CN111754397B

CN111754397B - Remote sensing image embedding method and device, electronic equipment and storage medium

Info

Publication number: CN111754397B
Application number: CN201910614757.2A
Authority: CN
Inventors: 徐翔; 曹子奇; 李聪
Original assignee: Beijing Sensetime Technology Development Co Ltd
Current assignee: Beijing Sensetime Technology Development Co Ltd
Priority date: 2019-07-09
Filing date: 2019-07-09
Publication date: 2023-03-21
Anticipated expiration: 2039-07-09
Also published as: CN111754397A

Abstract

The invention discloses a remote sensing image embedding method, a remote sensing image embedding device, electronic equipment and a storage medium, wherein each processing node is respectively responsible for embedding a remote sensing image of a target area in a distributed system, so that the remote sensing image embedding task of a large-range area can be quickly completed, and automatic and efficient image embedding is realized; and the effective area of the remote sensing image can be accurately determined based on the second geographic range and the cloud cover rate of the remote sensing image, and the target remote sensing image is selected according to the effective area of the remote sensing image, so that the quality of the embedded image can be effectively improved.

Description

Remote sensing image embedding method and device, electronic equipment and storage medium

Technical Field

The present disclosure relates to the field of remote sensing image processing technologies, and in particular, to a remote sensing image embedding method and apparatus, an electronic device, and a storage medium.

Background

Image mosaic refers to a process of splicing multi-scene adjacent remote sensing images into a large-range seamless image under certain mathematical basic control. Because the data shot by the satellite are all in a strip shape, if a required area spans a plurality of strips, splicing and inlaying processing needs to be carried out on the satellite data, the finally obtained result area is ensured to be integrated, and the splicing of a plurality of data cannot be seen.

The traditional method mostly adopts manual semi-automatic processing in Photoshop and other software, but for processing of large-scale (such as nationwide) images, the manual mode is time-consuming and labor-consuming, and the mosaic efficiency is low.

Disclosure of Invention

The disclosure provides a remote sensing image embedding method and device, electronic equipment and a storage medium.

In a first aspect, a remote sensing image mosaic method is provided, and is applied to a distributed system, where the distributed system includes multiple processing nodes, and the multiple processing nodes execute the mosaic method in parallel, and the mosaic method includes:

searching a plurality of remote sensing images matched with a target area to be embedded and remote sensing data respectively corresponding to the remote sensing images;

determining effective areas corresponding to the plurality of remote sensing images respectively based on the first geographical range of the target area and the remote sensing data corresponding to the plurality of remote sensing images respectively; the remote sensing data comprises a second geographical range and cloud cover rate of the remote sensing image;

selecting a target remote sensing image from the plurality of remote sensing images for a plurality of times based on the effective areas corresponding to the plurality of remote sensing images respectively;

and determining the embedded image based on the target remote sensing images selected for multiple times.

In one implementation, the determining, based on the first geographic range of the target region and the remote sensing data corresponding to the plurality of remote sensing images, effective areas corresponding to the plurality of remote sensing images respectively includes:

determining a first area of the target area based on a first geographic extent of the target area; determining a second area of an area in each remote sensing image, wherein the second geographical range overlaps with the first geographical range, based on the first geographical range and a second geographical range of each remote sensing image;

and determining the effective area corresponding to each remote sensing image based on the calculated first area, the second area corresponding to each remote sensing image and the cloud cover rate corresponding to each remote sensing image.

In yet another implementation, the determining the effective area corresponding to each remote sensing image based on the calculated first area, the second area corresponding to each remote sensing image, and the cloud cover rate corresponding to each remote sensing image includes:

determining the effective coverage rate of each remote sensing image according to a preset weight parameter and the cloud cover rate corresponding to each remote sensing image;

And determining the effective area corresponding to each remote sensing image according to the effective coverage rate of each remote sensing image, the first area and the second area corresponding to each remote sensing image.

In another implementation, the selecting a target remote sensing image from the plurality of remote sensing images for a plurality of times based on the effective areas corresponding to the plurality of remote sensing images, respectively, includes:

arranging the plurality of remote sensing images according to the sequence of the effective areas corresponding to the plurality of remote sensing images from large to small to obtain an arrangement result of the plurality of remote sensing images;

sequentially selecting a target remote sensing image from the plurality of remote sensing images according to the arrangement result;

the determining the mosaic image based on the target remote sensing images selected for multiple times comprises:

and assigning the pixels of the target area according to the sequentially selected pixels of the target remote sensing image until all the pixels of the target area are assigned.

In another implementation, before assigning values to the pixels of the target area according to the sequentially selected pixels of the target remote sensing image until all the pixels of the target area are determined to be assigned, the method further includes:

Excluding the first remote sensing image from the arrangement result in a case where a second geographical range of the first remote sensing image matching the target area is included in a second geographical range of the second remote sensing image matching the target area.

In another implementation, the searching for a plurality of remote sensing images matched with the target area to be embedded and remote sensing data corresponding to the plurality of remote sensing images respectively includes:

searching a plurality of remote sensing images which accord with the search condition and remote sensing data which respectively correspond to the plurality of remote sensing images;

wherein the search condition includes at least one of the following conditions:

the second geographical range of the remote sensing image is overlapped with the first geographical range of the target area;

the shooting time of the remote sensing image is within a set time range.

In yet another implementation, before searching for a plurality of remote sensing images matched with a target area to be embedded and remote sensing data respectively corresponding to the plurality of remote sensing images, the method further includes:

acquiring the remote sensing image and extracting remote sensing data of the remote sensing image;

and taking the extracted remote sensing data of the remote sensing image as index information, and correspondingly storing the remote sensing image and the remote sensing data.

In yet another implementation, prior to storing the remote sensing image and the remote sensing data in correspondence, the method further comprises:

acquiring a reference remote sensing image matched with the spectral characteristics of the target area;

and carrying out color homogenizing treatment on the remote sensing image based on the spectral characteristics of the remote sensing image and the spectral characteristics of the reference remote sensing image.

In yet another implementation, the target area to be tessellated comprises at least one mesh to be tessellated, or at least one cell in the mesh to be tessellated;

in case the target area to be tessellated comprises at least one mesh to be tessellated, the method further comprises:

obtaining the size of each unit cell in a first grid to be inlaid; the first grid is any one of the at least one grid to be inlaid;

and dividing the first grid into a plurality of cells according to the size of each cell in the first grid.

In yet another implementation, obtaining a size of each cell in a first grid to be tessellated includes:

determining the size of each unit cell in the row direction according to the number of pixels occupied by each unit cell in the row direction and the row resolution; determining the size of each unit cell in the column direction according to the number of pixels occupied by each unit cell in the column direction and the column resolution;

The dividing the first grid into a plurality of cells according to the size of each cell in the first grid includes:

determining the total row number of the divided cells according to the size of each cell in the row direction and the circumference of the earth, and determining the total column number of the divided cells according to the size of each cell in the column direction and the circumference of the earth;

and dividing the first grid according to the geographical range of the first grid and the total row number and the total column number of the cells divided by the first grid to obtain a plurality of cells.

