Aerial image stitching method, device, terminal equipment and storage medium

Abstract

The present invention discloses an aerial image stitching method, device, terminal equipment and storage medium. Obtain highway map information of the area to be stitched from an existing map, and generate standardized base maps of several levels according to the highway map information; sequentially obtain road block images in each of the aerial images of the corresponding level according to the standardized base map in combination with the aerial images; sequentially fill the road block images into the standardized base map of the corresponding level to obtain stitching images of different levels; and sequentially superimpose the stitching image of the current level onto the stitching image of the next level according to a predetermined level order to obtain a complete stitching image. The present application can not only avoid deformation of the stitching image, but also improve the integrity of the road structure in the stitching image and the clarity of the image.

Description

Technical Field

[0001] The present invention relates to the field of image processing, and in particular to an aerial image stitching method, device, terminal equipment and storage medium. Background Art

[0002] The existing aerial video image stitching method uses a feature matching method based on feature descriptors, including feature point matching, removal of mismatched feature point pairs, transformation matrix model parameter calculation, interpolation processing, and image stitching and fusion processing. At the same time, through the relationship between video image frames, the video image of the next frame is mapped to the previous frame, and superimposed in sequence until the last frame of the video image, and finally a stitching image is formed.

[0003] When processing urban scenes, the mosaic image generated based on the above method will show deformation of high-rise building targets in the mosaic image and unclear road structure in the image. This is mainly caused by the following reasons: 1. When selecting matching feature points for image mosaicking, the ideal situation is that the feature points are on the same horizontal plane. However, when processing urban scenes, the feature points of high-rise buildings can be easily selected incorrectly, resulting in feature points not being on the same horizontal plane, causing deformation of high-rise buildings; 2. Due to the presence of noise such as occlusions, the error rate of feature matching will be increased, resulting in discontinuity or inconsistency of the road in the mosaicked image, making the road structure in the mosaicked image inaccurate or unclear. Summary of the invention

[0004] The present invention provides an aerial image stitching method, apparatus, terminal device and storage medium to solve the technical problems of image deformation, unclear stitching and incomplete stitching caused by feature matching errors in the image stitching process, thereby improving the clarity and accuracy of the stitching image and providing complete road structure information.

[0005] In order to solve the above technical problems, in a first aspect, an embodiment of the present invention provides an aerial image stitching method, comprising:

[0006] Obtaining highway map information of the area to be spliced ​​from the existing map, and generating standardized base maps of several levels according to the highway map information; the standardized base maps include: the coordinates of the center point of each map road block of the corresponding level, several road nodes of each map road block of the corresponding level, and connectivity information between the road nodes;

[0007] According to the standardized base map and in combination with the aerial images, a road block image in each of the aerial images of the corresponding level is obtained;

[0008] Sequentially filling the road block images into the standardized base map of the corresponding levels to obtain mosaic images of different levels;

[0009] According to the predetermined hierarchical order, the mosaic image of the current level is superimposed on the mosaic image of the next level in turn to obtain a complete mosaic image.

[0010] Compared with the prior art, the above embodiment has the following beneficial effects: by acquiring highway map information from an existing map to generate a standardized map, accurate road structure and road connection information can be obtained when image stitching is subsequently performed according to the standardized map; by using a standardized base map, block images that need to be filled into the base map in the aerial image are extracted, and then the block images are filled into a standardized base map similar to a frame. This method can effectively avoid the problem of splicing image deformation caused by inaccurate feature matching based on a feature matching method; in addition, the splicing image is divided into different levels, and finally multiple levels are superimposed together, which can ensure the accuracy of the spliced ​​image under different scale information, and improve the resolution and accuracy of the final image.

[0011] In an embodiment of the first aspect, the step of sequentially obtaining a road block image in each of the aerial images of a corresponding level based on the standardized base map and in combination with the aerial images includes:

[0012] Acquire an aerial image taken vertically by an aerial photography device at each of the center point coordinates;

[0013] The road block image in each of the aerial images is obtained in sequence according to the road nodes of the map road block corresponding to the center point coordinates and the connectivity information between the road nodes.

[0014] Compared with the prior art, the above embodiment has the following beneficial effects: through the center point coordinate information, the image of each area in the base map can be accurately found from the existing captured images, and then the block images surrounded by roads of different levels can be accurately divided according to the obtained block node information and the connectivity information between the nodes, thereby improving the accuracy and clarity of the spliced ​​image while retaining the road structure.

