Seam line network generation method and apparatus, electronic device, and medium

Abstract

The present invention relates to the fields of computer vision, image processing, etc., and provides a seam line network generation method and apparatus, an electronic device, and a medium. The specific implementation solution comprises: determining respective effective areas of a plurality of orthophotos to be mosaicked; performing mosaicking processing multiple times on the basis of the respective effective areas of the plurality of orthophotos to obtain a target effective polygon group; and generating a seam line network for the plurality of orthophotos on the basis of the target effective polygon group.

Description

Mosaic line network generation method and device, electronic equipment and medium TECHNICAL FIELD

[0001] The present application relates to the technical fields of computer vision, image processing, and the like, and in particular to a mosaic line network generation method and device, electronic equipment, storage medium, and computer program product. BACKGROUND

[0002] Image mosaic is a process of splicing two or more orthographic images together to form a larger range image. The generation of a mosaic line network is a key step in image mosaic processing, and has an important influence on the quality and efficiency of image mosaic processing. SUMMARY

[0003] The present application provides a mosaic line network generation method and device, electronic equipment, storage medium, and computer program product.

[0004] According to one aspect of the present application, a mosaic line network generation method is provided, including: determining respective effective regions of a plurality of orthographic images to be mosaicked; performing a plurality of times of mosaic processing according to the respective effective regions of the plurality of orthographic images to obtain a target effective polygon group; and generating a mosaic line network for the plurality of orthographic images according to the target effective polygon group; wherein the mosaic processing includes: obtaining an nth starting image and an nth reference image used to perform an nth time of mosaic processing, and an (n-1)th effective polygon group obtained by the (n-1)th time of mosaic processing, wherein the nth starting image is obtained by merging the effective regions of all orthographic images participating in the previous (n-1) times of mosaic processing, the nth reference image is an orthographic image determined from the plurality of orthographic images that has not been mosaicked and has an intersection with the nth starting image, the (n-1)th effective polygon group includes effective mosaicked polygons of each orthographic image participating in the previous (n-1) times of mosaic processing, and n is an integer greater than 1; taking a preset direction of the nth reference image as a reference, constructing a directed skeleton line matched with the nth reference image in an effective overlap region between the nth starting image and the nth reference image to obtain a directed graph for the nth reference image, wherein the effective overlap region is determined according to the respective effective regions of the nth starting image and the nth reference image, the directed graph for the nth reference image indicates an effective mosaicked polygon of the nth reference image after the nth time of mosaic processing, and the preset direction of the nth reference image is used to indicate a tracking direction of a boundary of the effective region of the nth reference image; determining an nth effective polygon group corresponding to the nth time of mosaic processing based on the (n-1)th effective polygon group and the directed graph for the nth reference image; and in response to determining that there is no orthographic image that has not been mosaicked in the plurality of orthographic images after the nth time of mosaic processing, determining that the nth effective polygon group is the target effective polygon group.

[0005] According to another aspect of the present application, there is provided a mosaic line network generation apparatus, comprising: a determination module configured to determine respective effective areas of a plurality of orthographic images to be mosaicked; a mosaicking module configured to perform a plurality of times of mosaicking processing according to the respective effective areas of the plurality of orthographic images to obtain a target effective polygon group; and a generation module configured to generate a mosaic line network for the plurality of orthographic images according to the target effective polygon group; wherein the mosaicking module comprises: an acquisition unit configured to acquire an nth starting image and an nth reference image used for performing an nth time of mosaicking processing, and an (n-1)th effective polygon group obtained by performing an (n-1)th time of mosaicking processing, wherein the nth starting image is obtained by merging the effective areas of all orthographic images participating in the previous (n-1) times of mosaicking processing, the nth reference image is an orthographic image determined from the plurality of orthographic images and having no mosaicking and having an intersection with the nth starting image, and the (n-1)th effective polygon group comprises effective mosaicking polygons of respective orthographic images participating in the previous (n-1) times of mosaicking processing, and n is an integer greater than 1; a construction unit configured to take a preset direction of the nth reference image as a reference to construct a directed skeleton line matched with the nth reference image in an effective overlap area between the nth starting image and the nth reference image to obtain a directed graph for the nth reference image, wherein the effective overlap area is determined according to the respective effective areas of the nth starting image and the nth reference image, the directed graph for the nth reference image indicates effective mosaicking polygons of the nth reference image after the nth time of mosaicking processing, and the preset direction of the nth reference image is used to indicate a tracing direction of a boundary of the effective area of the nth reference image; a first determination unit configured to determine an nth effective polygon group corresponding to the nth time of mosaicking processing based on the (n-1)th effective polygon group and the directed graph for the nth reference image; and a second determination unit configured to determine that the nth effective polygon group is the target effective polygon group in response to determining that there is no orthographic image having no mosaicking in the plurality of orthographic images after the nth time of mosaicking processing.

[0006] According to another aspect of the present application, there is provided an electronic device, comprising: one or more processors; memory storing one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the method as described above.

[0007] According to another aspect of the present application, there is provided a computer-readable storage medium having stored thereon executable instructions that, when executed by a processor, cause the processor to perform the method as described above.

[0008] According to another aspect of the present application, there is provided a computer program product comprising a computer program which, when executed by a processor, performs the method as described above. BRIEF DESCRIPTION OF DRAWINGS

[0009] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort based on these drawings.

[0010] Fig. 1 is a flow chart of a method for generating a tessellation line network according to an embodiment of the present application;

[0011] Fig. 2A is a schematic diagram of determining a first group of valid polygons according to an embodiment of the present application;

[0012] Fig. 2B is a schematic diagram of determining a second group of valid polygons according to an embodiment of the present application;

[0013] Fig. 2C is a schematic diagram of a process of obtaining a tessellation image based on a target group of valid polygons according to an embodiment of the present application;

[0014] Fig. 3 is a schematic diagram of a process of constructing a two-dimensional triangular mesh according to an embodiment of the present application;

[0015] Fig. 4 is a schematic diagram of an effect of constructing a directed skeleton line segment in a triangular face with different attribute information according to an embodiment of the present application;

[0016] Fig. 5 is a schematic diagram of a process of constructing a directed skeleton line matching an nth reference image according to an embodiment of the present application;

[0017] Fig. 6 is a schematic diagram of a process of constructing a directed skeleton line matching an nth reference image according to another embodiment of the present application;

[0018] Fig. 7 is a schematic diagram of a process of constructing a directed skeleton line matching an nth reference image according to another embodiment of the present application;

[0019] Fig. 8 is a schematic diagram of a process of constructing a directed skeleton line matching an nth reference image according to another embodiment of the present application;

[0020] Fig. 9 is a schematic diagram of a process of constructing a directed skeleton line matching an nth reference image according to another embodiment of the present application;

[0021] Fig. 10 is a schematic diagram of a process of constructing a directed skeleton line matching an nth reference image according to another embodiment of the present application;

[0022] Fig. 11 is a schematic diagram of a process of constructing a directed skeleton line matching an nth reference image according to another embodiment of the present application;

[0023] FIG. 12 is a schematic diagram of a process of constructing a directional skeleton line matched with an n-th reference image according to another embodiment of the present application;

[0024] FIG. 13 is a schematic diagram of a process of constructing a directional skeleton line matched with an n-th reference image according to another embodiment of the present application;

[0025] FIG. 14 is a block diagram of a tessellation line network generation apparatus according to an embodiment of the present application;

[0026] FIG. 15 is a block diagram of an electronic device suitable for implementing a tessellation line network generation method according to an embodiment of the present application. DETAILED DESCRIPTION

[0027] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings of the embodiments of the present application and the embodiments. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application belong to the scope of protection of the present application.

[0028] It should be noted that the serial numbers of the operations in the following method are only used to represent the operations for description, and should not be regarded as representing the execution sequence of the operations. Unless explicitly stated, the method does not need to be executed in the order shown.

[0029] In addition, in the description of the present application, the terms "first", "second", "third" and the like (if any) are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0030] In the technical solutions of the present application, the collection, storage, use, processing, transmission, provision, disclosure and application of the data (for example, including but not limited to user personal information) involved all comply with the relevant legal regulations of the relevant countries and regions, and do not violate public order and good customs.

[0031] In addition, in the technical solutions of the present application, the authorization or consent of the data owner is obtained before obtaining or collecting relevant data.

[0032] FIG. 1 is a flowchart of a tessellation line network generation method according to an embodiment of the present application.

[0033] As shown in FIG. 1, the tessellation line network generation method 100 includes operations S110-S130.

[0034] In operation S110, the effective areas of a plurality of orthographic images to be tessellated are determined.

[0035] In operation S120, a plurality of tessellation processes are performed according to the effective areas of the plurality of orthographic images, and a target effective polygon group is obtained.

[0036] In operation S130, a mosaic line network for the plurality of orthographic images is generated according to the target effective polygon group.

[0037] In embodiments of the present application, each orthographic image to be mosaicked can be an orthographic image of various shapes and types, which are not limited by the present application. The type of the orthographic image can refer to whether the orthographic image is an orthographic image including a hole, for example. The shape of the orthographic image refers to the contour of the effective area of the orthographic image, which can be a quadrilateral or other polygon, and the like. In embodiments of the present application, an octant boundary tracking algorithm can be used to determine the effective area of each of the plurality of orthographic images to be mosaicked. In some embodiments, a Douglas-Peucker algorithm can also be used to thin the effective area of each of the plurality of orthographic images to be mosaicked, so as to simplify the effective area of each of the plurality of orthographic images to be mosaicked.

[0038] In embodiments of the present application, the target effective polygon group for mosaicking can be obtained by performing multiple mosaicking processes on the effective areas of the plurality of orthographic images. The target effective polygon group includes effective mosaicking polygons of each of the plurality of orthographic images after the multiple mosaicking processes, and according to the effective mosaicking polygons, the mosaic line network for the plurality of orthographic images can be generated.

[0039] According to embodiments of the present application, the mosaicking process in operation S120 can include operations S121-S124, which are described below.

[0040] In operation S121, an nth starting image and an nth reference image for performing an nth mosaicking process, and an (n-1)th effective polygon group obtained by an (n-1)th mosaicking process are obtained.

[0041] In operation S122, a directed skeleton line matching the nth reference image is constructed in an effective overlapping area between the nth starting image and the nth reference image with the direction of the nth reference image as a reference, so as to obtain a directed graph for the nth reference image.

[0042] In operation S123, an nth effective polygon group corresponding to the nth mosaicking process is determined based on the (n-1)th effective polygon group and the directed graph for the nth reference image.

[0043] In operation S124, in response to determining that there is no orthographic image that has not been mosaicked in the plurality of orthographic images after the nth mosaicking process, the nth effective polygon group is determined as the target effective polygon group.