In a second aspect, a remote sensing image mosaic device is provided, the device comprising:

the searching module is used for searching a plurality of remote sensing images matched with the target area to be embedded and remote sensing data respectively corresponding to the remote sensing images;

the first determining module is used for determining effective areas corresponding to the plurality of remote sensing images respectively based on a first geographical range of the target area and the remote sensing data corresponding to the plurality of remote sensing images respectively; the remote sensing data comprises a second geographical range and cloud cover rate of the remote sensing image;

The selection module is used for selecting a target remote sensing image from the plurality of remote sensing images for a plurality of times based on the effective areas corresponding to the plurality of remote sensing images respectively;

and the second determining module is used for determining the inlaid image based on the target remote sensing images selected for multiple times.

In one implementation, the first determining module includes:

a first determination unit, configured to determine a first area of the target area based on a first geographic range of the target area; determining a second area of an area in each remote sensing image, wherein the second geographical range overlaps with the first geographical range, based on the first geographical range and a second geographical range of each remote sensing image;

and the second determining unit is used for determining the effective area corresponding to each remote sensing image based on the calculated first area, the second area corresponding to each remote sensing image and the cloud cover rate corresponding to each remote sensing image.

In yet another implementation, the second determining unit is specifically configured to: determining the effective coverage rate of each remote sensing image according to a preset weight parameter and the cloud cover rate corresponding to each remote sensing image; and determining the effective area corresponding to each remote sensing image according to the effective coverage rate of each remote sensing image, the first area and the second area corresponding to each remote sensing image.

In yet another implementation, the selection module includes:

the arrangement unit is used for arranging the plurality of remote sensing images according to the sequence of the effective areas corresponding to the plurality of remote sensing images from large to small to obtain the arrangement result of the plurality of remote sensing images;

the selecting unit is used for sequentially selecting a target remote sensing image from the plurality of remote sensing images according to the arrangement result;

the second determination module is to: and assigning the pixels of the target area according to the sequentially selected pixels of the target remote sensing image until all the pixels of the target area are assigned.

In yet another implementation, the ranking unit is further configured to exclude the first remote sensing image matching the target area from the ranking result if the second geographical range of the first remote sensing image is included in the second geographical range of the second remote sensing image matching the target area.

In yet another implementation, the searching module is specifically configured to search for a plurality of remote sensing images that meet the search condition, and remote sensing data corresponding to the plurality of remote sensing images respectively;

wherein the search condition includes at least one of the following conditions:

the shooting time of the remote sensing image is within a set time range.

In yet another implementation, the apparatus further comprises:

the first acquisition module is used for acquiring the remote sensing image;

the extraction module is used for extracting the remote sensing data of the remote sensing image;

and the storage module is used for taking the extracted remote sensing data of the remote sensing image as index information and correspondingly storing the remote sensing image and the remote sensing data.

In yet another implementation, the apparatus further comprises:

the second acquisition module is used for acquiring a reference remote sensing image matched with the spectral characteristics of the target area;

and the color homogenizing module is used for homogenizing the remote sensing image based on the spectral characteristics of the remote sensing image and the spectral characteristics of the reference remote sensing image.

in case the target area to be tessellated comprises at least one mesh to be tessellated, the apparatus further comprises:

The third acquisition module is used for acquiring the size of each unit cell in the first grid to be inlaid; the first grid is any one of the at least one grid to be inlaid;

and the dividing module is used for dividing the first grid into a plurality of cells according to the size of each cell in the first grid.

In yet another implementation, the third obtaining module is specifically configured to determine a size of each cell in the row direction according to the number of pixels occupied by each cell in the row direction and the row resolution; determining the size of each unit cell in the column direction according to the number of pixels occupied by each unit cell in the column direction and the column resolution;

the dividing module includes:

a third determining unit, configured to determine a total number of rows of the divided cells according to the size of each cell in the row direction and the circumference of the earth, and determine a total number of columns of the divided cells according to the size of each cell in the column direction and the circumference of the earth;

and the dividing unit is used for dividing the first grid according to the geographical range of the first grid and the total row number and the total column number of the cells divided by the first grid to obtain a plurality of cells.

In a third aspect, an electronic device is provided, including: an input device, an output device, a memory, and a processor; wherein the memory stores a set of program codes and the processor is configured to call the program codes stored in the memory to perform the method as described in the first aspect or any one of the first aspects.

In a fourth aspect, there is provided a non-transitory computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement a method as described in the first aspect or any of the first aspects.

By adopting the scheme disclosed by the invention, the following technical effects are achieved:

each processing node is respectively responsible for inlaying the remote sensing image of the target area in the distributed system, so that the inlaying task of the remote sensing image of a large-range area can be quickly completed, and automatic and efficient image inlaying is realized; and the effective area of the remote sensing image can be accurately determined based on the second geographic range and the cloud cover rate of the remote sensing image, and the target remote sensing image is selected according to the effective area of the remote sensing image, so that the quality of the embedded image can be effectively improved.

Drawings

FIG. 1 is a schematic diagram of a distributed system architecture;

fig. 2 is a schematic flowchart illustrating a remote sensing image mosaic method according to an embodiment of the present disclosure;

fig. 3 is a schematic flowchart of a remote sensing image mosaic method according to an embodiment of the present disclosure;

FIG. 4a is an image before shading;

FIG. 4b is an image after leveling;

FIG. 5 is a schematic diagram of the partitioning of a grid;

FIG. 6 is a schematic diagram of image mosaic;

fig. 7 is a schematic diagram of a mesh partitioning process provided by an embodiment of the present disclosure;

FIG. 8 is a schematic diagram of a grid coordinate system;

fig. 9 is a schematic structural diagram of a remote sensing image mosaic device according to an embodiment of the present disclosure;

fig. 10 is a schematic structural diagram of an electronic device provided in an embodiment of the present disclosure.

Detailed Description

The technical solutions in the embodiments of the present disclosure will be described below with reference to the drawings in the embodiments of the present disclosure.

Because the remote sensing image data volume is very huge, on one hand, in the storage process of the remote sensing image, the storage requirement of large data volume is difficult to meet by adopting a single physical storage node mode, and on the other hand, when the mosaic task of the remote sensing image is executed, the processing pressure is too large and the processing efficiency is too low when the mosaic task is executed by adopting a single physical processing node. Therefore, the distributed system can be used for storing and inlaying the remote sensing images.

Fig. 1 is a schematic structural diagram of a distributed system, which includes n processing nodes, where n is a positive integer. Each processing node can be used for storing a certain amount of information such as remote sensing images and corresponding remote sensing data and can also be used for executing the mosaic task of the remote sensing images. For example, each processing node may execute a remote sensing image mosaic task for a target area to be mosaiced, and each processing node may process the storage task and the mosaic task in parallel. By adopting the remote sensing image mosaic method in which the plurality of processing nodes execute in parallel in the distributed system, the mosaic task of the remote sensing image in a large-range area can be quickly completed, and automatic and efficient image mosaic is realized. In addition, in the present disclosure, the effective area of the remote sensing image can be accurately determined based on the second geographic range and the cloud cover rate of the remote sensing image, and the target remote sensing image is selected according to the effective area of the remote sensing image, so that the quality of the embedded image can be effectively improved.

Fig. 2 is a schematic flowchart of a remote sensing image mosaic method provided by an embodiment of the present disclosure, and the method may be applied to the distributed system exemplarily shown in fig. 1, where the distributed system includes multiple processing nodes, and the multiple processing nodes execute the following mosaic method in parallel. Illustratively, the method may comprise the steps of:

S101, searching a plurality of remote sensing images matched with a target area to be embedded and remote sensing data respectively corresponding to the remote sensing images.