[0015] In an embodiment of the first aspect, the step of acquiring the road block image in each of the aerial images according to the road nodes of the map road block corresponding to the center point coordinates and the connectivity information between the road nodes includes:

[0016] Calculating the actual distance from the center point coordinate to each road node according to the center point coordinate and each road node of the corresponding map road block in sequence;

[0017] Acquire the shooting height when shooting the image at the center point coordinates, and sequentially obtain the mapping node of each road node in the aerial image by combining the shooting height and the actual distance;

[0018] According to the connectivity information between all the mapping nodes and the road nodes, the road block image in the aerial image is divided.

[0019] Compared with the prior art, the above embodiment has the following beneficial effects: by calculating the actual distance from the center point coordinates to each road node, combined with information such as the shooting height, the mapping node of each road node in the aerial image can be obtained, thereby achieving accurate mapping of the road nodes, and improving the continuity of the road structure in the stitched image and the consistency with the actual road structure; combined with the high resolution of the aerial image, more detailed and clear road block images can be obtained, providing high-quality support data for subsequent image stitching, and improving the accuracy of subsequent image stitching.

[0020] In an embodiment of the first aspect, sequentially filling the road block images into standardized base maps of corresponding levels to obtain mosaic images of different levels includes:

[0021] Combining the scale information of the existing map and the shooting height when the image is shot at the center point coordinates, and applying a scale transformation algorithm, the road block image is converted into a road block image of a standard size;

[0022] The road block images of the standard size are sequentially filled into the road blocks corresponding to the center point coordinates of the standardized base map of the corresponding level until the road blocks in the standardized base maps of different levels are filled.

[0023] Compared with the prior art, the above embodiment has the following beneficial effects: a scale transformation algorithm is used to convert the road block image into a road block image of a standard size, ensuring that the road block image has a consistent scale in standardized base maps at different levels, thereby improving the coherence of image stitching; by adopting a scale transformation algorithm and precise filling, the geometric shape of each area in the image can be effectively maintained, ensuring that the road block image can maintain its original shape after being filled into the base map, so that the final stitching image is not distorted or stretched, thereby improving the quality of the stitching image.

[0024] In an embodiment of the first aspect, obtaining road map information of the area to be spliced ​​from an existing map, and generating standardized base maps of several levels according to the road map information, includes:

[0025] Extract detailed road maps of each level in the area to be spliced ​​from the existing map;

[0026] Divide the area to be spliced ​​according to the road refinement map of each level to obtain a map road block division map of the corresponding level;

[0027] According to the road block division map of the corresponding level, combined with the scale information of the existing map and the preset shooting height of the aerial image, the standardized base map of the corresponding level is obtained through the scale change algorithm.

[0028] Compared with the prior art, the above embodiment has the following beneficial effects: by extracting road refinement maps of different levels of the area to be spliced ​​in the existing map, accurate road structure information is retained and the accuracy of block division is improved; and through the scale change algorithm, the scale of the road block can be adaptively adjusted, so that the standardized base map is more in line with the actual scene, and image distortion caused by scale change during image splicing is reduced.

[0029] In an embodiment of the first aspect, dividing the to-be-joined area according to the road refinement map at each level to obtain a map road block division map at a corresponding level includes:

[0030] Extracting the road node network of each level from the road refinement graph of each level, and obtaining the connectivity information between the road nodes; the road node network is composed of a plurality of road nodes;

[0031] According to the road node network, the area to be spliced ​​is divided into a number of closed map road blocks of corresponding levels; there are no empty areas between the map road blocks of the same level.

[0032] Compared with the prior art, the above embodiment has the following beneficial effects: by extracting road refinement maps of different levels of the area to be spliced ​​in the existing map, the road node information can be accurately identified, and the accurate road structure information can be retained. The road block segmentation is performed in combination with the connectivity information between the nodes, thereby improving the clarity and accuracy of the road structure in the spliced ​​image during subsequent image splicing.

[0033] In an embodiment of the first aspect, pixel points in the road refinement graph that have two or more different adjacent pixel similarity points are set as the road nodes.

[0034] Compared with the prior art, the above embodiment has the following beneficial effects: through the similarity of pixel points, the difference between road pixel points and pixel points of other target objects can be accurately identified, thereby improving the accuracy of subsequent road block division.

[0035] In an embodiment of the first aspect, the method of obtaining a standardized base map of a corresponding level by a scale change algorithm based on a map road block division diagram of a corresponding level, combining scale information of an existing map and a preset shooting height of an aerial image, includes:

[0036] Calculate scale transformation information based on the scale information of the existing map and the preset shooting height of the aerial image;

[0037] According to the scale transformation information, the map road block division diagram is converted into the standardized base map of the corresponding level through a scale change algorithm.