[0044] According to an embodiment of the present application, the nth starting image is obtained by merging the effective regions of all orthographic images participating in the previous n-1 times of the tiling process, the nth reference image is an orthographic image determined from the plurality of orthographic images and not tiled and having intersection with the nth starting image, the n-1th effective polygon set comprises the effective tiling polygons of each orthographic image participating in the previous n-1 times of the tiling process, and n is an integer greater than 1.

[0045] For example, the 2th starting image for performing the 2th tiling process is obtained by merging the effective regions of all orthographic images participating in the 1th tiling process (denoted as the 1th starting image and the 1th reference image), wherein the 1th starting image is an orthographic image arbitrarily selected from the plurality of orthographic images to be tiled, and the 1th reference image is an orthographic image determined from the plurality of orthographic images and not tiled and having intersection with the 1th starting image. The 2th reference image is an orthographic image determined from the plurality of orthographic images and not tiled and having intersection with the 2th starting image. The 1th effective polygon set obtained by the 1th tiling process comprises the effective tiling polygons of each orthographic image participating in the 1th tiling process (i.e. the 1th starting image and the 1th reference image).

[0046] For another example, the 3th starting image for performing the 3th tiling process is obtained by merging the effective regions of all orthographic images participating in the previous 2 times of the tiling process (i.e. the 1th starting image, the 1th reference image and the 2th reference image). The 3th reference image is an orthographic image determined from the plurality of orthographic images and not tiled and having intersection with the 3th starting image. The 2th effective polygon set obtained by the 2th tiling process comprises the effective tiling polygons of each orthographic image participating in the previous 2 times of the tiling process (i.e. the 1th starting image, the 1th reference image and the 2th reference image).

[0047] By analogy, the nth starting image and the nth reference image for performing the nth tiling process, and the n-1th effective polygon set obtained by the n-1th tiling process can be acquired.

[0048] According to an embodiment of the present application, the effective overlap area between the nth initial image and the nth reference image refers to the overlapping part between the effective area of the nth initial image and the effective area of the nth reference image, which can be determined according to the effective area of each of the nth initial image and the nth reference image. For example, the effective overlap area between the 1st initial image and the 1st reference image is determined according to the effective area of each of the 1st initial image and the 1st reference image. The effective overlap area between the 2nd initial image and the 2nd reference image is determined according to the effective area of each of the 2nd initial image and the 2nd reference image. Since the 2nd initial image is obtained by merging the effective areas of all the orthographic images involved in the 1st tiling process (i.e. the 1st initial image and the 1st reference image), the effective overlap area between the 2nd initial image and the 2nd reference image can be considered as being determined according to the effective area of each of the 1st initial image, the 1st reference image and the 2nd reference image.

[0049] According to an embodiment of the present application, the preset direction of the nth reference image is used to indicate the tracing direction of the boundary of the effective area of the nth reference image. In each tiling process, the preset direction of the nth reference image can be used as a reference to uniquely determine the direction of the directional skeleton line matched with the nth reference image, so as to facilitate faster and better generation of the directional graph for the nth reference image, and further facilitate improvement of the generation efficiency and generation quality of the tiling line network.

[0050] After the effective overlap area between the nth initial image and the nth reference image is determined, a directional skeleton line matched with the nth reference image can be constructed within the effective overlap area between the nth initial image and the nth reference image, with the preset direction of the nth reference image as a reference. The directional skeleton line indicates the tiling line between the nth reference image and the nth initial image, and based on the directional skeleton line matched with the nth reference image, a directional graph for the nth reference image can be obtained. The directional graph for the nth reference image indicates the effective tiling polygon of the nth reference image after the nth tiling process.

[0051] Since the (n-1)th effective polygon set includes the effective tiling polygons of each orthographic image involved in the first n-1 tiling processes, and the directional graph for the nth reference image indicates the effective tiling polygon of the nth reference image after the nth tiling process, the (n)th effective polygon set corresponding to the nth tiling process can be determined by using the (n-1)th effective polygon set and the directional graph for the nth reference image. The (n)th effective polygon set includes the effective tiling polygons of each orthographic image involved in the nth tiling process.

[0052] In some embodiments, the following operation can be employed to determine the nth effective polygon group corresponding to the nth tiling process.

[0053] For example, the effective tiling polygon of the nth reference image after the nth tiling process can be determined according to the directed graph of the nth reference image. Then, the effective tiling polygon of each orthographic image in the (n-1)th effective polygon group is updated by removing the target overlap region from the effective tiling polygon of the orthographic image, where the target overlap region refers to the effective overlap region between the effective tiling polygon of the orthographic image and the effective tiling polygon of the nth reference image after the nth tiling process. Next, the effective tiling polygon of the nth reference image after the nth tiling process is added to the updated effective polygon group, thereby obtaining the nth effective polygon group.

[0054] After determining the nth effective polygon group corresponding to the nth tiling process, it can be determined whether there is an un-tiled orthographic image among the plurality of orthographic images after the nth tiling process. If it is determined that there is no un-tiled orthographic image among the plurality of orthographic images after the nth tiling process, the nth effective polygon group corresponding to the nth tiling process can be determined as the target effective polygon group.

[0055] In some embodiments, if it is determined that there is still an un-tiled orthographic image among the plurality of orthographic images after the nth tiling process, the effective region of the nth starting image and the effective region of the nth reference image can be merged to obtain an (n+1)th starting image for performing an (n+1)th tiling process. Then, an (n+1)th reference image for performing the (n+1)th tiling process is obtained, where the (n+1)th reference image is an un-tiled orthographic image determined from the plurality of orthographic images and having an intersection with the (n+1)th starting image. Next, the (n+1)th starting image, the (n+1)th reference image, and the nth effective polygon group are updated based on the (n+1)th starting image, the (n+1)th reference image, and the nth effective polygon group, and the operation of determining whether the nth effective polygon group is the target effective polygon group is repeated until the target effective polygon group is determined, or it is determined that the un-tiled orthographic image has no intersection with the updated (n+1)th starting image, and the above operations S121-S124 are stopped.

[0056] According to the technical scheme of the present application, in each time of mosaic processing, on the one hand, the n-th starting image and the n-th reference image are mosaicked by taking the direction of the preset n-th reference image as a reference, so that the direction of the directed skeleton line matched with the n-th reference image can be uniquely calibrated, thereby facilitating faster and better generation of the directed graph for the n-th reference image, and further facilitating improvement of the generation efficiency and generation quality of the mosaic line network; on the other hand, in the process of mosaicking the n-th starting image and the n-th reference image, the effective mosaic polygons of each orthographic image participating in the n times of mosaic processing can also be output, thereby ensuring the completeness, seamless and non-overlapping of the finally generated mosaic line network, and further facilitating improvement of the generation efficiency and generation quality of the mosaic line network.

[0057] FIG. 2A is a schematic diagram of determining a first effective polygon group according to an embodiment of the present application, FIG. 2B is a schematic diagram of determining a second effective polygon group according to an embodiment of the present application, and FIG. 2C is a schematic diagram of a process of obtaining a mosaic image based on a target effective polygon group according to an embodiment of the present application. The process of the orthographic image mosaic will be exemplarily described below with reference to FIGS. 2A, 2B and 2C. It should be noted that the number, shape and type of the orthographic images to be mosaicked shown in FIGS. 2A, 2B and 2C are only exemplary, and the embodiments of the present application are not limited thereto.

[0058] As shown in 201 of FIG. 2A, the plurality of orthographic images to be mosaicked include, for example, an orthographic image Polygon_b1, an orthographic image Polygon_a1 and an orthographic image Polygon_a2.

[0059] Firstly, the effective areas of the plurality of orthographic images to be mosaicked are determined, for example, the effective area of the orthographic image Polygon_b1, the effective area of the orthographic image Polygon_a1 and the effective area of the orthographic image Polygon_a2 are obtained. The effective area of the orthographic image Polygon_b1, the effective area of the orthographic image Polygon_a1 and the effective area of the orthographic image Polygon_a2 intersect with each other.

[0060] Next, one of the plurality of orthographic images is arbitrarily selected as a first starting image, for example, the orthographic image Polygon_b1 is selected as the first starting image. Then, one of the plurality of orthographic images which has not been mosaicked and has an intersection with the first starting image is determined as a first reference image, for example, the orthographic image Polygon_a1 which has an intersection with the orthographic image Polygon_b1 can be determined as the first reference image.

[0061] Next, a directed skeleton line matching the first reference image is constructed in the effective overlap region between the first reference image and the first start image, with the direction of the first reference image as a reference, to obtain a directed graph for the first reference image.

[0062] Please continue to refer to 201 in FIG. 2A, first, the effective overlap region ABCD between the first reference image (i.e. orthographic image Polygon_a1) and the first start image (i.e. orthographic image Polygon_b1) is determined according to the effective regions of the first reference image and the first start image respectively. Then, a directed skeleton line 21 matching the first reference image is constructed in the effective overlap region ABCD, with the direction of the first reference image as a reference (for example, the tracing direction of the outer boundary of the effective region of the first reference image is counterclockwise), to obtain a directed graph DG1 for the first reference image.

[0063] Next, based on the effective region of the first start image (i.e. orthographic image Polygon_b1) and the directed graph DG1 for the first reference image, a first effective polygon group corresponding to the first tiling process is determined.

[0064] As shown in 201 and 202 in FIG. 2A, the directed graph DG1 for the first reference image indicates the effective tiling polygon of the first reference image (i.e. orthographic image Polygon_a1) after the first tiling process. According to the directed graph DG1 for the first reference image, the effective tiling polygon S_a1_1 of the first reference image after the first tiling process can be determined. For example, the directed graph DG1 for the first reference image can be mapped into the first reference image (i.e. orthographic image Polygon_a1) to obtain the effective tiling polygon S_a1_1.

[0065] Next, the effective tiling polygon S_b1_1 of the first start image after the first tiling process is determined based on the positional relationship between the effective tiling polygon S_a1_1 of the first reference image after the first tiling process and the effective region of the first start image (i.e. orthographic image Polygon_b1). For example, the target overlap region M1 between the effective region of the first start image and the effective tiling polygon S_a1_1 can be removed from the effective region of the first start image to obtain the effective tiling polygon S_b1_1 of the first start image after the first tiling process.

[0066] Next, based on the effective mosaicked polygon S_a1_1 of the first reference image after the first mosaicking process and the effective mosaicked polygon S_b1_1 of the first starting image, a first effective polygon group Y1 corresponding to the first mosaicking process is obtained.

[0067] Next, it is determined whether there is an unmosicked orthographic image in the plurality of orthographic images after the first mosaicking process. If not, the first effective polygon group Y1 corresponding to the first mosaicking process can be determined as the target effective polygon group. If yes, the effective area of the first starting image and the effective area of the first reference image are merged to obtain a second starting image used for performing a second mosaicking process. Then, a second reference image used for performing the second mosaicking process is obtained. Next, the second mosaicking process is performed based on the second starting image, the second reference image, and the first effective polygon group Y1.