For the shooting of the remote sensing images in a large range, the remote sensing images shot by the shooting device have a certain geographical range, and the remote sensing images in the same geographical range can be obtained at different times. The remote sensing image and the corresponding remote sensing data are stored on each processing node in the distributed system in a distributed mode.

Illustratively, the remote sensing data corresponding to the remote sensing image includes at least one of the following information: the second geographical range of the remote sensing image, the cloud cover rate, the shooting time, the identification of the remote sensing image, the line number, the column number, the resolution ratio and the like of the remote sensing image.

In a possible implementation manner, a second geographic range, shooting time and the like of the remote sensing image can be respectively determined based on operating environment parameters of the remote sensing image shot by the shooting device, such as shooting time, longitude and latitude coordinates and other parameters; the number of lines, columns, resolution and the like of the remote sensing image can be respectively determined based on shooting configuration parameters of the shooting device, such as the size, resolution and other configuration parameters of the shot image; the cloud cover rate in the remote sensing image can be determined based on the image characteristics of the shot remote sensing image, for example, the cloud cover rate can be calculated according to an automatic cloud cover rate evaluation algorithm, or the remote sensing image is input into a cloud detection model to obtain the cloud cover rate; in addition, the remote sensing images can be obtained by shooting and stored in the distributed system, and the stored remote sensing images can be automatically numbered to obtain the identification of the remote sensing images. Of course, the above-mentioned manner of determining the remote sensing data is merely an exemplary illustration, and the present application is not limited thereto.

A plurality of processing nodes in the distributed system execute the tessellation method of this embodiment in parallel. For a mosaic area of a certain geographic range to be mosaiced, the mosaic area can be divided into a plurality of target areas, wherein each target area to be mosaiced corresponds to a first geographic range, and the first geographic range represents the geographic range in which the target area is located. Each processing node may be assigned to perform a mosaicking task of remotely sensing an image of a target area to be mosaicked.

In a possible implementation manner, each processing node may search, according to a first geographic range corresponding to a target area to be embedded, a plurality of remote sensing images matched with the target area to be embedded, and remote sensing data corresponding to the plurality of remote sensing images respectively. It can be understood that the plurality of remote sensing images and the corresponding remote sensing data searched by the processing node may be stored in the processing node, and may also be stored in other processing nodes of the distribution system, which is not limited by the present disclosure.

S102, determining effective areas corresponding to the remote sensing images respectively based on a first geographical range of a target area and the remote sensing data corresponding to the remote sensing images respectively; and the remote sensing data comprises a second geographical range and cloud cover rate of the remote sensing image.

For example, the second geographic range corresponding to the searched remote sensing image may be located entirely within the first geographic range of the target area, or may intersect with the first geographic range of the target area (i.e., only a part of the second geographic range corresponding to the remote sensing image is located within the first geographic range). In addition, when the remote sensing image contains the cloud amount, the influence of the coverage rate of the cloud amount in the remote sensing image on the quality of the mosaic image should be considered. Therefore, the effective areas corresponding to the remote sensing images are determined based on the first geographical range of the target area and the remote sensing data corresponding to the remote sensing images.

And S103, selecting a target remote sensing image from the plurality of remote sensing images for a plurality of times based on the effective areas corresponding to the plurality of remote sensing images respectively.

In the disclosure, after determining the effective areas corresponding to the multiple remote sensing images, since the sizes of the found effective areas of the multiple remote sensing images may be inconsistent, in order to improve the integrity of the mosaic image, in a possible implementation manner, the corresponding remote sensing images may be sequentially selected as the target remote sensing image according to the order of the effective areas from large to small.

And S104, determining the inlaid image based on the target remote sensing images selected for multiple times.

In the method, the pixels of the target area can be assigned according to the sequentially selected pixels of the target remote sensing image until all the pixels of the target area are assigned.

In a possible implementation manner, each time the selected target remote sensing image is used to assign a value to the target area, a target sub-area corresponding to the position of the pixel point in the target remote sensing image in the target area (the target sub-area may also be understood as a sub-area overlapping with the target remote sensing image in the target area) may be first determined, then, whether the pixel point of the target sub-area is assigned or not is determined, if all the pixel points in the target sub-area are not assigned, the corresponding position of the target sub-area is assigned by using the pixel point of the selected target remote sensing image, and if some pixel points in the target sub-area are assigned, the corresponding position of the non-assigned area in the target sub-area is assigned by using the pixel point of the selected target remote sensing image.

Or, in another possible implementation, when the selected target remote sensing image is used to assign a value to the target area, the target sub-area corresponding to the position of the pixel point in the target remote sensing image in the target area may also be undetermined, that is, the position to be embedded of the target area is determined directly for the position of each pixel point in the target remote sensing image, and then it is determined whether the pixel at the position to be embedded is assigned, if the position to be embedded is assigned, the value assignment operation is not performed, and if the position to be embedded is not assigned, the pixel at the position to be embedded is assigned by using the pixel at the corresponding position in the target remote sensing image.

According to the remote sensing image mosaic method provided by the embodiment of the disclosure, each processing node is respectively responsible for mosaic of the remote sensing image of the target area in the distributed system, so that the mosaic task of the remote sensing image of a large area can be rapidly completed, and automatic and efficient image mosaic is realized; and the effective area of the remote sensing image can be accurately determined based on the second geographic range and the cloud cover rate of the remote sensing image, and the target remote sensing image is selected according to the effective area of the remote sensing image, so that the quality of the embedded image can be effectively improved.

Fig. 3 is a schematic flow chart of another remote sensing image mosaic method provided by an embodiment of the present disclosure, which may also be applied to the distributed system shown in fig. 1, where the distributed system includes multiple processing nodes, and the multiple processing nodes execute the following mosaic method in parallel. The processing node may also perform processing such as storage and search of the remote sensing image in the mosaic execution method. Illustratively, the method may comprise the steps of:

s201, obtaining the remote sensing image and extracting remote sensing data of the remote sensing image.

For the shooting of remote sensing images in a large range, each remote sensing image shot by the shooting device has a certain geographical range, and the remote sensing images in the same geographical range can be obtained at different times. And after the remote sensing image shot by the shooting device is obtained, extracting the remote sensing data of the remote sensing image. The remote sensing data corresponding to the remote sensing image comprises a second geographical range (boundary) of the remote sensing image, cloud coverage (cloud _ cover), shooting time (acquisition _ date), identification (id) of the remote sensing image, row number (rows) of the remote sensing image, column number (cols), resolution (resolution) and the like. Illustratively, the obtained telemetry data is stored in extensible markup language (XML) format.

Illustratively, the data about the remote sensing image stored in the XML file, which includes the remote sensing data, is as shown in table 1 below:

TABLE 1

S202, obtaining a reference remote sensing image matched with the spectral characteristics of the target area.

Due to weather conditions, sensor factors and the like, the remote sensing images have the problem of inconsistent color and brightness. As shown in fig. 4a, the two remote sensing images before color homogenization are dark in the whole, and a certain color difference exists between the two remote sensing images, so that the boundary of the boundary is obvious.

The embodiment selects the remote sensing image matched with the spectral characteristics of the target area as the reference remote sensing image. Illustratively, the spectral characteristics of the target region include at least one of information of a wavelength band of the target region, a range of gray levels of each wavelength band, and a pixel corresponding to each gray level. The target region may refer to the entire mosaic region in a wide range, or may refer to a target region corresponding to each processing node.