[0038] Compared with the prior art, the above embodiment has the following beneficial effects: the scale information of the base map that needs to be transformed is accurately calculated according to the scale information of the map and the shooting height of the aerial image, and the standard degree of the base map finally obtained is improved by combining the scale change algorithm.

[0039] In a second aspect, an embodiment of the present invention provides an aerial image stitching device, the aerial image stitching device comprising: a standard base map acquisition module, a road block image acquisition module, an image filling module and an image overlay module;

[0040] The standard base map acquisition module is used to obtain the road map information of the area to be spliced ​​from the existing map, and generate standardized base maps of several levels according to the road map information; the standardized base map includes: the center point coordinates of each map road block of the corresponding level, several road nodes of each map road block of the corresponding level, and the connectivity information between the road nodes;

[0041] The road block image acquisition module is used to sequentially acquire the road block image in each of the aerial images of the corresponding level based on the standardized base map and in combination with the aerial images;

[0042] The image filling module is used to sequentially fill the road block images into the standardized base map of the corresponding level to obtain mosaic images of different levels;

[0043] The image superposition module is used to superimpose the mosaic image of the current level onto the mosaic image of the next level in sequence according to a predetermined level order to obtain a complete mosaic image.

[0044] In an embodiment of the second aspect, the road block image acquisition module includes: an aerial image acquisition unit and a road block image acquisition unit;

[0045] Wherein, the aerial image acquisition unit is used to acquire the aerial image taken vertically by the aerial photography device at each of the center point coordinates;

[0046] The road block image acquisition unit is used to sequentially acquire the road block image in each of the aerial images according to the road nodes of the map road block corresponding to the center point coordinates and the connectivity information between the road nodes.

[0047] In an embodiment of the second aspect, the road block image acquisition unit includes: a distance calculation subunit, a node mapping subunit and an image division subunit;

[0048] The distance calculation subunit is used to calculate the actual distance from the center point coordinate to each road node according to the center point coordinate and each road node of the corresponding map road block in sequence;

[0049] The node mapping subunit is used to obtain the shooting height when the image is shot at the center point coordinates, and to sequentially obtain the mapping node of each road node in the aerial image by combining the shooting height and the actual distance;

[0050] The image division subunit is used to divide the road block image in the aerial image according to the connectivity information between all the mapping nodes and the road nodes.

[0051] In an embodiment of the second aspect, the image filling module includes: an image standardization unit and an image filling unit;

[0052] The image standardization unit is used to sequentially combine the scale information of the existing map and the shooting height when the image is shot at the center point coordinates, and use a scale transformation algorithm to convert the road block image into a road block image of a standard size;

[0053] The image filling unit is used to sequentially fill the road block images of the standard size into the road blocks corresponding to the center point coordinates of the standardized base map of the corresponding level until the road blocks in the standardized base maps of different levels are filled.

[0054] In an embodiment of the second aspect, the standard base map acquisition module includes: a road refinement map extraction unit, a map road block division unit, and a standardized base map generation unit;

[0055] The road detailed map extraction unit is used to extract the road detailed maps of each level in the area to be spliced ​​from the existing map;

[0056] The map road block division unit is used to divide the to-be-joined area according to the road refinement map of each level to obtain the map road block division map of the corresponding level;

[0057] The standardized base map generation unit is used to obtain the standardized base map of the corresponding level through a scale change algorithm according to the map road block division map of the corresponding level, combined with the scale information of the existing map and the preset shooting height of the aerial image.

[0058] In an embodiment of the second aspect, the map road block division unit includes: a map road node extraction subunit and a map road block division subunit;

[0059] The map road node extraction subunit is used to extract the road node network of each level from the road refinement map of each level, and obtain the connectivity information between the road nodes; the road node network is composed of a number of road nodes;

[0060] The map road block division subunit is used to divide the area to be spliced ​​into a number of closed map road blocks of corresponding levels according to the road node network; there are no empty areas between the map road blocks of the same level.

[0061] In an embodiment of the second aspect, the map road node extraction subunit is further used to set the pixel points in the road refinement map that have more than two different adjacent pixel similarity points as the road nodes.