[0068] It can be understood that since there is still an unmosicked orthographic image Polygon_a2 in the plurality of orthographic images, the effective area of the first starting image (i.e., the orthographic image Polygon_b1) and the effective area of the first reference image (i.e., the orthographic image Polygon_a1) can be merged to obtain a merged image, denoted as Polygon_b2 (as shown in 203 in FIG. 2B). Then, the merged image Polygon_b2 is taken as the second starting image, and the orthographic image Polygon_a2 having an intersection with the second starting image (i.e., the merged image Polygon_b2) is taken as the second reference image. Next, the second mosaicking process is performed based on the second starting image, the second reference image, and the first effective polygon group Y1.

[0069] Next, with the direction of the preset second reference image as a reference, a directed skeleton line matching the second reference image is constructed in the effective overlapping area between the second reference image and the second starting image to obtain a directed graph for the second reference image.

[0070] Referring to 203 in FIG. 2B, first, the effective overlapping region EFAGHI between the second reference image and the second starting image is determined according to the effective regions of the second reference image (i.e., the orthographic image Polygon_a2) and the second starting image (i.e., the merged image Polygon_b2). Then, a directed skeleton line 22 matched with the second reference image is constructed in the above-mentioned effective overlapping region EFAGHI with reference to a preset direction of the second reference image (for example, the tracing direction of the outer boundary of the effective region of the second reference image is counterclockwise), so as to obtain a directed graph DG2 for the second reference image. It should be noted that the preset direction of the second reference image in the second embedding process can be the same as or different from the preset direction of the first reference image, which is set according to actual needs, and the present application does not make any limitation in this regard.

[0071] Next, a second effective polygon group corresponding to the second embedding process is determined based on the first effective polygon group Y1 and the directed graph DG2 for the second reference image.

[0072] As shown in 203 and 204 in FIG. 2B, the directed graph DG2 for the second reference image indicates the effective embedded polygon of the second reference image (i.e., the orthographic image Polygon_a2) after the second embedding process. According to the directed graph DG2 for the second reference image, the effective embedded polygon S_a2_2 of the second reference image after the second embedding process can be determined. Similarly, the directed graph DG2 for the second reference image can be mapped into the second reference image (i.e., the orthographic image Polygon_a2), so as to obtain the above-mentioned effective embedded polygon S_a2_2.

[0073] Next, the effective embedded polygon of each orthographic image in the first effective polygon group Y1 is obtained by removing the corresponding target overlapping region from the effective embedded polygon of the orthographic image, and an updated effective polygon group is obtained according to the effective embedded polygon of each orthographic image after the second embedding process. The corresponding target overlapping region refers to the effective overlapping region between the effective embedded polygon of the corresponding orthographic image in the first effective polygon group Y1 and the effective embedded polygon S_a2_2 of the second reference image after the second embedding process.

[0074] For example, for the effective mosaicking polygon S_a1_1 of the first reference image (i.e. orthographic image Polygon_a1) in the first effective polygon group Y1, the target overlapping area M21 is removed from the effective mosaicking polygon S_a1_1 of the first reference image, to obtain the effective mosaicking polygon S_a1_2 of the first reference image after the second mosaicking processing. The target overlapping area M21 is the effective overlapping area between the effective mosaicking polygon S_a1_1 of the first reference image (i.e. orthographic image Polygon_a1) after the first mosaicking processing and the effective mosaicking polygon S_a2_2 of the second reference image (i.e. orthographic image Polygon_a2) after the second mosaicking processing.

[0075] Similarly, for the effective mosaicking polygon S_b1_1 of the first starting image (i.e. orthographic image Polygon_b1) in the first effective polygon group Y1, the target overlapping area M22 is removed from the effective mosaicking polygon S_b1_1 of the first starting image, to obtain the effective mosaicking polygon S_b1_2 of the first starting image after the second mosaicking processing. The target overlapping area M22 is the effective overlapping area between the effective mosaicking polygon S_b1_1 of the first starting image (i.e. orthographic image Polygon_b1) after the first mosaicking processing and the effective mosaicking polygon S_a2_2 of the second reference image (i.e. orthographic image Polygon_a2) after the second mosaicking processing.

[0076] Based on the effective mosaicking polygon S_a1_2 of the first reference image after the second mosaicking processing and the effective mosaicking polygon S_b1_2 of the first starting image after the second mosaicking processing, an updated effective polygon group Y1’ is obtained.

[0077] Next, the effective mosaicking polygon S_a2_2 of the second reference image after the second mosaicking processing is added to the updated effective polygon group Y1’, to obtain a second effective polygon group Y2 corresponding to the second mosaicking processing.

[0078] Next, it is determined whether there is an un-tiled orthographic image in the plurality of orthographic images after the second tiling process. If not, the second effective polygon group Y2 corresponding to the second tiling process can be determined as the target effective polygon group. If yes, the effective area of the second starting image and the effective area of the second reference image are merged to obtain a third starting image for performing a third tiling process. Then, a third reference image for performing the third tiling process is obtained from the remaining orthographic images, and the third tiling process is performed based on the third starting image, the third reference image, and the second effective polygon group Y2 according to the above-described process; the operation is repeated until the target effective polygon group is determined, or it is determined that there is no intersection between the un-tiled orthographic image and the updated nth (n is an integer greater than 1) starting image, and the above-described operation is stopped.

[0079] According to the above description, after the second tiling process, there is no un-tiled orthographic image in the plurality of orthographic images (including the orthographic images Polygon_b1, Polygon_a1, and Polygon_a2), and therefore, the second effective polygon group Y2 corresponding to the second tiling process can be determined as the target effective polygon group.

[0080] In some embodiments, a tiling line network for the plurality of orthographic images can also be generated according to the target effective polygon group, and the plurality of orthographic images can be tiled based on the tiling line network to obtain a tiled image.

[0081] As shown in FIG. 2C, after the target effective polygon group (i.e., the second effective polygon group Y2) is obtained, the common edges between adjacent effective tiling polygons in the target effective polygon group can be determined, and a tiling line network can be generated according to the common edges between all adjacent effective tiling polygons.

[0082] For example, the common edge between the effective tiling polygon S_a1_2 of the first reference image after the second tiling process and the effective tiling polygon S_b1_2 of the first starting image after the second tiling process can be determined to obtain the common edge La1_b1. Similarly, the common edge La1_a2 between the effective tiling polygon S_a1_2 of the first reference image after the second tiling process and the effective tiling polygon S_a2_2 of the second reference image after the second tiling process, and the common edge Lb1_a2 between the effective tiling polygon S_b1_2 of the first starting image after the second tiling process and the effective tiling polygon S_a2_2 of the second reference image after the second tiling process can be obtained.

[0083] Then, according to the common edges La1_b1, La1_a2 and Lb1_a2, a tessellation line network for a plurality of orthographic images (including orthographic images Polygon_b1, Polygon_a1 and Polygon_a2) can be generated.

[0084] Next, based on the tessellation line network for the plurality of orthographic images, the plurality of orthographic images are tessellated to obtain a tessellated image. As shown in FIG. 2C, the orthographic images Polygon_b1, Polygon_a1 and Polygon_a2 can be tessellated according to the above-mentioned tessellation line network to obtain a tessellated image Polygon_X.

[0085] In an embodiment of the present application, in the operation S122 described above, the direction of the preset nth reference image is taken as a reference, and a directed skeleton line matched with the nth reference image is constructed in the effective overlapping region between the nth starting image and the nth reference image to obtain a directed graph for the nth reference image. For example, the operation can include the following operations. It should be noted that the scheme of constructing the directed skeleton line matched with the nth reference image in the effective overlapping region is applicable to any one-time tessellation process (including the first-time tessellation process), and will not be described in detail hereinafter.

[0086] Firstly, based on the constrained Delaunay triangulation algorithm, a two-dimensional triangular mesh is constructed according to all the vertices in the effective overlapping region between the nth starting image and the nth reference image. In the two-dimensional triangular mesh, each triangular face contains at most one intersection point, and the intersection point refers to the intersection point between the boundary of the effective region of the nth starting image and the boundary of the effective region of the nth reference image. It should be noted that, unless otherwise specified, the intersection point referred to in the present application refers to the intersection point between the boundary of the effective region of the nth starting image and the boundary of the effective region of the nth reference image, and the effective overlapping region refers to the effective overlapping region between the nth starting image and the nth reference image.

[0087] The process of constructing the two-dimensional triangular mesh will be described below with reference to FIG. 3.

[0088] Firstly, based on the Delaunay triangulation algorithm, an initial two-dimensional triangular mesh is constructed according to all the vertices in the effective overlapping region between the nth starting image and the nth reference image.

[0089] As shown in FIG. 3, first determine the effective overlapping area EFAGHI between the nth initial image Polygon_bn and the nth reference image Polygon_an, all the vertices in the effective overlapping area EFAGHI include vertices E, F, A, G, H, and I. Then, construct an initial two-dimensional triangular mesh based on all the vertices in the effective overlapping area EFAGHI.

[0090] Next, determine each intersection point in the effective overlapping area between the nth initial image and the nth reference image. Please continue to refer to FIG. 3, all the intersection points in the effective overlapping area EFAGHI include intersection points E and H.

[0091] Next, for any one intersection point, construct a constraint edge corresponding to the intersection point in the above effective overlapping area. The constraint edge is connected according to the first constraint point and the second constraint point associated with the intersection point. Wherein, the first constraint point and the second constraint point are respectively any point on the two boundary line segments of the effective overlapping area where the intersection point is located, except the intersection point. In the embodiments of the present application, in addition to the boundary line segments where the first constraint point and the second constraint point are located respectively, the constraint edge does not intersect with other boundary line segments in the effective overlapping area. Then, update the initial two-dimensional triangular mesh based on the constraint edge to obtain an updated two-dimensional triangular mesh. Next, repeat the operation of updating the initial two-dimensional triangular mesh based on the constraint edge until the constraint processing for all intersection points is completed, and obtain the two-dimensional triangular mesh.

[0092] As shown in FIG. 3, for example, first for the intersection point H, construct a constraint edge a1b1 corresponding to the intersection point H in the effective overlapping area EFAGHI. Wherein, the constraint edge a1b1 is connected according to the first constraint point a1 and the second constraint point b1, the first constraint point a1 and the second constraint point b1 are respectively any point on the two boundary line segments of the effective overlapping area EFAGHI where the intersection point H is located, except the intersection point H, for example, the first constraint point a1 is any point on the boundary line segment HI except the intersection point H, and the second constraint point b1 is any point on the boundary line segment HG except the intersection point H. In some embodiments, the first constraint point a1 and the second constraint point b1 may, for example, be the midpoints of the boundary line segments HI and HG respectively. In addition to the boundary line segments (i.e. HI and HG) where the first constraint point a1 and the second constraint point b1 are located respectively, the constraint edge a1b1 does not intersect with other boundary line segments in the effective overlapping area EFAGHI.