And S203, carrying out color homogenizing treatment on the remote sensing image based on the spectral characteristics of the remote sensing image and the spectral characteristics of the reference remote sensing image.

And carrying out color homogenizing treatment on the remote sensing image according to the spectral characteristics of the remote sensing image and the spectral characteristics of the reference remote sensing image. The remote sensing images are subjected to color homogenizing treatment based on the spectral characteristics of the reference remote sensing images, so that the spectral characteristics of the remote sensing images after color homogenizing can be consistent with the spectral characteristics of the reference remote sensing images as much as possible, and the consistency of the color and the brightness of the remote sensing images can be ensured as much as possible.

As shown in fig. 4b, the two remote sensing images after color homogenization are integrally bright, and the two remote sensing images have almost no color difference.

Specifically, S203 may include the following steps a to D:

and A, respectively obtaining the cumulative histogram of each wave band of the reference remote sensing image and the cumulative histogram of each wave band of each remote sensing image.

The reference remote sensing image is a remote sensing image capable of representing spectral characteristics of the target area.

The remote sensing image includes a plurality of bands, such as Red Green Blue (RGB). And respectively obtaining the cumulative histogram of each wave band of the reference remote sensing image and the cumulative histogram of each wave band of each remote sensing image. The cumulative histogram is an accumulated value of the number of pixels corresponding to each gray level in the remote sensing image. Each band corresponds to a cumulative histogram.

For example, taking a certain wavelength band as an example, assuming that the gray scale is 0 to 7 (actually, the range of the gray scale may be 0 to 255), one remote sensing image of 8 × 8=64 pixels, where the number of pixels corresponding to each gray scale is:

P(0)＝12，P(1)＝10，P(2)＝8，P(3)＝6，P(4)＝20，P(5)＝5，P(6)＝1，P(7)＝2。

where P (i) represents the number of pixels with a gray level of i.

According to the calculation formula of the cumulative histogram:

wherein k =1,2, \8230, 7,

The cumulative histogram of the remote sensing image is: s (0) =0, S (1) =10, S (2) =18, S (3) =24, S (4) =44, S (5) =49, S (6) =50, S (7) =52. Note that during the calculation, i ≧ 1, i.e., the point at which the pixel value is 0, is excluded.

In the same way, the cumulative histogram of each wave band of the reference remote sensing image can be obtained. Wherein, the cumulative histogram of a band corresponding to a certain band of the remote sensing image is:

S(0)＝0，S(1)＝14，S(2)＝20，S(3)＝32，S(4)＝32，S(5)＝38，S(6)＝48，S(7)＝48。

and B, respectively carrying out normalization processing on the cumulative histogram of each wave band of the reference remote sensing image and the cumulative histogram of each wave band of each remote sensing image.

The cumulative histogram of each band of the reference remote sensing image and the cumulative histogram of each band of each remote sensing image are normalized respectively, so that the subsequent calculation can be simplified.

Specifically, the histogram of each gray level is divided by the cumulative histogram of the highest level. As in the above example, after normalization processing is performed on the cumulative histogram of a certain band of the remote sensing image, the following results are obtained: s '(0) =0/52=0, S' (1) =10/52=0.192, S '(2) =18/52=0.346, S' (3) =24/52=0.462, S '(4) =44/52=0.846, S' (5) =49/52=0.942, S '(6) =50/52=0.962, S' (7) =52/52=1.

After normalization processing is carried out on the cumulative histogram of the corresponding wave band of the reference remote sensing image, the following results are obtained: s '(0) =0/48=0, S' (1) =14/48=0.292, S '(2) =20/48=0.417, S' (3) =32/48=0.667, S '(4) =32/48=0.667, S' (5) =38/48=0.792, S '(6) =48/48=1, S' (7) =48/48=1.

And C, calculating the absolute value of the difference between the value of each gray scale in the cumulative histogram of each wave band of each remote sensing image after normalization processing and the value of each gray scale in the cumulative histogram of the corresponding wave band of the reference remote sensing image after normalization processing to obtain the absolute value of the difference between the values of each gray scale.

The method comprises the steps of taking a first wave band as any wave band, taking a first gray scale as any gray scale, and aiming at the value of the first gray scale in the cumulative histogram of the first wave band of the remote sensing image, if the absolute value of the difference between the value of the first gray scale in the cumulative histogram of the first wave band of the remote sensing image and the value of the target gray scale of the cumulative histogram of the first wave band of the reference remote sensing image is minimum, indicating that the pixel value of the first gray scale of the remote sensing image is closer to the pixel value of the target gray scale in the reference remote sensing image.

Taking the calculated S '(0) and S' (3) of the remote sensing image as an example, if the absolute value of the difference between S '(0) of the remote sensing image and S' (0) of the reference remote sensing image is obviously the smallest, it means that the pixel value of S '(0) of the remote sensing image is closer to the pixel value of S' (0) of the reference remote sensing image; if the absolute value of the difference between S '(3) of the remote-sensing image and S' (2) of the reference remote-sensing image is the smallest, it means that the pixel value of S '(3) of the remote-sensing image is closer to the pixel value of S' (2) of the reference remote-sensing image.

And mapping the pixel values after calculating the absolute value of the difference between the value of each gray level in the cumulative histogram of each wave band of each remote sensing image and the value of each gray level in the cumulative histogram of the corresponding wave band of the reference remote sensing image. The specific mapping manner is shown in step D, for example.

Step D, mapping the pixel value of the first gray level in the cumulative histogram of the first wave band of each remote sensing image into the pixel value of the target gray level in the cumulative histogram of the first wave band of the reference remote sensing image; wherein an absolute value of a difference between the gray value of the first gray level and the gray value of the target gray level is smallest.

Still taking the above example as an example, the pixel value of the grayscale level P (0) of the remote-sensing image is mapped to the pixel value of the grayscale level P (0) of the reference remote-sensing image, the pixel value of the grayscale level P (1) of the remote-sensing image is mapped to the pixel value of the grayscale level P (1) of the reference remote-sensing image, the pixel value of the grayscale level P (2) of the remote-sensing image is mapped to the pixel value of the grayscale level P (1) of the reference remote-sensing image, the pixel value of the grayscale level P (3) of the remote-sensing image is mapped to the pixel value of the grayscale level P (2) of the reference remote-sensing image, the pixel value of the grayscale level P (4) of the remote-sensing image is mapped to the pixel value of the grayscale level P (5) of the reference remote-sensing image, the pixel value of the grayscale level P (5) of the remote-sensing image is mapped to the pixel value of the grayscale level P (6) of the reference remote-sensing image, and the pixel value of the grayscale level P (6) of the remote-sensing image is mapped to the pixel value of the grayscale level P (6) of the reference remote-sensing image, and the pixel value of the reference remote-sensing image is mapped to the pixel value of the grayscale level P (6) of the pixel of the reference remote-sensing image.

And finishing the color homogenizing of each remote sensing image, so that the pixel value of each gray scale of each wave band of each remote sensing image is closest to the pixel value of a certain gray scale of the corresponding wave band of the reference remote sensing image.

The step 202 and the step 203 may be optional steps, that is, in the process of storing the remote sensing image, the process of the color homogenizing process may be optional steps, which is not limited in the present disclosure.

