodes with the tile area image data according to the acquisition time;

[0057] Step S4: Annotate the three-dimensional image of each scene object with dynamic elements or static elements, adopt the process of step S3 to generate the three-dimensional image of the scene object, set several ground-air scene image models on the tile area image data, and annotate the motion trajectories of the ground-air scene image models at different time nodes, match and map the dynamic elements and static elements with the image data of each tile area, and then obtain a complete scene visualization spatial image.

[0058] Example 2

[0059] This embodiment further limits the embodiment 1, and the step S1 is implemented by the following process:

[0060] Num drones are set up in a target scene, each of which is equipped with a wireless signal transmission device, a GPS positioning device, and multiple cameras. The target scene can be a mountainous area, a forest, a city, or other location.

[0061] Each drone is connected to a satellite through a wireless signal transmission device. The satellite is equipped with a data processing unit and a variety of remote sensing devices, including optical sensors, multispectral sensors, etc.

[0062] Then the satellite sets a number for each drone based on the communication results, where the numbers are a1, a2, a3, ..., a Num , obtain the initial geographical location of each UAV through the GPS positioning device;

[0063] The satellite generates a data collection command and sends it to each drone simultaneously. After receiving the data collection command, each drone sends a reception confirmation prompt to the satellite. Then the drone flies randomly in the target scene and collects scene image data of the target scene through the camera. The satellite collects scene image data of the target scene through various remote sensing devices. At the same time, the initial geographical location of each drone is used as the starting point, and the flight path of each drone is obtained through the GPS positioning device.

[0064] It should be noted that, in the process of collecting scene image data by satellites and drones, time tags are set for the collected scene image data according to their current time nodes;

[0065] Furthermore, the satellite generates a collection stop command and sends it to the UAV, and then each UAV sends the scene image data collected during the flight to the satellite via a wireless signal transmission device;

[0066] The satellite sets a corresponding drone number for each scene image data and flight route, and integrates the scene image data with the same number to obtain a scene dataset.

[0067] Example 3

[0068] This embodiment further limits the embodiment 1, and the step S2 is implemented by the following process:

[0069] The satellite sends the scene data set and the scene image data collected by the satellite to the data processing device, and the data processing device labels the scene image data collected by the satellite as ground-to-air scene image data and labels the scene data in the scene data set as ground scene image data;

[0070] The data processing device divides all ground-air scene image data into N tile area image data of the same size, where N is a natural number greater than 0.

[0071] Example 4

[0072] This embodiment further limits the embodiment 1, and the step S3 is implemented by the following process:

[0073] The data processing device establishes a three-dimensional spatial coordinate system, maps the ground scene image data and the flight path in each scene data set onto the three-dimensional coordinate system, sets a number of time marks on the flight path according to the generation time of the flight path, matches the time marks on the ground scene image data with the time marks on the flight path, and arranges the ground scene image data on the flight path in sequence according to the matching results;

[0074] Grayscale processing is performed on each ground scene image data, and regional pixel values ​​of each grayscale pixel in the ground scene image data are averaged;

[0075] The process of regional pixel value averaging includes: setting a square pixel frame to sequentially select a plurality of grayscale pixels in the ground scene image data, accumulating the pixel values ​​of each grayscale pixel in the square pixel frame to obtain an average value, assigning the average value to each grayscale pixel in the square pixel frame as a new pixel value, and so on, until the square pixel frame selects the last grayscale pixel in the ground scene image data;

[0076] Setting a pixel value threshold, comparing the pixel value of the grayscale pixel in each ground scene image data with the pixel value threshold, and setting a label for the grayscale pixel if the pixel value of the grayscale pixel is greater than or equal to the pixel value threshold;

[0077] If the pixel value of the grayscale pixel is less than the pixel value threshold, no operation is performed;

[0078] Connect adjacent labeled grayscale pixels in sequence, and then divide several scene elements in the ground scene image data;

[0079] Match the ground scene image data with the same time stamp in scene datasets with different numbers. If the two have the same scene elements, copy the corresponding scene elements in the two and classify them for storage. If the two do not have the same scene elements, do nothing.

[0080] Since the ground scene image data corresponding to the same scene element is taken by different drones at different angles, the angles of the target scene in the ground scene image data taken at the same time are different;

[0081] When the ground scene image data with the same time mark is matched, the classified and stored scene elements will be overlapped and mapped to obtain the corresponding three-dimensional image of the scene object, and the corresponding time mark will be set;

[0082] Repeat the above operation to obtain the three-dimensional images of the same scene object at different time nodes.