[0093] After constructing the constraint edge a1b1 corresponding to the intersection point H, the initial two-dimensional triangular mesh can be updated based on the constraint edge a1b1 to obtain an updated two-dimensional triangular mesh.

[0094] Next, for the intersection point E, a constraint edge a2b2 corresponding to the intersection point E is constructed in the effective overlap region EFAGHI. The constraint edge a2b2 is connected according to a first constraint point a2 and a second constraint point b2, for example, any point on two boundary line segments (i.e., EI and EF) of the effective overlap region EFAGHI where the intersection point E is located except the intersection point E. In some embodiments, the first constraint point a2 and the second constraint point b2 can be the midpoints of the boundary line segments EI and EF, respectively. In addition to the boundary line segments (i.e., EI and EF) where the first constraint point a2 and the second constraint point b2 are located, the constraint edge a2b2 does not intersect with other boundary line segments in the effective overlap region EFAGHI.

[0095] After the constraint edge a2b2 corresponding to the intersection point E is constructed, the updated two-dimensional triangular mesh can be updated based on the constraint edge a2b2 to obtain another updated two-dimensional triangular mesh. Since the effective overlap region EFAGHI only includes the intersection points E and H, the construction of the two-dimensional triangular mesh is completed after the constraint processing for the intersection points H and E is sequentially completed. The obtained another updated two-dimensional triangular mesh is the final two-dimensional triangular mesh.

[0096] It should be noted that the order of the constraint processing for each intersection point in the construction of the two-dimensional triangular mesh is not limited to the above example, and can be set according to actual needs, which is not limited by the present application.

[0097] After the two-dimensional triangular mesh is constructed, next, a directed skeleton line matched with the nth reference image is constructed in the effective overlap region according to the two-dimensional triangular mesh with the direction of the preset nth reference image as a reference to obtain a directed graph for the nth reference image.

[0098] The process of constructing the directed skeleton line matched with the nth reference image will be described below with reference to specific embodiments.

[0099] First, attribute information of each triangular face in the two-dimensional triangular mesh is determined, which indicates the source of each vertex constituting the corresponding triangular face.

[0100] Next, with the direction of the preset nth reference image as a reference, a directed skeleton line segment is constructed in the corresponding triangular face in the effective overlap region according to the attribute information of each triangular face to obtain the directed skeleton line matched with the nth reference image.

[0101] It can be understood that the source of the three vertices constituting the triangular face may include, for example, the following cases. It should be noted that the following first boundary point refers to any point on the boundary of the effective region of the nth reference image except the intersection point, and the second boundary point refers to any point on the boundary of the effective region of the nth starting image except the intersection point, and the following will not be repeated.

[0102] The first case is that the three vertices constituting the triangular face include one first boundary point from the nth reference image and two second boundary points from the nth starting image.

[0103] The second case is that the three vertices constituting the triangular face include two first boundary points from the nth reference image and one second boundary point from the nth starting image.

[0104] The third case is that the three vertices constituting the triangular face include one first boundary point from the nth reference image, one second boundary point from the nth starting image, and one intersection point.

[0105] The fourth case is that the three vertices constituting the triangular face include three first boundary points from the nth reference image.

[0106] The fifth case is that the three vertices constituting the triangular face include three second boundary points from the nth starting image.

[0107] In other words, there are the above five possibilities for the attribute information of each triangular face in the two-dimensional triangular mesh. In the embodiment of the present application, different construction results can be obtained when the directed skeleton line segment is constructed in the triangular face with different attribute information.

[0108] The construction of the directed skeleton line segment in the triangular face will be introduced below in conjunction with FIG. 4. It should be noted that the vertices a, b, and c in FIG. 4 represent that the vertex is a first boundary point from the nth reference image, a second boundary point from the nth starting image, and an intersection point, respectively.

[0109] In one example, for any one triangular face, in response to determining that the attribute information of the triangular face is the first attribute information, the first attribute information indicates that the three vertices constituting the triangular face include one first boundary point from the nth reference image and two second boundary points from the nth starting image, the midpoint of the line segment constituted by the first boundary point and one second boundary point is determined as the first entry point, the midpoint of the line segment constituted by the first boundary point and the other second boundary point is determined as the first exit point, and then a first directed skeleton line segment is constructed between the first entry point and the first exit point with the preset direction of the nth reference image as a reference. The first directed skeleton line segment is used to divide the two second boundary points from the nth starting image to the inside of the effective tessellation polygon of the nth reference image after the nth tessellation processing.

[0110] As shown in 401 of FIG. 4, for any one triangular face in the two-dimensional triangular mesh, when the three vertices constituting the triangular face include one first boundary point a from the nth reference image and two second boundary points b from the nth starting image, the first entry point r1 and the first exit point q1 can be determined according to the method described in the above example, and the first directed skeleton line segment described above is constructed between the first entry point r1 and the first exit point q1.

[0111] It can be understood that, since the direction of the first directed skeleton line segment is determined according to the direction of the preset nth reference image, when there are two possibilities for the direction of the preset nth reference image (for example, clockwise direction or counterclockwise direction), correspondingly, there are also two possibilities for the direction of the first directed skeleton line segment. Exemplarily, when the direction of the preset nth reference image is used to indicate the tracing direction of the outer boundary of the effective region of the nth reference image as the counterclockwise direction, the first directed skeleton line segment constructed may be, for example, the first directed skeleton line segment 4111; when the direction of the preset nth reference image is used to indicate the tracing direction of the outer boundary of the effective region of the nth reference image as the clockwise direction, the first directed skeleton line segment constructed may be, for example, the first directed skeleton line segment 4112. In the embodiment of the application, no matter what the direction of the preset nth reference image is, the first directed skeleton line segment (for example, 4111 or 4112) constructed is used to divide the two second boundary points b from the nth starting image to the inside of the effective tessellation polygon of the nth reference image after the nth tessellation processing.

[0112] In another example, for any one triangular face, in response to determining that the attribute information of the triangular face is second attribute information, the second attribute information indicates that the three vertices constituting the triangular face include two first boundary points from the nth reference image and one second boundary point from the nth starting image, the midpoint of the line segment constituted by one first boundary point and the second boundary point is determined as a second entry point, the midpoint of the line segment constituted by the other first boundary point and the second boundary point is determined as a second exit point, and then the second directed skeleton line segment is constructed between the second entry point and the second exit point with the direction of the preset nth reference image as a reference, the second directed skeleton line segment is used to divide one second boundary point from the nth starting image to the inside of the effective tessellation polygon of the nth reference image after the nth tessellation processing.

[0113] As shown in 402 of FIG. 4, for any one triangular face in the two-dimensional triangular mesh, when the three vertices constituting the triangular face include two first boundary points a from the nth reference image and one second boundary point b from the nth starting image, the second entry point r2 and the second exit point q2 can be determined according to the method described in the above example, and the second directed skeleton line segment described above is constructed between the second entry point r2 and the second exit point q2.

[0114] Similarly, when the preset direction of the nth reference image is used to indicate the tracing direction of the outer boundary of the effective region of the nth reference image as the counterclockwise direction, the constructed second directed skeleton line segment can be, for example, the second directed skeleton line segment 4113; when the preset direction of the nth reference image is used to indicate the tracing direction of the outer boundary of the effective region of the nth reference image as the clockwise direction, the constructed second directed skeleton line segment can be, for example, the second directed skeleton line segment 4114. The second directed skeleton line segment 4113 or 4114 is used to divide one second boundary point b from the nth starting image to the inside of the effective tessellation polygon of the nth reference image after the nth tessellation processing.

[0115] In another example, for any one triangular face, in response to determining that the attribute information of the triangular face is third attribute information, the third attribute information indicates that the three vertices constituting the triangular face include one first boundary point from the nth reference image, one second boundary point from the nth starting image, and an intersection point, the midpoint of the line segment constituted by the first boundary point and the second boundary point is determined as the third entry point or the third exit point, and then the third directed skeleton line segment is constructed between the third entry point and the intersection point or between the third exit point and the intersection point with the preset direction of the nth reference image as the reference, the third directed skeleton line segment is used to divide one second boundary point from the nth starting image to the inside of the effective tessellation polygon of the nth reference image after the nth tessellation processing.

[0116] As shown in 403 of FIG. 4, for any one triangular face in the two-dimensional triangular mesh, when the three vertices constituting the triangular face include one first boundary point a from the nth reference image, one second boundary point b from the nth starting image, and an intersection point c, the third exit point q3 or the third entry point r3 can be determined according to the method described in the above example, and the third directed skeleton line segment is constructed between the third exit point q3 and the intersection point c or between the third entry point r3 and the intersection point c.

[0117] Similarly, when the direction of the preset nth reference image is used to indicate the tracing direction of the outer boundary of the effective region of the nth reference image is counterclockwise, the third directed skeleton line segment constructed may be, for example, the third directed skeleton line segment 4115; when the direction of the preset nth reference image is used to indicate the tracing direction of the outer boundary of the effective region of the nth reference image is clockwise, the third directed skeleton line segment constructed may be, for example, the third directed skeleton line segment 4116. The third directed skeleton line segment 4115 or 4116 is used to divide a second boundary point b from the nth starting image to the inside of the effective tessellation polygon of the nth reference image after the nth tessellation processing.

[0118] In another example, for any one triangular face, in response to determining that the attribute information of the triangular face is fourth attribute information, the fourth attribute information indicates that the three vertices constituting the triangular face include three first boundary points from the nth reference image, no directed skeleton line segment is constructed in the triangular face. As shown in 404 in FIG. 4, for any one triangular face in the two-dimensional triangular mesh, when the three vertices constituting the triangular face include three first boundary points a from the nth reference image, no directed skeleton line segment is constructed in the triangular face.

[0119] In another example, for any one triangular face, in response to determining that the attribute information of the triangular face is fifth attribute information, the fifth attribute information indicates that the three vertices constituting the triangular face include three second boundary points from the nth starting image, no directed skeleton line segment is constructed in the triangular face. As shown in 405 in FIG. 4, for any one triangular face in the two-dimensional triangular mesh, when the three vertices constituting the triangular face include three second boundary points b from the nth starting image, no directed skeleton line segment is constructed in the triangular face.