And S204, taking the extracted remote sensing data of the remote sensing image as index information, and correspondingly storing the remote sensing image and the remote sensing data.

After the remote sensing data of the remote sensing image is extracted in the above steps, the extracted remote sensing data of the remote sensing image can be used as index information, and the remote sensing image and the corresponding remote sensing data thereof can be stored in a distributed manner on each processing node in the distributed system.

The steps S202 to S203 are to perform the color-uniformizing process on the remote sensing image, and the step S204 is to store the remote sensing image and the remote sensing data corresponding thereto. It is understood that the color homogenizing process of the present embodiment may be performed before the remote sensing image and the remote sensing data corresponding thereto are stored (i.e., step S204), or may be performed after the remote sensing image and the remote sensing data corresponding thereto are stored and before the mosaic operation, which is not limited by the present disclosure.

S205, searching a plurality of remote sensing images which accord with the search condition and remote sensing data which respectively correspond to the remote sensing images, wherein the search condition comprises at least one of the following conditions: the second geographical range of the remote sensing image is overlapped with the first geographical range of the target area; the shooting time of the remote sensing image is within a set time range.

Before searching the remote sensing image, a first geographical range of the target area needs to be determined. In a possible embodiment, the target area to be tessellated may be a tessellation minimum unit, for example, the tessellation minimum unit is referred to as a cell, where a cell may be understood as being further divided based on the mesh of the target area. In this case, a first geographical range of the target area is determined, i.e. the geographical range of the cells is determined.

As shown in the schematic diagram of the cells under the grid in fig. 5, the coordinate range of each cell under the geographic coordinate system can be determined by its 4 vertices. As shown in fig. 5, cell a includes four vertices a, b, c, d. The coordinates of vertex a are (-world + col × tileColSize × colRes, world-row × tileRowSize × rowRes). The coordinates of vertices b, c, d can then be obtained from vertex a plus or minus the height or width of a single cell. Wherein the content of the first and second substances,

row＝floor((x+world)/(tileRowSize·rowRes))

col＝floor((world-y)/(tileColSize·colRes))

The geographic coordinates of any point in the first geographic range of the target area are (x, y), row is the row number of the cell intersected with the first geographic range, and col is the column number of the cell intersected with the first geographic range.

In a possible implementation process, after the first geographical range of the target area is obtained, a plurality of remote sensing images meeting the conditions and corresponding remote sensing data thereof can be searched according to the first geographical range and/or the set time range of the target area. For example, the remote sensing image with the second geographic range overlapped with the first geographic range of the target area is searched, and/or the remote sensing image with the shooting time within a set time range is searched.

S206, determining a first area of the target area based on the first geographical range of the target area; and determining a second area of an area in each remote sensing image, in which the second geographical range overlaps with the first geographical range, based on the first geographical range and the second geographical range of each remote sensing image.

After the first geographic range of the target area is determined, a first area of the target area may be calculated. And calculating a second area of an area in each remote sensing image, in which the second geographical range overlaps with the first geographical range, based on the first geographical range and the second geographical range of each remote sensing image.

And S207, determining the effective area corresponding to each remote sensing image based on the calculated first area, the second area corresponding to each remote sensing image and the cloud cover rate corresponding to each remote sensing image.

The effective area corresponding to each remote sensing image needs to consider a second area of the second geographical range of each remote sensing image, which is overlapped with the first geographical range of the target area, and the cloud cover rate corresponding to each remote sensing image, and the effective area corresponding to each remote sensing image can be determined according to the first area, the second area and the cloud cover rate corresponding to each remote sensing image. The effective area is an effective image portion for the mosaic image.

Specifically, S207 includes: determining the effective coverage rate of each remote sensing image according to a preset weight parameter and the cloud cover rate corresponding to each remote sensing image; and determining the effective area corresponding to each remote sensing image according to the effective coverage rate and the first area of each remote sensing image and the second area corresponding to each remote sensing image.

Specifically, the effective area Sv corresponding to each remote sensing image is:

Sv＝(Sr/S)×(1–w·c)

wherein Sr is the area of one or more remote sensing images in the target region, S is the area of the target region, w is a weight parameter not less than 1, and c is the cloud cover rate.

And S208, arranging the plurality of remote sensing images according to the sequence of the effective areas corresponding to the plurality of remote sensing images from large to small to obtain an arrangement result of the plurality of remote sensing images.

The remote sensing images matched with the target area are different in shape and size, so that the effective areas are different in size. In order to improve the integrity of the mosaic image, the remote sensing images can be arranged according to the sequence of the effective areas corresponding to the remote sensing images from large to small, so that the arrangement result of the remote sensing images can be obtained.

As shown in the schematic diagram of image mosaic shown in fig. 6, the remote sensing image set matched with the target area includes remote sensing images 1 to 4. The remote sensing images 1 to 4 are different in size and shape. The remote-sensing image 1 is entirely contained in the remote-sensing image 4, and the remote-sensing image 2 and the remote-sensing image 3 are partially overlapped.

As can be seen from the above-mentioned fig. 6 that all the remote sensing images 1 are contained in the remote sensing image 4, generally, the remote sensing images have repeatability, that is, a large number of images are taken in different times for the same target area, so that in the case of searching the remote sensing images according to the shooting time of the remote sensing images, the remote sensing images acquired at different shooting times in the same target area intersect with each other. As shown in fig. 6, the remote sensing images intersecting with a target area include remote sensing images 1 to 4, where the remote sensing image 1 and the remote sensing image 4 are images of the same area obtained at different times, and the remote sensing image 1 is included in the remote sensing image 4. Further, the method may further include: in a case where the second geographical range of the first remote sensing image matching the target area is included in the second geographical range of the second remote sensing image matching the target area, the first remote sensing image is excluded from the arrangement result. That is, as shown in fig. 6, the remote sensing image 1 is directly excluded from the arrangement result to reduce the amount of subsequent calculation.

It should be noted that the process of excluding the first remote sensing image may be completed during the arrangement process, before the arrangement process, or after the arrangement process, which is not limited by the present disclosure.

As shown in fig. 6, when the effective areas of the remote sensing images 2 to 4 are sorted, the sorting result is: remote sensing image 4, remote sensing image 3 and remote sensing image 2.

And S209, sequentially selecting a target remote sensing image from the plurality of remote sensing images according to the arrangement result.

According to the arrangement result of the sizes of the effective areas of the plurality of remote sensing images, the remote sensing image with the larger effective area is preferentially selected as the target remote sensing image, and then the remote sensing image with the smaller effective area than the previously selected remote sensing image is selected as the target remote sensing image selected next time.

Illustratively, as shown in fig. 6, the remote sensing image 4, the remote sensing image 3, and the remote sensing image 2 are sequentially selected as the target remote sensing image according to the arrangement result.

And S210, assigning the pixels of the target area according to the sequentially selected pixels of the target remote sensing image until all the pixels of the target area are assigned.

After the target remote sensing image is sequentially selected from the plurality of remote sensing images matched with the target area according to the arrangement result, the pixels of the target area can be assigned according to the sequentially selected pixels of the target remote sensing image until all the pixels of the target area are assigned.