[0083] Furthermore, the flight paths of each drone and the tile area image data are simultaneously mapped onto the 3D spatial model. The positions on each flight path are matched according to the time stamps on each tile area image data, and the real-time positions of each drone are then marked on each tile area image data.

[0084] Based on the ground scene image data captured by each drone at different time points, the direction and distance of each scene element in the ground scene image data relative to the drone are determined, and then the corresponding scene element position is marked in the corresponding drone's surrounding position in each tile area image data according to the determination result;

[0085] Match the scene target 3D image with the same time tag with the tile area image data, and then map the scene target 3D image to the scene element position. Repeat the above operation until the scene target 3D images at all time nodes are mapped to the tile area image data.

[0086] Example 5

[0087] This embodiment further limits the embodiment 1, and the step S4 is implemented by the following process:

[0088] Arranging tile region image data with the same number according to time sequence, and overlapping and mapping the same three-dimensional images of scene objects on the tile region image data in adjacent sequences;

[0089] If all pixels on the three-dimensional images of the two scene objects completely overlap, it is determined that the corresponding scene objects have not moved within the corresponding time interval;

[0090] If more than x pixels in the three-dimensional images of two scene objects do not overlap, then the corresponding scene objects are determined to have moved within the corresponding time interval, where x is 1 / 10 of the total number of pixels in the three-dimensional images of the corresponding scene objects;

[0091] If a 3D image of a scene object in the tile area image data has displacement within three consecutive time intervals, the 3D image of the scene object at the corresponding time node is marked as a dynamic element;

[0092] If there is no three-dimensional image of a scene object in the tile area image data that has displacement within three consecutive time intervals, the three-dimensional image of the scene object at the corresponding time node is marked as a static element;

[0093] Furthermore, the process of generating a three-dimensional image of the scene object in step S3 is used to set a plurality of ground-air scene image models on the tile region image data at different time nodes, and the same ground-air scene image models on the tile region image data with the same number but different time tags are overlapped and mapped to obtain the motion trajectory of each ground-air scene image model;

[0094] Match and map the dynamic and static elements with the motion trajectories of each ground-air scene image model. If the dynamic and static elements match the motion trajectories of each ground-air scene image model by more than 80%, overlap and map the corresponding three-dimensional image of the scene object with the ground-air scene image model.

[0095] Otherwise, the scene image data collected by the UAV and satellite are re-acquired simultaneously until each ground-air scene image model has an overlapping mapped three-dimensional image of the scene object;

[0096] The tile area image data are spliced ​​in sequence according to the splitting order to obtain a complete scene visualization spatial image.

[0097] The above embodiments are only used to illustrate the technical method of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A data rendering method applied to satellite tile data, characterized in that: The following steps are involved: Step S1: several drones are set up in the target scene, and each drone is connected to a satellite for communication, so that each drone and the satellite simultaneously collect scene image data of the target scene; Step S2: Label the scene image data collected by the drone as ground scene image data, label the scene image data collected by the satellite as ground-air scene image data, and then divide the ground-air scene image data into a number of tile area image data; Step S3: Establishing a three-dimensional spatial coordinate system, generating a number of three-dimensional images of scene objects at different time nodes based on each ground scene image data, and then matching and splicing the three-dimensional images of scene objects at different time nodes with the tile area image data according to the acquisition time; Step S4: Annotate the three-dimensional image of each scene object with dynamic elements or static elements, adopt the process of step S3 to generate the three-dimensional image of the scene object, set several ground-air scene image models on the tile area image data, and annotate the motion trajectories of the ground-air scene image models at different time nodes, match and map the dynamic elements and static elements with the image data of each tile area, and then obtain a complete scene visualization spatial image.

2. A data rendering method applied to satellite tile data according to claim 1, characterized in that: The process of collecting scene image data includes: The drones are provided with a wireless signal transmission device, a GPS positioning device and a plurality of cameras, and each drone is connected to a satellite for communication via the wireless signal transmission device. The satellite is provided with a data processing unit and a plurality of remote sensing devices; The satellite then assigns a number to each drone based on the communication results and obtains the initial geographic location of each drone through the GPS positioning device; The satellite generates a data collection command and sends it to each drone simultaneously. After receiving the data collection command, each drone sends a reception confirmation prompt to the satellite. Then the drone flies randomly in the target scene and collects scene image data of the target scene through the camera. The satellite collects scene image data of the target scene through various remote sensing devices. At the same time, the initial geographical location of each drone is used as the starting point, and the flight path of each drone is obtained through the GPS positioning device. The satellite sets a corresponding drone number for each scene image data and flight route, and integrates the scene image data with the same number to obtain a scene dataset.