[0120] In some embodiments, in the process of constructing the directed skeleton line matched with the nth reference image in the effective overlapping region, if it is determined that the out-point of the directed skeleton line segment constructed in any one of the triangular faces is an intersection point, the direction of the preset nth reference image is used as a reference, and the intersection point is used as a starting point to trace along the boundary of the effective region of the nth reference image until another intersection point is traced. Then, the above-mentioned scheme of constructing the directed skeleton line segment is used to continue to construct the directed skeleton line segment.

[0121] In the embodiments of the present application, the above-mentioned nth reference image and nth starting image at least include one of the following.

[0122] Firstly, the nth reference image and the nth starting image are both orthographic images without holes.

[0123] Second, one of the nth reference image and the nth initial image is an orthoimage including at least one first hole.

[0124] Third, the nth reference image is an orthoimage including at least one second hole, the nth initial image is an orthoimage including at least one third hole, and the at least one second hole and the at least one third hole are completely disjoint.

[0125] Fourth, the nth reference image is an orthoimage including at least one fourth hole, the nth initial image is an orthoimage including at least one fifth hole, and the at least one fourth hole and the at least one fifth hole at least partially intersect.

[0126] Fifth, the nth reference image is an orthoimage including at least one sixth hole, the nth initial image is an orthoimage including at least one seventh hole, and the at least one sixth hole includes at least one sixth hole that encloses the seventh hole.

[0127] Sixth, the nth reference image is an orthoimage including at least one eighth hole, the nth initial image is an orthoimage including at least one ninth hole, and the at least one ninth hole includes at least one ninth hole that encloses the eighth hole.

[0128] It should be understood that the nth reference image and the nth initial image are not limited to the above embodiments, and for the sake of brevity, are not listed here. In addition, when the nth reference image and / or the nth initial image is an orthoimage including at least one hole, the hole in each image can at least partially be in the overlap region between the nth reference image and the nth initial image. Here, the overlap region refers to the portion common to the nth reference image and the nth initial image in geographic space. Even if there is a hole in the image, the overlap region still includes all the overlapping pixels in the two images, regardless of their quality and whether they are continuous. In the embodiments of the present application, the effective overlap region is a subset of the overlap region, which refers to the continuous region in the overlap region, which generally excludes those pixel regions that are not suitable for further analysis due to cloud cover, shadows, obstructions or other quality problems. In the case of holes in the image, the effective overlap region only includes those continuous regions that are not separated by holes, and the pixel quality of these regions is high enough to be used for subsequent analysis and processing.

[0129] It can be understood that different types of the nth reference image and the nth initial image correspond to different effective overlapping areas. Based on the above described scheme for constructing the directed skeleton line, corresponding two-dimensional triangular meshes can be constructed in the effective overlapping areas, and then the corresponding directed skeleton lines are obtained. The following will be described with specific examples. For the convenience of description, the effective areas of each nth reference image and each nth initial image and the areas corresponding to each hole are exemplarily represented by quadrilaterals in the following examples, and the number of holes (if any) contained in each image is exemplarily taken as 1. It should be understood that the scheme of the present application is not limited thereto. In addition, unless otherwise specified, the overlapping area referred to below refers to the overlapping area between the nth initial image and the nth reference image.

[0130] FIG. 5 is a schematic diagram of a process for constructing a directed skeleton line matched with the nth reference image according to an embodiment of the present application. As shown in FIG. 5, the nth reference image Polygon_an and the nth initial image Polygon_bn are both orthographic images without holes. The following will briefly introduce the process for constructing the directed skeleton line with reference to FIG. 5.

[0131] First, the effective overlapping area between the nth initial image Polygon_bn and the nth reference image Polygon_an is determined, which is the same as the overlapping area ABCD between the nth initial image Polygon_bn and the nth reference image Polygon_an. Then, based on the above scheme for constructing the two-dimensional triangular mesh, a two-dimensional triangular mesh is constructed in the effective overlapping area.

[0132] Next, with the preset direction of the nth reference image Polygon_an as a reference, a directed skeleton line matched with the nth reference image Polygon_an is constructed in the effective overlapping area according to the two-dimensional triangular mesh.

[0133] For example, the attribute information of each triangular face in the two-dimensional triangular mesh is determined first. Exemplarily, the attribute information of the triangular faces Ca51b51, a51b51b52, a51a52b52, Aa52b52, Bb51b52 and Da51a52 is determined as the third attribute information, the first attribute information, the second attribute information, the third attribute information, the fifth attribute information and the fourth attribute information, respectively.

[0134] Then, with the preset direction of the nth reference image Polygon_an as a reference, a directed skeleton line segment is constructed in the corresponding triangular face in the effective overlapping area according to the attribute information of each triangular face, so as to obtain a directed skeleton line matched with the nth reference image Polygon_an.

[0135] In the embodiment of the present application, in the case that the nth reference image Polygon_an is a orthographic image without holes, the preset direction of the nth reference image Polygon_an includes a first direction of the outer boundary (for example, the boundary of the boundary points A, C, D) of the effective area of the nth reference image Polygon_an, and the first direction indicates the tracing direction of the outer boundary of the effective area of the nth reference image Polygon_an. The first direction can be, for example, a counterclockwise direction, or a clockwise direction, and can be set according to actual needs.

[0136] As shown in FIG. 5, in the example of the first direction being a counterclockwise direction, based on the attribute information of each of the triangular faces, a directed skeleton line segment is constructed in the corresponding triangular face in the effective overlap area, for example, the directed skeleton line segments Cq51, q51q52, q52q53, q53A are constructed in the triangular faces Ca51b51, a51b51b52, a51a52b52, Aa52b52 in sequence, and the directed skeleton line 51 matched with the nth reference image Polygon_an is obtained according to the directed skeleton line segments. Similarly, in the example of the first direction being a clockwise direction, based on the attribute information of each of the triangular faces, a directed skeleton line segment is constructed in the corresponding triangular face in the effective overlap area (i.e., the triangular faces Aa52b52, a51a52b52, a51b51b52, Ca51b51), so as to obtain the directed skeleton line 51' matched with the nth reference image Polygon_an. Based on the directed skeleton line 51 or 51', the directed graph for the nth reference image Polygon_an is obtained, and then the effective tessellation polygon of the nth reference image Polygon_an after the nth tessellation processing is obtained. As can be seen from FIG. 5, no matter whether the first direction is a clockwise direction or a counterclockwise direction, the directed skeleton line segment constructed in the corresponding triangular face can divide the second boundary point from the nth starting image Polygon_bn to the inside of the effective tessellation polygon of the nth reference image Polygon_an after the nth tessellation processing.

[0137] According to the above-mentioned method of constructing the directed skeleton line in the triangular face, it can be seen that, in the case that the attribute information of the triangular face is determined, the position of the directed skeleton line constructed in the triangular face does not change although the direction of the directed skeleton line changes correspondingly with the direction (for example, the first direction) of the preset nth reference image Polygon_an. Therefore, when the first direction is changed from the counterclockwise direction to the clockwise direction, the position of the directed skeleton line composed of the directed skeleton lines does not change, and only the direction of the directed skeleton line is changed, i.e., the direction from the intersection point C to the intersection point A is changed to the direction from the intersection point A to the intersection point C. It can be seen that the above-mentioned method of constructing the directed skeleton line can ensure the uniqueness of the constructed directed skeleton line.

[0138] It can be understood that the influence of the direction of the preset nth reference image on the direction of the directed skeleton line is also applicable to the embodiments of constructing the directed skeleton line by using other types of nth reference image and nth starting image, and the details are not described herein again in order to save space.

[0139] FIG. 6 is a schematic diagram of a process of constructing the directed skeleton line matched with the nth reference image according to another embodiment of the present application. As shown in FIG. 6, the nth reference image Polygon_an is an orthographic image including at least one hole, and the nth starting image Polygon_bn is an orthographic image not including the hole. The process of constructing the directed skeleton line is briefly introduced below with reference to FIG. 6.

[0140] Firstly, the effective overlapping area between the nth starting image Polygon_bn and the nth reference image Polygon_an is determined, i.e., the effective area in the overlapping area ABCD between the nth starting image Polygon_bn and the nth reference image Polygon_an except the hole JKLM. The hole JKLM is the hole in the nth reference image Polygon_an. Then, the two-dimensional triangular mesh is constructed in the effective overlapping area based on the above-mentioned method of constructing the two-dimensional triangular mesh.

[0141] Next, the attribute information of each triangular face in the two-dimensional triangular mesh is determined. For example, the attribute information of the triangular faces Ca61b61, a61b61M is determined as the third attribute information and the second attribute information respectively.

[0142] Next, the direction of the preset nth reference image Polygon_an is taken as a reference, and the directed skeleton line segment is constructed in the corresponding triangular face in the effective overlapping area according to the attribute information of each triangular face, so as to obtain the directed skeleton line matched with the nth reference image Polygon_an.

[0143] According to the embodiment of the present application, in the case that the nth reference image Polygon_an is a orthographic image including at least one hole, the preset direction of the nth reference image Polygon_an includes a first direction for the outer boundary (e.g. the boundary where the boundary points A, C, D are located) of the effective area of the nth reference image Polygon_an and a second direction for the inner boundary of the effective area of the nth reference image Polygon_an. The inner boundary is determined according to the boundary (e.g. the boundary where the boundary points J, K, L, M are located) of the hole (e.g. the hole JKLM) in the nth reference image Polygon_an. The first direction can be, for example, a counterclockwise direction or a clockwise direction, which can be set according to actual needs. The second direction is opposite to the first direction. For example, when the first direction is counterclockwise, the second direction is clockwise; when the first direction is clockwise, the second direction is counterclockwise.

[0144] As shown in FIG. 6, in the case that the first direction is counterclockwise and the second direction is clockwise, according to the above scheme of constructing the directed skeleton line, the directed skeleton line 61 can be constructed in the corresponding triangular surface in the effective overlapping area shown in FIG. 6. By comparing FIG. 5 and FIG. 6, it can be seen that when the overlapping area ABCD contains a hole, the effective overlapping area will change, thereby affecting the attribute information of the corresponding triangular surface in the two-dimensional triangular mesh, and further causing the position of the directed skeleton line to change.

[0145] FIG. 7 is a process diagram of constructing a directed skeleton line matched with the nth reference image according to another embodiment of the present application. Compared with the embodiment shown in FIG. 6, the hole EFGH in the overlapping area ABCD between the nth starting image Polygon_bn and the nth reference image Polygon_an in FIG. 7 is a hole in the nth starting image Polygon_bn. Correspondingly, the effective overlapping area between the nth starting image Polygon_bn and the nth reference image Polygon_an refers to the effective area in the overlapping area ABCD excluding the hole EFGH.