For example, with continued reference to fig. 6, after obtaining the arrangement result of the target remote sensing image, the process of mosaicing the target area may include:

firstly, determining a first target subregion corresponding to the position of a pixel point in the remote sensing image 4 in a target region; then, it is determined whether the pixel points of the first target sub-region have been assigned, as shown in fig. 6, all the pixel points in the first target sub-region have not been assigned, so that the pixels of the selected remote sensing image 4 can be used to assign the corresponding positions of the first target sub-region.

Secondly, firstly, determining a second target subregion corresponding to the position of the pixel point in the remote sensing image 3 in the target region; then, whether the pixel points of the second target sub-region are assigned or not is judged, as shown in fig. 6, all the pixel points in the second target sub-region are not assigned, so that the selected pixels of the remote sensing image 3 can be used for assigning the corresponding positions of the second target sub-region respectively.

Thirdly, determining a third target subarea corresponding to the positions of the pixel points in the remote sensing image 2 in the target area; then, whether the pixel points of the third target sub-region are assigned or not is judged, as shown in fig. 6, since the remote sensing image 2 and the remote sensing image 3 are partially overlapped, and part of the pixel points in the third target sub-region are assigned, the pixels in the selected remote sensing image 2 can be used for assigning the corresponding positions of the regions which are not assigned in the third target sub-region.

And sequentially selecting the target remote sensing image and assigning the pixels of the target area according to the mode, and finally obtaining the remote sensing image after the mosaic of the target area.

Specifically, in one implementation, the pixel coordinates of any one point in the target remote sensing image in the target area can be determined according to the following formula:

x'＝(x+world-col·tileColSize·colRes)/colRes

y'＝(world-row·tileRowSize·rowRes-y)/rowRes

the geographic coordinate of any point in the target remote sensing image is (x, y), row is the line number of a target area intersected with the target remote sensing image, col is the column number of a cell intersected with the target remote sensing image, (x ', y') is the pixel coordinate of the point (x, y) in the cell, and floor (·) is a downward integral function.

As can be seen from the above, each processing node can perform the mosaic method in the minimum mosaic unit of the target area. Therefore, aiming at the large-range area, each processing node can be respectively responsible for processing the mosaic task in each target area in the large-range area so as to quickly finish the mosaic task of the remote sensing image in the large-range area.

According to the remote sensing image mosaic method provided by the embodiment of the disclosure, each processing node is respectively responsible for mosaic of the remote sensing image of the target area in the distributed system, so that the mosaic task of the remote sensing image of a large area can be rapidly completed, and automatic and efficient image mosaic is realized; the remote sensing image is subjected to color homogenizing treatment, so that the consistency of the color and the brightness of the embedded image is effectively improved; and the pixel value of the remote sensing image with larger intersection area with the grid is preferentially adopted for assignment, so that the integrity of the mosaic image can be improved; in addition, when the intersection area of the image and the grid is calculated, the cloud cover rate is considered, so that the image quality of the mosaic image can be improved; and for contained remote sensing images, the remote sensing images are directly discarded during assignment, so that the calculation amount can be reduced.

In addition, in the present disclosure, if the target area to be mosaicked is a mesh, the mesh may be further divided to obtain a minimum mosaicking unit, that is, the above-mentioned cell, before the above-mentioned searching, determining the effective area of the remote sensing image, and selecting the target remote sensing image.

Fig. 7 is a flowchart illustrating a method for partitioning a grid according to an embodiment of the present disclosure, where the method illustratively includes the following steps:

s301, obtaining the size of each unit cell in the first grid to be inlaid.

And acquiring the size of each unit cell in the first grid to be inlaid. The first mesh is any one of at least one mesh to be tessellated.

Specifically, S301 includes: determining the size of each unit cell in the row direction according to the number of pixels occupied by each unit cell in the row direction and the row resolution; and determining the size of each unit cell in the column direction according to the number of pixels occupied by each unit cell in the column direction and the column resolution.

First, a coordinate system, such as the schematic diagram of the grid coordinate system shown in fig. 8, is set up, the coordinate system with O as the origin is the geographic coordinate system, and the coordinate system with Og as the origin is the grid coordinate system. The geographic coordinate system takes a point with longitude and latitude of 0 as an origin, the coordinate of the origin O is (0, 0), and the unit of the coordinate of the geographic coordinate system is meter; the coordinates of the grid coordinate system are in units of pixels, and the coordinates of the origin Og in the geographic coordinate system are (-world, world), world is half the circumference of the earth, the value is 20037508.342787 meters, and the coordinates of the origin Og in the grid coordinate system are (0,0).

Then, according to the setting of the parameters of the grid coordinate system, the size of each cell in the first grid is obtained: the size of each cell in the row direction is: tileRowSize. RowRes; and the dimension of each cell in the column direction is: tileColSize. ColRes. Wherein tileRowSize is the number of pixels occupied by each cell in the row direction, tileColSize is the number of pixels occupied by each cell in the column direction, rowRes is the row resolution, and colRes is the column resolution. The tillerowsize, tileconsize, rowRes, and colRe can be obtained from the remote sensing data of the extracted remote sensing image.

S302, dividing the first grid into a plurality of cells according to the size of each cell in the first grid.

After the size of each cell in the first grid is obtained, the size of the first grid is determined, so that the first grid can be divided into a plurality of cells.

Specifically, S302 includes: determining the total row number of the divided cells according to the size of each cell in the row direction and the circumference of the earth, and determining the total column number of the divided cells according to the size of each cell in the column direction and the circumference of the earth; and dividing the first grid according to the geographical range of the first grid and the total row number and the total column number of the cells divided by the first grid to obtain a plurality of cells.

Specifically, the total number of rows and the total number of columns of the divided cells may be determined according to the following formulas:

Row＝ceil(2·world/(tileRowSize·rowRes))

Col＝ceil(2·world/(tileColSize·colRes))

where Row is the total number of rows of the divided cells, col is the total number of columns of the divided cells, world is half of the circumference of the earth, tillerowsize is the number of pixels occupied by each cell in the Row direction, tilcolisize is the number of pixels occupied by each cell in the column direction, rowRes is the Row resolution, colRes is the column resolution, and ceil (·) is an upward rounding function.

Thus, the first grid may be divided into Row × Col cells.

Specifically, as shown in the schematic division diagram of the grid shown in fig. 5, there are 4 × 5=20 cells in total, each cell has tillerowsize × tileconsize pixels, and the row resolution and the column resolution of each pixel are rowRes and colRes, respectively.

After the first grid is divided into a plurality of cells, each cell corresponds to one spliced image.

By adopting the grid division method, the grid can be accurately divided. Subsequently, the remote sensing image mosaic process can be executed respectively for each divided cell, and the mosaic remote sensing image corresponding to the target area is obtained.

The method of the embodiments of the present disclosure is set forth above in detail and the apparatus of the embodiments of the present disclosure is provided below.

Based on the same concept of the remote sensing image mosaic method in the above embodiments, as shown in fig. 8, the embodiment of the present disclosure further provides a remote sensing image mosaic device. The tessellation device may be any of the processing nodes in fig. 1. The damascene device 1000 includes: the searching module 101, the first determining module 102, the selecting module 103, and the second determining module 104 may further include a first obtaining module 105, an extracting module 106, a storing module 107, a second obtaining module 108, a color homogenizing module 109, a third obtaining module 110, and a dividing module 111 (connected by a dotted line in the figure). Exemplarily, the following steps are carried out:

the searching module 101 is configured to search a plurality of remote sensing images matched with a target area to be embedded, and remote sensing data corresponding to the plurality of remote sensing images respectively;

a first determining module 102, configured to determine, based on a first geographic range of the target region and remote sensing data corresponding to the plurality of remote sensing images, effective areas corresponding to the plurality of remote sensing images, respectively; the remote sensing data comprises a second geographical range and cloud cover rate of the remote sensing image;

the selection module 103 is configured to select a target remote sensing image from the plurality of remote sensing images for multiple times based on effective areas corresponding to the plurality of remote sensing images respectively;

And a second determining module 104, configured to determine a mosaic image based on the target remote sensing images selected multiple times.