3. A data rendering method applied to satellite tile data according to claim 2, characterized in that: The process of processing the ground scene image data includes: Establish a three-dimensional spatial coordinate system, map the ground scene image data and flight routes in each scene dataset onto the three-dimensional coordinate system, set several time annotations on the flight routes based on the generation time of the flight routes, and then match the time annotations on the ground scene image data with the time annotations on the flight routes. Based on the matching results, arrange the ground scene image data in sequence on the flight routes. Each ground scene image data is gray-scaled, and the regional pixel values ​​of each gray-scale pixel in the ground scene image data are averaged.

4. A data rendering method applied to satellite tile data according to claim 3, characterized in that: The process of averaging the regional pixel values ​​includes: A square pixel frame is set to select several grayscale pixels in the ground scene image data in turn, and the pixel values ​​of each grayscale pixel in the square pixel frame are accumulated and averaged, and the average value is assigned to each grayscale pixel in the square pixel frame as a new pixel value until the square pixel frame selects the last grayscale pixel in the ground scene image data.

5. A data rendering method applied to satellite tile data according to claim 4, characterized in that: The process of creating a three-dimensional image of a scene object includes: Setting a pixel value threshold, comparing the pixel value of each grayscale pixel in the ground scene image data with the pixel value threshold, and setting a label for each grayscale pixel according to the comparison result; Connect adjacent labeled grayscale pixels in sequence, and then divide several scene elements in the ground scene image data; Match the ground scene image data with the same time mark in scene data sets with different numbers. If the two have the same scene elements, copy the corresponding scene elements in the two and classify them for storage. If the two do not have the same scene elements, do nothing. After the matching is completed, the classified and stored scene elements will be overlapped and mapped to obtain the corresponding three-dimensional image of the scene object, and the corresponding time mark will be set.

6. A data rendering method for satellite tile data according to claim 5, characterized in that: The process of matching and splicing the 3D image of the scene object and the tile area image data according to the acquisition time includes: The flight paths and tile image data of each drone are simultaneously mapped onto the 3D spatial model. The positions on each flight path are matched according to the time stamps of each tile image data, and the real-time positions of each drone are then marked on each tile image data. Based on the ground scene image data captured by each drone at different time points, the direction and distance of each scene element in the ground scene image data relative to the drone are determined, and then the corresponding scene element position is marked in the corresponding drone's surrounding position in each tile area image data according to the determination result; Match the scene target 3D image with the same time tag with the tile area image data, and then map the scene target 3D image to the scene element position. Repeat the above operation until the scene target 3D images at all time nodes are mapped to the tile area image data.

7. A data rendering method for satellite tile data according to claim 6, characterized in that: The process of annotating dynamic or static elements includes: Arranging tile region image data with the same number according to time sequence, and overlapping and mapping the same three-dimensional images of scene objects on the tile region image data in adjacent sequences; If all pixels on the three-dimensional images of the two scene objects completely overlap, it is determined that the corresponding scene objects have not moved within the corresponding time interval; If more than x pixels in the three-dimensional images of two scene objects do not overlap, then the corresponding scene objects are determined to have moved within the corresponding time interval, where x is 1 / 10 of the total number of pixels in the three-dimensional images of the corresponding scene objects; It is determined whether a three-dimensional image of a scene object in the tile area image data has displacement within three consecutive time intervals, and then the three-dimensional image of the scene object is marked as a static element or a dynamic element.

8. A data rendering method for satellite tile data according to claim 7, characterized in that: The process of establishing a complete scene visualization spatial image includes: Setting a number of ground-air scene image models on tile area image data at different time nodes, overlapping and mapping the same ground-air scene image models on tile area image data with the same number but different time tags, and then obtaining the motion trajectory of each ground-air scene image model; Match and map the dynamic and static elements with the motion trajectories of each ground-air scene image model. If the dynamic and static elements match the motion trajectories of each ground-air scene image model by more than 80%, overlap and map the corresponding three-dimensional image of the scene object with the ground-air scene image model. Otherwise, the scene image data collected by the UAV and satellite are re-acquired simultaneously until each ground-air scene image model has an overlapping mapped three-dimensional image of the scene object; The tile area image data are spliced ​​in sequence according to the splitting order to obtain a complete scene visualization spatial image.

Images