[0146] In the example shown in FIG. 7, the preset direction of the nth reference image Polygon_an includes a first direction of the outer boundary (e.g., the boundary where the boundary points A, C, and D are located) of the effective area of the nth reference image Polygon_an. As shown in FIG. 7, in the example where the first direction is a clockwise direction, the directed skeleton line 71 is constructed in the corresponding triangular face in the effective overlapping area shown in FIG. 7 according to the above scheme for constructing a directed skeleton line. By comparing FIG. 6 and FIG. 7, it can be seen that when the holes contained in the overlapping area ABCD belong to different images, the attribute information of the corresponding triangular face in the two-dimensional triangular mesh is affected, which further causes the position of the directed skeleton line to change.

[0147] FIG. 8 is a schematic diagram of a process for constructing a directed skeleton line according to another embodiment of the present application. Compared with the embodiment shown in FIG. 7, the hole EFGH in the nth starting image Polygon_bn in FIG. 8 is partially located in the overlapping area ABCD (as shown by OQGH in FIG. 8). Correspondingly, the effective overlapping area between the nth starting image Polygon_bn and the nth reference image Polygon_an refers to the effective area in the overlapping area ABCD excluding the part of the hole OQGH that is located in the overlapping area ABCD.

[0148] In the example shown in FIG. 8, the preset direction of the nth reference image Polygon_an includes a first direction of the outer boundary (e.g., the boundary where the boundary points A, C, and D are located) of the effective area of the nth reference image Polygon_an.

[0149] As shown in FIG. 8, in the example where the first direction is a counterclockwise direction, the directed skeleton line 81 is constructed in the corresponding triangular face in the effective overlapping area shown in FIG. 8 according to the above scheme for constructing a directed skeleton line. It should be noted that when the starting point of the directed skeleton line segment constructed in a triangular face is an intersection point, the direction of the nth reference image Polygon_an can be taken as a reference, and the intersection point can be taken as a starting point to track along the boundary of the effective area of the nth reference image until another intersection point is tracked, and then the above scheme for constructing a directed skeleton line segment can be used to continue to construct a directed skeleton line segment. For example, the first direction (i.e., O→Q→A) can be taken as a reference, and the intersection point O can be taken as a starting point to track along the boundary of the effective area of the nth reference image Polygon_an (i.e., the boundary where the points O, Q, and A are located) until another intersection point Q is tracked according to the above method in FIG. 8. Then, the directed skeleton line segment is continued to be constructed in the triangular face Qa 83b 83 until the directed skeleton line is constructed. By comparing FIG. 7 and FIG. 8, it can be seen that when the hole part is located in the overlapping area, the attribute information of the corresponding triangular face in the two-dimensional triangular mesh is also affected, which further causes the position of the directed skeleton line to change.

[0150] FIG. 9 is a schematic diagram of a process of constructing a directed skeleton line according to another embodiment of the present application. As shown in FIG. 9, the nth reference image Polygon_an and the nth start image Polygon_bn are both orthographic images including at least one hole, the at least one hole in the nth reference image Polygon_an is completely disjointed with the at least one hole in the nth start image Polygon_bn, and the holes are both in the overlapping region ABCD. The process of constructing a directed skeleton line is briefly introduced below with reference to FIG. 9.

[0151] First, the effective overlapping region between the nth start image Polygon_bn and the nth reference image Polygon_an is determined, i.e., the effective region in the overlapping region ABCD excluding the hole EFGH in the nth start image Polygon_bn and the hole JKLM in the nth reference image Polygon_an. Then, a two-dimensional triangular mesh is constructed in the effective overlapping region based on the above scheme of constructing a two-dimensional triangular mesh.

[0152] Next, the direction of the preset nth reference image Polygon_an is taken as a reference, and a directed skeleton line matching the nth reference image Polygon_an is constructed in the effective overlapping region according to the two-dimensional triangular mesh.

[0153] For example, the attribute information of each triangular face in the two-dimensional triangular mesh is determined first. For example, the attribute information of the triangular face Ca91b91 is determined as the third attribute information, the attribute information of the triangular face a91b91M is determined as the second attribute information, and so on.

[0154] Then, the direction of the preset nth reference image Polygon_an is taken as a reference, and a directed skeleton line segment is constructed in the corresponding triangular face in the effective overlapping region according to the attribute information of each triangular face, so as to obtain a directed skeleton line matching the nth reference image Polygon_an.

[0155] In the example shown in FIG. 9, the preset direction of the nth reference image Polygon_an includes a first direction for the outer boundary (e.g., the boundary where the boundary points A, C, D are located) of the effective area of the nth reference image Polygon_an and a second direction for the inner boundary of the effective area of the nth reference image Polygon_an. The inner boundary is determined according to the boundary of the hole (e.g., the hole JKLM) in the nth reference image Polygon_an. The first direction and the second direction are opposite. For example, when the first direction is counterclockwise, the second direction is clockwise, in other words, when the tracking direction of the outer boundary (i.e., the boundary where the boundary points A, C, D are located) of the effective area of the nth reference image Polygon_an is A→C→D→A, the tracking direction of the corresponding inner boundary is J→M→L→K→J.

[0156] As shown in FIG. 9, in the example where the first direction is counterclockwise and the second direction is clockwise, based on the attribute information of each triangle face, a directed skeleton line segment is constructed in the corresponding triangle face in the effective overlap area, for example, a directed skeleton line segment is sequentially constructed in the triangle faces Ca91b91, a91b91M, b91MB, and so on, and a directed skeleton line 91 matched with the nth reference image Polygon_an is obtained according to the directed skeleton line segments.

[0157] FIG. 10 is a schematic diagram of a process of constructing a directed skeleton line matched with the nth reference image according to another embodiment of the present application. Compared with the embodiment shown in FIG. 9, the hole JKLM in the nth reference image Polygon_an partially intersects with the hole EFGH in the nth starting image Polygon_bn in FIG. 10, and the intersecting part of the hole is in the overlap area ABCD. Correspondingly, the effective overlap area between the nth starting image Polygon_bn and the nth reference image Polygon_an refers to the effective area in the overlap area ABCD except the hole EFGH and the hole JKLM.

[0158] Similarly, in the example shown in FIG. 10, the preset direction of the nth reference image Polygon_an includes a first direction for the outer boundary (e.g., the boundary where the boundary points A, C, D are located) of the effective area of the nth reference image Polygon_an and a second direction for the inner boundary of the effective area of the nth reference image Polygon_an. The inner boundary is determined according to the boundary of the hole (e.g., the hole JKLM) in the nth reference image Polygon_an.

[0159] As shown in FIG. 10, in an example where the first direction is counterclockwise and the second direction is clockwise, according to the above scheme of constructing the directional skeleton line, the directional skeleton line 101 is constructed in the corresponding triangular face in the effective overlapping region shown in FIG. 10. By comparing FIG. 9 and FIG. 10, it can be seen that when the cavities belonging to different images intersect in the overlapping region ABCD, the attribute information of the corresponding triangular face in the two-dimensional triangular mesh is also affected, which further causes the position of the directional skeleton line to change.

[0160] FIG. 11 is a schematic diagram of a process of constructing a directional skeleton line matched with the nth reference image according to another embodiment of the present application. Compared with the embodiment shown in FIG. 9, in FIG. 11, the cavity EFGH in the nth starting image Polygon_bn covers the cavity JKLM in the nth reference image Polygon_an. Correspondingly, the effective overlapping region between the nth starting image Polygon_bn and the nth reference image Polygon_an refers to the effective region in the overlapping region ABCD except the cavity EFGH.

[0161] In the example shown in FIG. 11, the preset direction of the nth reference image Polygon_an includes a first direction of the outer boundary (for example, the boundary of the boundary point A, C, D) of the effective region of the nth reference image Polygon_an and a second direction of the inner boundary of the effective region of the nth reference image Polygon_an. The inner boundary is determined according to the boundary of the cavity (for example, the cavity JKLM) in the nth reference image Polygon_an. As shown in FIG. 11, in an example where the first direction is clockwise and the second direction is counterclockwise, according to the above scheme of constructing the directional skeleton line, the directional skeleton line 111 is constructed in the corresponding triangular face in the effective overlapping region shown in FIG. 11.

[0162] FIG. 12 is a schematic diagram of a process of constructing a directional skeleton line matched with the nth reference image according to another embodiment of the present application. Compared with the embodiment shown in FIG. 11, in FIG. 12, the cavity JKLM in the nth reference image Polygon_an covers the cavity EFGH in the nth starting image Polygon_bn. Correspondingly, the effective overlapping region between the nth starting image Polygon_bn and the nth reference image Polygon_an refers to the effective region in the overlapping region ABCD except the cavity JKLM.

[0163] In the example shown in FIG. 12, the preset direction of the nth reference image Polygon_an includes a first direction of the outer boundary (e.g., the boundary where the boundary points A, C, D are located) of the effective area of the nth reference image Polygon_an and a second direction of the inner boundary of the effective area of the nth reference image Polygon_an. The inner boundary is determined according to the boundary of the hole (e.g., the hole JKLM) in the nth reference image Polygon_an. As shown in FIG. 12, in the example where the first direction is counterclockwise and the second direction is clockwise, the directed skeleton line 121 is constructed in the corresponding triangular surface in the effective overlap area shown in FIG. 12 according to the above scheme of constructing the directed skeleton line.

[0164] FIG. 13 is a schematic diagram of a process of constructing a directed skeleton line matched with the nth reference image according to another embodiment of the present application. Compared with the embodiment shown in FIG. 10, the hole JKLM in the nth reference image Polygon_an and the hole EFGH in the nth start image Polygon_bn intersect with each other in FIG. 13, and the intersecting part of the hole is in the overlap area ABCD. Correspondingly, the effective overlap area between the nth start image Polygon_bn and the nth reference image Polygon_an refers to the effective area in the overlap area ABCD except the hole EFGH and the hole JKLM.

[0165] In the example shown in FIG. 13, the preset direction of the nth reference image Polygon_an includes a first direction of the outer boundary (e.g., the boundary where the boundary points A, C, D are located) of the effective area of the nth reference image Polygon_an and a second direction of the inner boundary of the effective area of the nth reference image Polygon_an. The inner boundary is determined according to the boundary of the hole (e.g., the hole JKLM) in the nth reference image Polygon_an.

[0166] As shown in FIG. 13, in the example where the first direction is counterclockwise and the second direction is clockwise, the directed skeleton line 131 is constructed in the corresponding triangular surface in the effective overlap area shown in FIG. 13 according to the above scheme of constructing the directed skeleton line. By comparing FIG. 10 and FIG. 13, it can be found that when the degree of intersection of the holes belonging to different images in the overlap area ABCD changes, the attribute information of the corresponding triangular surface in the two-dimensional triangular mesh will also be affected, thereby causing the position of the directed skeleton line to change.

[0167] FIG. 14 is a block diagram of a tiling line network generation device according to an embodiment of the present application.

[0168] As shown in FIG. 14, the tessellation line network generation apparatus 1400 comprises a determination module 1410, a tessellation module 1420 and a generation module 1430.