In one implementation, the first determining module 102 includes:

a first determining unit 1021, configured to determine a first area of the target area based on a first geographical range of the target area; determining a second area of an area in each remote sensing image, wherein the second geographical range overlaps with the first geographical range, based on the first geographical range and a second geographical range of each remote sensing image;

a second determining unit 1022, configured to determine an effective area corresponding to each remote sensing image based on the calculated first area, the second area corresponding to each remote sensing image, and the cloud cover rate corresponding to each remote sensing image.

In yet another implementation, the second determining unit 1022 is specifically configured to: determining the effective coverage rate of each remote sensing image according to a preset weight parameter and the cloud cover rate corresponding to each remote sensing image; and determining the effective area corresponding to each remote sensing image according to the effective coverage rate of each remote sensing image, the first area and the second area corresponding to each remote sensing image.

In yet another implementation, the selection module 103 includes:

the arrangement unit 1031 is configured to arrange the plurality of remote sensing images according to a descending order of effective areas corresponding to the plurality of remote sensing images, so as to obtain an arrangement result of the plurality of remote sensing images;

a selecting unit 1032, configured to sequentially select a target remote sensing image from the multiple remote sensing images according to the arrangement result;

the second determining module 104 is configured to: and assigning the pixels of the target area according to the sequentially selected pixels of the target remote sensing image until all the pixels of the target area are assigned.

In yet another implementation, the arrangement unit 1031 is further configured to exclude the first remote sensing image matching the target area from the arrangement result if the second geographical range of the first remote sensing image is included in the second geographical range of the second remote sensing image matching the target area.

In yet another implementation, the searching module 101 is specifically configured to search for a plurality of remote sensing images that meet the search condition, and remote sensing data corresponding to the plurality of remote sensing images respectively;

wherein the search condition includes at least one of the following conditions:

the shooting time of the remote sensing image is within a set time range.

In yet another implementation, the first obtaining module 105 is configured to obtain the remote sensing image;

an extraction module 106, configured to extract remote sensing data of the remote sensing image;

and the storage module 107 is used for taking the extracted remote sensing data of the remote sensing image as index information and correspondingly storing the remote sensing image and the remote sensing data.

In yet another implementation, the second obtaining module 108 is configured to obtain a reference remote sensing image matched with the spectral feature of the target region;

and the color homogenizing module 109 is used for homogenizing the remote sensing image based on the spectral characteristics of the remote sensing image and the spectral characteristics of the reference remote sensing image.

a third obtaining module 110, configured to obtain a size of each cell in the first grid to be tessellated; the first grid is any one of the at least one grid to be inlaid;

A dividing module 111, configured to divide the first grid into a plurality of cells according to a size of each cell in the first grid.

In yet another implementation, the third obtaining module 110 is specifically configured to determine, according to the number of pixels occupied by each cell in the row direction and the row resolution, a size of each cell in the row direction; determining the size of each unit cell in the column direction according to the number of pixels occupied by each unit cell in the column direction and the column resolution;

the dividing module 111 includes:

a third determining unit 1111, configured to determine a total number of rows of the divided cells according to the size of each cell in the row direction and the circumference of the earth, and determine a total number of columns of the divided cells according to the size of each cell in the column direction and the circumference of the earth;

a dividing unit 1112, configured to divide the first grid according to the geographic range of the first grid and the total number of rows and the total number of columns of the cells divided by the first grid, to obtain multiple cells.

For the detailed description of the above modules and units, reference may be made to the related description in the method embodiments shown in fig. 2 and fig. 3, and details are not repeated here.

According to the remote sensing image embedding device provided by the embodiment of the disclosure, each processing node is respectively responsible for embedding the remote sensing image of a target area in a distributed system, so that the remote sensing image embedding task of a large-range area can be quickly completed, and automatic and efficient image embedding is realized; the remote sensing image is subjected to color homogenizing treatment, so that the color and the brightness of the embedded image can be ensured to be consistent; and the pixel value of the remote sensing image with larger intersection area with the grid is preferentially adopted for assignment, so that the integrity of the mosaic image can be improved; in addition, when the intersection area with the grid is calculated, the cloud cover rate is considered, so that the image quality of the mosaic image can be improved; and for contained remote sensing images, the remote sensing images are directly discarded during assignment, so that the calculation amount can be reduced.

Fig. 9 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure, where the electronic device may be any one of the processing nodes shown in fig. 1. The electronic device 2000 includes: at least one input device 201; at least one output device 202; at least one processor 203, such as a CPU; and a memory 204, the input device 201, the output device 202, the processor 203, and the memory 204 being connected by a bus 205.

The input device 201 may be a touch panel, a physical button, or a mouse.

The output device 202 may be a display screen.

The memory 204 may be a high-speed RAM memory, or may be a non-volatile memory (non-volatile memory), such as a magnetic disk memory. The memory 204 is used for storing a set of program codes, and the input device 201, the output device 202 and the processor 203 are used for calling the program codes stored in the memory 204 and executing the following operations:

and determining the mosaic image based on the target remote sensing images selected for multiple times.

In one implementation, the processor 203 performs the operation of determining the effective areas corresponding to the remote sensing images based on the first geographic range of the target area and the remote sensing data corresponding to the remote sensing images, including:

In yet another implementation, the processor 203 performs the operation of determining the effective area corresponding to each remote sensing image based on the calculated first area, the second area corresponding to each remote sensing image, and the cloud cover rate corresponding to each remote sensing image, including:

In yet another implementation, the processor 203 performs the operation of selecting a target remote sensing image from the plurality of remote sensing images for a plurality of times based on the effective areas corresponding to the plurality of remote sensing images, respectively, including:

In another implementation, before performing the operation of assigning the pixels of the target area according to the sequentially selected pixels of the target remote sensing image until determining that all the pixels of the target area are assigned, the processor 203 further performs the following operation:

In yet another implementation, the processor 203 performs the operation of searching for a plurality of remote sensing images matched with the target area to be mosaicked, and remote sensing data corresponding to the plurality of remote sensing images, respectively, including:

wherein the search condition includes at least one of the following conditions:

the shooting time of the remote sensing image is within a set time range.