[0169] The determination module 1410 is configured to determine respective effective areas of a plurality of orthographic images to be tessellated.

[0170] The tessellation module 1420 is configured to perform a plurality of times of tessellation processing according to the respective effective areas of the plurality of orthographic images, to obtain a target effective polygon group.

[0171] The generation module 1430 is configured to generate a tessellation line network for the plurality of orthographic images according to the target effective polygon group.

[0172] In the embodiment of the present application, the tessellation module 1420 comprises an acquisition unit 1421, a construction unit 1422, a first determination unit 1423 and a second determination unit 1424.

[0173] The acquisition unit 1421 is configured to acquire an nth starting image and an nth reference image used for performing an nth time of tessellation processing, and an (n-1)th effective polygon group obtained by an (n-1)th time of tessellation processing, wherein the nth starting image is obtained by merging effective areas of all orthographic images participating in the first (n-1) times of tessellation processing, the nth reference image is an orthographic image determined from the plurality of orthographic images, which is not tessellated and has an intersection with the nth starting image, the (n-1)th effective polygon group comprises effective tessellation polygons of respective orthographic images participating in the first (n-1) times of tessellation processing, and n is an integer greater than 1.

[0174] The construction unit 1422 is configured to take a preset direction of the nth reference image as a reference, to construct a directed skeleton line matched with the nth reference image in an effective overlap area between the nth starting image and the nth reference image, to obtain a directed graph for the nth reference image, wherein the effective overlap area is determined according to the respective effective areas of the nth starting image and the nth reference image, the directed graph for the nth reference image indicates an effective tessellation polygon of the nth reference image after the nth time of tessellation processing, and the preset direction of the nth reference image is used to indicate a tracking direction of a boundary of the effective area of the nth reference image.

[0175] The first determination unit 1423 is configured to determine an nth effective polygon group corresponding to the nth time of tessellation processing based on the (n-1)th effective polygon group and the directed graph for the nth reference image.

[0176] The second determination unit 1424 is configured to determine that the nth effective polygon group is the target effective polygon group in response to determining that there is no orthographic image not tessellated in the plurality of orthographic images after the nth time of tessellation processing.

[0177] It should be noted that the implementation of the modules in the device part embodiments, the technical problems solved, the functions implemented, and the technical effects achieved are the same as or similar to the implementation of the corresponding steps in the method part embodiments, the technical problems solved, the functions implemented, and the technical effects achieved, and will not be described here.

[0178] FIG. 15 schematically shows a block diagram of an electronic device suitable for implementing the tessellation line network generation method according to an embodiment of the present application.

[0179] As shown in FIG. 15, the electronic device 1500 according to an embodiment of the present application includes a processor 1501 that can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM) 1502 or loaded from a storage section 1508 into a random access memory (RAM) 1503. The processor 1501 can include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor, and / or a related chipset, and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), and the like. The processor 1501 can also include an on-board memory for cache use. The processor 1501 can include a single processing unit or multiple processing units for performing different actions of the method processes according to embodiments of the present application.

[0180] In the RAM 1503, various programs and data required for the operation of the electronic device 1500 are stored. The processor 1501, the ROM 1502, and the RAM 1503 are connected to each other through a bus 1504. The processor 1501 performs various operations of the method processes according to embodiments of the present application by executing the programs in the ROM 1502 and / or the RAM 1503. It should be noted that the programs can also be stored in one or more memories other than the ROM 1502 and the RAM 1503. The processor 1501 can also perform various operations of the method processes according to embodiments of the present application by executing the programs stored in the one or more memories.

[0181] According to an embodiment of the present application, the electronic device 1500 can further include an input / output (I / O) interface 1505 that is also connected to the bus 1504. The electronic device 1500 can further include one or more of the following components connected to the I / O interface 1505: an input part 1506 including, for example, a keyboard and a mouse; an output part 1507 including, for example, a cathode ray tube (CRT), a liquid crystal display (LCD), and a speaker; a storage part 1508 including, for example, a hard disk; and a communication part 1509 including, for example, a LAN card, a modem, and the like. The communication part 1509 performs communication processing via a network such as the Internet. A drive 1510 is also connected to the I / O interface 1505 as necessary. A removable medium 1511 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, and the like is mounted on the drive 1510 as necessary, so that a computer program read therefrom is installed in the storage part 1508 as necessary.

[0182] The present application also provides a computer readable storage medium, which can be included in the device / apparatus / system described in the above embodiments, or can exist separately without being assembled into the device / apparatus / system. The computer readable storage medium carries one or more programs, which, when executed, implement the tessellation line network generation method according to the embodiments of the present application.

[0183] According to an embodiment of the present application, the computer readable storage medium can be a non-volatile computer readable storage medium, which can include, but is not limited to, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present application, the computer readable storage medium can be any tangible medium that contains or stores a program, which can be used by or in connection with an instruction execution system, apparatus, or device. For example, according to an embodiment of the present application, the computer readable storage medium can include one or more memories such as the ROM 1502 and / or the RAM 1503 described above and / or one or more memory components other than the ROM 1502 and the RAM 1503.

[0184] The embodiments of the present application also include a computer program product including a computer program containing program codes for executing the methods shown in the flowcharts. When the computer program product is run in a computer system, the program codes are used to make the computer system implement the tessellation line network generation method provided by the embodiments of the present application.

[0185] The computer program, when executed by the processor 1501, performs the above-described functions defined in the system / apparatus / module / processor according to the embodiments of the present application. According to embodiments of the present application, the system, apparatus, module, processor, etc. described above can be implemented by using computer program modules.

[0186] In one embodiment, the computer program can be stored in a tangible storage medium, such as an optical, magnetic, or semiconductor storage. In another embodiment, the computer program can be transmitted over a network using a transmission medium or a carrier wave. In some embodiments, the computer program can be downloaded or uploaded via the communication unit 1509. In some embodiments, the computer program can be embodied in a carrier wave that is downloaded over a network or other communications medium.

[0187] In such an embodiment, the computer program can be downloaded or uploaded to the system via the communication unit 1509 or the removable medium 1511. When the computer program is executed by the processor 1501, the above-described functions defined in the system according to the embodiments of the present application are performed. According to embodiments of the present application, the system, apparatus, module, processor, etc. described above can be implemented by using computer program modules.

[0188] According to embodiments of the present application, program code portions for executing the computer program of the embodiments of the present application can be written in any combination of one or more programming languages, and can be implemented by using high-level procedures and / or object-oriented programming languages, and / or assembly / machine languages. The programming languages include, but are not limited to, Java, C++, python, "C", or the like. The program code can execute entirely on the user's computing device, partly on the user's device, and partly on a remote computing device, or entirely on the remote computing device or server. In the latter scenario, the remote computing device can be connected to the user's computing device through any kind of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computing device, such as through the Internet using an Internet Service Provider (ISP).

[0189] The computer program product of the present application can be a computer program product comprising a computer readable storage medium and a computer program mechanism embedded in the computer readable storage medium. Such computer program product can further include a computer readable storage medium and program means for causing a processor or other programmable processing apparatus to function in a particular manner, such that the computer program mechanism embedded in the computer readable storage medium can be used to actually effect the apparatus functions.

[0190] Those skilled in the art will appreciate that the features recited in the various embodiments of the present application can be combined and / or integrated in a variety of ways, even if such combinations or integrations are not expressly contemplated in the present application. In particular, the features recited in the various embodiments of the present application can be combined and / or integrated in a variety of ways without departing from the spirit and scope of the present application. All such combinations and / or integrations are within the scope of the present application.

[0191] The above-described embodiments of the application are merely descriptive of its application and not limiting. Although the above-described embodiments have been described in some detail for purposes of clarity and the known examples, various changes, substitutions, and alterations can be made without departing from the spirit and scope of the application as defined by the appended claims.

Claims

1. A method for generating tessellated wire networks, characterized in that, include: Determine the effective regions of each of the multiple orthophotos to be mosaicked; Multiple mosaicking processes are performed on the effective regions of the multiple orthophotos to obtain a target effective polygon group; as well as Based on the target valid polygon set, generate a mosaic line network for the plurality of orthophotos; The inlay process includes: Obtain the nth starting image and the nth reference image for performing the nth mosaicking process, and the (n-1)th effective polygon set obtained from the (n-1)th mosaicking process. The nth starting image is obtained by merging the effective regions of all orthophotos participating in the previous (n-1)th mosaicking processes. The nth reference image is an orthophoto that has not been mosaicked and has an intersection with the nth starting image, determined from the plurality of orthophotos. The (n-1)th effective polygon set includes the effective mosaicking polygons of each orthophoto participating in the previous (n-1)th mosaicking processes, where n is an integer greater than 1. Using the direction of a preset nth reference image as a reference, a directed skeleton line matching the nth reference image is constructed within the effective overlap area between the nth starting image and the nth reference image to obtain a directed graph for the nth reference image. The effective overlap area is determined based on the effective areas of the nth starting image and the nth reference image, and the directed graph for the nth reference image indicates the effective mosaic polygon of the nth reference image after the nth mosaicking process. The direction of the preset nth reference image is used to indicate the tracking direction of the boundary of the effective area of ​​the nth reference image. Based on the (n-1)th effective polygon group and the directed graph for the nth reference image, determine the nth effective polygon group corresponding to the nth mosaicking process; In response to determining that there is no un-mosaved orthophoto among the plurality of orthophotos after the nth mosaicking process, the nth effective polygon group is determined as the target effective polygon group.

2. The method according to claim 1, characterized in that, The method further includes: In response to the determination that there are still un-mosaved orthophotos among the plurality of orthophotos after the nth mosaicking process, the effective regions of the nth starting image and the nth reference image are merged to obtain the n+1th starting image for performing the n+1th mosaicking process; Obtain the (n+1)th reference image for performing the (n+1)th mosaicking process, wherein the (n+1)th reference image is an orthophoto that has not been mosaicked and has an intersection with the (n+1)th starting image, determined from the plurality of orthophotos; Based on the (n+1)th starting image, the (n+1)th reference image, and the nth effective polygon group, update the nth starting image, the nth reference image, and the (n-1)th effective polygon group accordingly, and repeatedly perform the operation of determining whether the nth effective polygon group is the target effective polygon group.

3. The method according to claim 1, characterized in that, The step of constructing a directed skeleton line matching the nth reference image within the effective overlap area between the nth starting image and the nth reference image, using the direction of a preset nth reference image as a reference, to obtain a directed graph for the nth reference image includes: Based on the constrained Delaunay triangulation algorithm, a two-dimensional triangular mesh is constructed according to all vertices in the effective overlapping region. Each triangular face in the two-dimensional triangular mesh contains at most one intersection point, which is the intersection point between the boundary of the effective region of the nth starting image and the boundary of the effective region of the nth reference image. Using the direction of the preset nth reference image as a reference, a directed skeleton line matching the nth reference image is constructed within the effective overlapping area according to the two-dimensional triangular mesh, so as to obtain a directed graph for the nth reference image.