In yet another implementation, before the processor 203 performs the operation of searching for the plurality of remote sensing images matched with the target area to be mosaiced and the remote sensing data respectively corresponding to the plurality of remote sensing images, the following operations are further performed:

In yet another implementation, before the processor 203 performs the operation of storing the remote sensing image and the remote sensing data correspondingly, the following operation is further performed:

in case the target area to be tessellated comprises at least one mesh to be tessellated, the processor 203 further performs the following:

In yet another implementation, the processor 203 performs the operation of obtaining the size of each cell in the first mesh to be tessellated, including:

The processor 203 performs the operation of dividing the first grid into a plurality of cells according to the size of each cell in the first grid, including:

According to the electronic equipment provided by the embodiment of the disclosure, each processing node is respectively responsible for inlaying the remote sensing image of the target area in the distributed system, so that the inlaying task of the remote sensing image of a large-range area can be quickly completed, and automatic and efficient image inlaying is realized; the remote sensing image is subjected to color homogenizing treatment, so that the color and the brightness of the embedded image can be ensured to be consistent; and the pixel value of the remote sensing image with larger intersection area with the grid is preferentially adopted for assignment, so that the integrity of the mosaic image can be improved; in addition, when the intersection area of the image and the grid is calculated, the cloud cover rate is considered, so that the image quality of the mosaic image can be improved; and for contained remote sensing images, the remote sensing images are directly discarded during assignment, so that the calculation amount can be reduced.

It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.

In the several embodiments provided in the present disclosure, it should be understood that the disclosed system, apparatus, and method may be implemented in other ways. For example, the division of the unit is only one logical function division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. The shown or discussed mutual coupling, direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some interfaces, and may be in an electrical, mechanical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one position, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. The procedures or functions according to the embodiments of the disclosure are wholly or partially generated when the computer program instructions are loaded and executed on a computer. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored on or transmitted over a computer-readable storage medium. The computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wire (e.g., coaxial cable, fiber optic, digital Subscriber Line (DSL)), or wirelessly (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that includes one or more of the available media. The usable medium may be a read-only memory (ROM), or a Random Access Memory (RAM), or a magnetic medium, such as a floppy disk, a hard disk, a magnetic tape, a magnetic disk, or an optical medium, such as a Digital Versatile Disk (DVD), or a semiconductor medium, such as a Solid State Disk (SSD).

Claims

1. A remote sensing image mosaic method is applied to a distributed system, the distributed system comprises a plurality of processing nodes, the processing nodes execute the mosaic method in parallel, and the mosaic method comprises the following steps:

searching a plurality of remote sensing images matched with a target area to be inlaid and remote sensing data corresponding to the remote sensing images respectively, wherein the target area to be inlaid comprises at least one grid to be inlaid, each grid in the at least one grid to be inlaid comprises a plurality of unit grids, the unit grids are obtained by dividing according to the geographic range of each grid and the total row number and the total column number of the unit grids divided by each grid, the total row number of the divided unit grids is determined according to the size of each unit grid in the row direction and the earth perimeter, and the total column number of the divided unit grids is determined according to the geographic range of a first grid and the total row number and the total column number of the unit grids divided by the first grid;

2. The tessellation method of claim 1, wherein determining an effective area corresponding to each of the plurality of remotely sensed images based on the first geographic area of the target region and the remotely sensed data corresponding to each of the plurality of remotely sensed images comprises:

determining a first area of the target area based on a first geographic extent of the target area;

determining a second area of an area in each remote sensing image, wherein the second geographical range overlaps with the first geographical range, based on the first geographical range and a second geographical range of each remote sensing image;

3. The mosaicing method of claim 2, wherein determining the effective area corresponding to each remote-sensing image based on the calculated first area, the second area corresponding to each remote-sensing image, and the cloud cover rate corresponding to each remote-sensing image comprises:

4. The mosaic method according to any one of claims 1 to 3, wherein said selecting a target remote sensing image from said plurality of remote sensing images a plurality of times based on the effective areas corresponding to said plurality of remote sensing images, respectively, comprises:

5. The tessellation method of claim 4, wherein, prior to assigning the pixels of the target area based on sequentially selected pixels of the target remote sensing image until all of the pixels of the target area are determined to be assigned, the method further comprises:

6. The mosaicking method according to any one of claims 1 to 3, wherein the searching for a plurality of remote sensing images matched with the target area to be mosaicked, the remote sensing data corresponding to each of the plurality of remote sensing images, comprises:

wherein the search condition includes at least one of the following conditions:

the shooting time of the remote sensing image is within a set time range.

7. The mosaicking method according to any one of claims 1 to 3, wherein before searching for a plurality of remote sensing images matching a target area to be mosaicked, the method further comprises, before searching for remote sensing data corresponding to each of the plurality of remote sensing images:

8. The tessellation method of claim 7, wherein prior to storing the remote sensing image and the remote sensing data in correspondence, the method further comprises:

9. The embedding method according to any one of claims 1 to 3,

the method further comprises the following steps:

10. The method of claim 9, wherein obtaining the size of each cell in the first mesh to be tessellated comprises:

Determining the size of each unit cell in the row direction according to the number of pixels occupied by each unit cell in the row direction and the row resolution; and determining the size of each unit cell in the column direction according to the number of pixels occupied by each unit cell in the column direction and the column resolution.

11. An apparatus for embedding remote sensing images, the apparatus comprising:

the system comprises a searching module, a searching module and a processing module, wherein the searching module is used for searching a plurality of remote sensing images matched with a target area to be embedded and remote sensing data respectively corresponding to the remote sensing images, the target area to be embedded comprises at least one grid to be embedded, each grid in the at least one grid to be embedded comprises a plurality of cells, the cells are obtained by dividing according to the geographic range of each grid and the total row number and the total column number of the cells divided by each grid, the total row number of the divided cells is determined according to the size of each cell in the row direction and the circumference of the earth, and the total column number of the divided cells is determined according to the geographic range of a first grid and the total row number and the total column number of the cells divided by the first grid;

12. The tessellation device of claim 11, wherein the first determination module comprises:

13. The tessellation device of claim 12, wherein the second determination unit is specifically configured to: determining the effective coverage rate of each remote sensing image according to a preset weight parameter and the cloud cover rate corresponding to each remote sensing image; and determining the effective area corresponding to each remote sensing image according to the effective coverage rate of each remote sensing image, the first area and the second area corresponding to each remote sensing image.

14. The tessellation device of any of claims 11 to 13, wherein the selection module comprises:

15. The mosaic device according to claim 14, wherein said arrangement unit is further adapted to exclude said first remote sensing image matching said target area from said arrangement result if said second geographical range of said first remote sensing image is included in said second geographical range of said second remote sensing image matching said target area.

16. The mosaic device according to any one of claims 11 to 13, wherein said finding module is specifically configured to find a plurality of remote sensing images that meet the search condition, the remote sensing data corresponding to each of the plurality of remote sensing images;

wherein the search condition includes at least one of the following conditions:

the shooting time of the remote sensing image is within a set time range.

17. The damascene device as claimed in any one of claims 11 to 13, further comprising:

the first acquisition module is used for acquiring the remote sensing image;

18. The damascene apparatus of claim 17, the apparatus further comprising:

19. The damascene device as claimed in any one of claims 11 to 13, further comprising:

20. The apparatus according to claim 19, wherein the third obtaining module is specifically configured to determine a size of each cell in the row direction according to the number of pixels occupied by each cell in the row direction and a row resolution; and determining the size of each unit cell in the column direction according to the number of pixels occupied by each unit cell in the column direction and the column resolution.

21. An electronic device, comprising: an input device, an output device, a memory, and a processor; wherein the memory stores a set of program codes, and the processor is configured to call the program codes stored in the memory to execute the method according to any one of claims 1 to 10.

22. A non-transitory computer readable storage medium having computer program instructions stored thereon, wherein the computer program instructions, when executed by a processor, implement the method of any one of claims 1 to 10.