4. The method according to claim 3, characterized in that, The constrained Delaunay triangulation algorithm, which constructs a two-dimensional triangular mesh based on all vertices in the effective overlapping region, includes: Based on the Delaunay triangulation algorithm, an initial two-dimensional triangular mesh is constructed according to all vertices in the effective overlapping region; Determine each intersection point in the effective overlapping region; For any one of the intersection points, a constraint edge corresponding to the intersection point is constructed within the effective overlapping area. The constraint edge is formed by connecting a first constraint point and a second constraint point associated with the intersection point. The first constraint point and the second constraint point are any points on the two boundary segments of the effective overlapping area where the intersection point is located, excluding the intersection point itself. Except for the boundary segments where the first constraint point and the second constraint point are located, the constraint edge does not intersect with other boundary segments in the effective overlapping area. The initial two-dimensional triangular mesh is updated based on the constraint edges to obtain the updated two-dimensional triangular mesh; Repeat the operation of updating the initial two-dimensional triangular mesh based on the constraint edges until the constraint processing for all intersection points is completed, and the two-dimensional triangular mesh is obtained.

5. The method according to claim 3, characterized in that, The step of constructing a directed skeleton line matching the nth reference image within the effective overlapping area based on the two-dimensional triangular mesh, using the direction of a preset nth reference image as a reference, includes: Determine the attribute information of each triangular face in the two-dimensional triangular mesh, wherein the attribute information indicates the source of each vertex that constitutes the corresponding triangular face; Using the direction of the preset nth reference image as a reference, and based on the attribute information of each triangle, a directed skeleton line segment is constructed in the corresponding triangle in the effective overlapping area to obtain a directed skeleton line that matches the nth reference image.

6. The method according to claim 5, characterized in that, The step of using the direction of a preset nth reference image as a reference, and constructing directed skeleton line segments within the corresponding triangular faces in the effective overlapping area according to the attribute information of each triangular face, includes: For any of the triangles, in response to determining the attribute information of the triangle as first attribute information, the first attribute information indicates that the three vertices constituting the triangle include a first boundary point from the nth reference image and two second boundary points from the nth starting image. The midpoint of the line segment formed by the first boundary point and one second boundary point is determined as the first entry point, and the midpoint of the line segment formed by the first boundary point and the other second boundary point is determined as the first exit point. The first boundary point refers to any point on the boundary of the effective area of ​​the nth reference image other than the intersection point, and the second boundary point refers to any point on the boundary of the effective area of ​​the nth starting image other than the intersection point. Using the direction of the preset nth reference image as a reference, a first directed skeleton segment is constructed between the first in point and the first out point. The first directed skeleton segment is used to segment the two second boundary points from the nth starting image into the interior of the effective mosaic polygon of the nth reference image after the nth mosaic process.

7. The method according to claim 6, characterized in that, The step of using the direction of a preset nth reference image as a reference, and constructing directed skeleton line segments within the corresponding triangular faces in the effective overlapping area according to the attribute information of each triangular face, further includes: For any one of the triangles, in response to determining the attribute information of the triangle as second attribute information, the second attribute information indicates that the three vertices constituting the triangle include two first boundary points from the nth reference image and a second boundary point from the nth starting image, the midpoint of the line segment formed by one first boundary point and the second boundary point is determined as the second entry point, and the midpoint of the line segment formed by the other first boundary point and the second boundary point is determined as the second exit point. Using the direction of the preset nth reference image as a reference, a second directed skeleton segment is constructed between the second in point and the second out point. The second directed skeleton segment is used to segment a second boundary point from the nth starting image into the interior of the effective mosaic polygon of the nth reference image after the nth mosaic process.

8. The method according to claim 6, characterized in that, The step of using the direction of a preset nth reference image as a reference, and constructing directed skeleton line segments within the corresponding triangular faces in the effective overlapping area according to the attribute information of each triangular face, further includes: For any of the triangles, in response to determining the attribute information of the triangle as third attribute information, the third attribute information indicates that the three vertices constituting the triangle include a first boundary point from the nth reference image, a second boundary point from the nth starting image, and an intersection point, the midpoint of the line segment formed by the first boundary point and the second boundary point is determined as the third entry point or the third exit point. Using the direction of the preset nth reference image as a reference, a third directed skeleton segment is constructed between the third in-point and the intersection point or the third out-point and the intersection point. The third directed skeleton segment is used to segment a second boundary point from the nth starting image into the interior of the effective mosaic polygon of the nth reference image after the nth mosaic process.

9. The method according to claim 6, characterized in that, The step of using the direction of a preset nth reference image as a reference, and constructing directed skeleton line segments within the corresponding triangular faces in the effective overlapping area according to the attribute information of each triangular face, further includes: For any of the triangles, in response to determining that the attribute information of the triangle is the fourth attribute information or the fifth attribute information, no directed skeleton line segment is constructed within the triangle, wherein the fourth attribute information indicates that the three vertices constituting the triangle include three first boundary points from the nth reference image, and the fifth attribute information indicates that the three vertices constituting the triangle include three second boundary points from the nth starting image.

10. The method according to claim 5, characterized in that, The step of using the direction of a preset nth reference image as a reference to construct a directed skeleton line matching the nth reference image within the effective overlapping area based on the two-dimensional triangular mesh further includes: In response to determining the point of intersection of the directed skeleton line segment constructed within any of the triangular faces, the direction of the preset nth reference image is used as a reference. Starting from the intersection point, the line is traced along the boundary of the effective area of ​​the nth reference image until another intersection point is found.

11. The method according to any one of claims 1 to 10, characterized in that, The nth reference image and the nth starting image include at least one of the following: Both the nth reference image and the nth starting image are orthophotos that do not contain holes; One of the nth reference image and the nth starting image is an orthophoto image including at least one first hole; The nth reference image is an orthophoto image including at least one second hole, and the nth starting image is an orthophoto image including at least one third hole, wherein the at least one second hole and the at least one third hole do not intersect at all. The nth reference image is an orthophoto image including at least one fourth hole, and the nth starting image is an orthophoto image including at least one fifth hole, wherein the at least one fourth hole and the at least one fifth hole at least partially intersect. The nth reference image is an orthophoto image including at least one sixth hole, and the nth starting image is an orthophoto image including at least one seventh hole, wherein at least one of the at least one sixth hole includes a sixth hole that covers the seventh hole. The nth reference image is an orthophoto image including at least one eighth hole, and the nth starting image is an orthophoto image including at least one ninth hole, wherein at least one ninth hole includes a ninth hole that covers the eighth hole.

12. The method according to claim 11, characterized in that, In response to the nth reference image being an orthophoto image excluding holes, the preset direction of the nth reference image includes a first direction relative to the outer boundary of the effective area of ​​the nth reference image; In response to the nth reference image being an orthophoto image including at least one hole, the preset direction of the nth reference image includes a first direction relative to the outer boundary of the effective area of ​​the nth reference image and a second direction relative to the inner boundary of the effective area of ​​the nth reference image, wherein the inner boundary is determined based on the boundary of at least one hole in the nth reference image, and the first direction and the second direction are opposite.

13. The method according to any one of claims 1 to 10, characterized in that, The step of determining the nth effective polygon group corresponding to the nth mosaicking process based on the (n-1)th effective polygon group and the directed graph for the nth reference image includes: Based on the directed graph for the nth reference image, determine the effective mosaic polygon of the nth reference image after the nth mosaicking process; For each effective mosaic polygon of the orthophoto in the (n-1)th effective polygon group, the target overlapping region is removed from the effective mosaic polygon of the orthophoto to obtain an updated effective polygon group. The target overlapping region refers to the effective overlapping region between the effective mosaic polygon of the orthophoto and the effective mosaic polygon of the nth reference image after the nth mosaic processing. The effective mosaic polygons of the nth reference image after the nth mosaicking process are added to the updated effective polygon group to obtain the nth effective polygon group.

14. The method according to claim 13, characterized in that, The step of generating a mosaic line network for the plurality of orthophotos based on the target effective polygon group includes: Determine the common edges between adjacent valid tessellated polygons in the target valid polygon group, and generate the tessellation network based on the common edges between all adjacent valid tessellated polygons.

15. The method according to any one of claims 1 to 10, characterized in that, The method further includes: Based on the mosaicking network, the multiple orthophotos are mosaicked to obtain a mosaicked image.

16. A mosaic line network generation apparatus, characterized in that, include: The determination module is used to determine the effective regions of each of the multiple orthophotos to be mosaicked; The mosaicking module is used to perform multiple mosaicking processes based on the effective regions of the multiple orthophotos to obtain a group of effective polygons of the target. as well as A generation module is used to generate a mosaic line network for the plurality of orthophotos based on the target effective polygon group; The inlay module includes: The acquisition unit is configured to acquire the nth starting image and the nth reference image for performing the nth mosaicking process, and the (n-1)th effective polygon group obtained from the (n-1)th mosaicking process. The nth starting image is obtained by merging the effective regions of all orthophotos participating in the previous (n-1)th mosaicking processes. The nth reference image is an orthophoto that has not been mosaicked and has an intersection with the nth starting image, determined from the plurality of orthophotos. The (n-1)th effective polygon group includes the effective mosaicking polygons of each orthophoto participating in the previous (n-1)th mosaicking processes, where n is an integer greater than 1. A construction unit is used to construct a directed skeleton line matching the nth reference image within the effective overlap area between the nth starting image and the nth reference image, with the direction of a preset nth reference image as a reference, to obtain a directed graph for the nth reference image. The effective overlap area is determined based on the effective areas of the nth starting image and the nth reference image, and the directed graph for the nth reference image indicates the effective mosaic polygon of the nth reference image after the nth mosaicking process. The direction of the preset nth reference image is used to indicate the tracking direction of the boundary of the effective area of ​​the nth reference image. The first determining unit is used to determine the nth effective polygon group corresponding to the nth mosaicking process based on the (n-1)th effective polygon group and the directed graph for the nth reference image. The second determining unit is configured to determine the nth effective polygon group as the target effective polygon group in response to determining that there is no un-mosaved orthophoto among the plurality of orthophotos after the nth mosaicking process.

17. An electronic device, characterized in that, include: One or more processors; Memory, used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors perform the method according to any one of claims 1 to 15.

18. A computer-readable storage medium, characterized in that, It stores executable instructions that, when executed by a processor, cause the processor to perform the method according to any one of claims 1 to 15.

19. A computer program product, characterized in that, It includes a computer program that, when executed by a processor, implements the method according to any one of claims 1 to 15.

Images