[0062] In an embodiment of the second aspect, the standardized base map generation unit includes: a transformation scale calculation subunit and an image scale transformation subunit;

[0063] The transformation scale calculation subunit is used to calculate the scale transformation information according to the scale information of the existing map and the preset shooting height of the aerial image;

[0064] The image scale conversion subunit is used to convert the map road block division diagram into the standardized base map of the corresponding level through a scale change algorithm according to the scale conversion information.

[0065] In a third aspect, the present invention further provides a terminal device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the above-mentioned aerial image stitching method when executing the computer program.

[0066] In a fourth aspect, the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the above-mentioned aerial image stitching method. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 A schematic diagram of a flow chart of an embodiment of an aerial image stitching method provided by the present invention;

[0068] Figure 2 A schematic structural diagram of an embodiment of an aerial image stitching device provided by the present invention;

[0069] Figure 3 A schematic structural diagram of an embodiment of a standard base map acquisition module in an aerial image stitching device provided by the present invention;

[0070] Figure 4 A schematic structural diagram of an embodiment of a road block image acquisition module in an aerial image stitching device provided by the present invention;

[0071] Figure 5 A schematic structural diagram of an embodiment of an image filling module in an aerial image stitching device provided by the present invention. DETAILED DESCRIPTION

[0072] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0073] Embodiment 1

[0074] Please refer to Figure 1 , an aerial image stitching method provided by an embodiment of the present invention, includes S101 to S104, specifically:

[0075] S101: Obtaining highway map information of the area to be spliced ​​from an existing map, and generating standardized base maps of several levels according to the highway map information; the standardized base maps include: the coordinates of the center point of each map road block of the corresponding level, several road nodes of each map road block of the corresponding level, and connectivity information between the road nodes;

[0076] In one embodiment, the road map information of the area to be spliced ​​is obtained from the existing map, and several levels of standardized base maps are generated according to the road map information, including: extracting the road detailed map of each level of the area to be spliced ​​from the existing map; dividing the area to be spliced ​​according to the road detailed map of each level to obtain the map road block division map of the corresponding level; according to the map road block division map of the corresponding level, combined with the scale information of the existing map and the preset shooting height of the aerial image, the standardized base map of the corresponding level is obtained through a scale change algorithm.

[0077] Furthermore, the area to be spliced ​​is divided according to the road refinement map of each level to obtain a map road block division map of the corresponding level, including: extracting the road node network of each level from the road refinement map of each level to obtain connectivity information between road nodes; the road node network is composed of a number of road nodes; according to the road node network, the area to be spliced ​​is divided into a number of closed map road blocks of the corresponding level; there is no empty area between the map road blocks of the same level.

[0078] Preferably, the road nodes may be obtained by setting the pixel points having two or more different adjacent pixel similarity points in the road refinement map as the road nodes.

[0079] Furthermore, the standardized base map of the corresponding level is obtained through a scale change algorithm based on the map road block division diagram of the corresponding level, in combination with the scale information of the existing map and the preset shooting height of the aerial image, including: calculating scale transformation information based on the scale information of the existing map and the preset shooting height of the aerial image; and converting the map road block division diagram into the standardized base map of the corresponding level through a scale change algorithm based on the scale transformation information.

[0080] In order to better illustrate the above embodiment, the following examples are used for supplementary explanation:

[0081] A standardized basemap can be generated by following the steps below:

[0082] S1011: Determine the area to be spliced ​​according to the requirements, and find the highway map information of the corresponding area in the existing map, including national roads (first-level roads), provincial roads (second-level roads), urban roads (third-level roads) and internal roads;

[0083] S1012: using a thinning line algorithm, converting the primary roads, secondary roads, tertiary roads, and internal roads into a primary road thinning line map, a secondary road thinning line map, a tertiary road thinning line map, and an internal road thinning line map respectively;

[0084] S1013: extracting the road nodes in the road thinning line graph obtained in S1012 respectively to form a primary road node network, a secondary road node network, a tertiary road node network, and an internal road node network; wherein the road node is defined as having two or more non-adjacent pixel similarity points in the thinning line graph, that is, the pixel point is called a road node;

[0085] S1014: The closed loop areas surrounded by the road node networks at all levels are divided respectively, and the closed loop areas are convex polygons without any hollow areas, so as to form primary road blocks, secondary road blocks, tertiary road blocks, and internal road blocks;

[0086] S1015: Based on the above multi-level blocks, combined with the highway map scale information and the designated aerial image shooting height, a scale transformation algorithm is applied to form standardized base maps at various levels.

[0087] Compared with the prior art, the above embodiment has the following beneficial effects: by extracting road refinement maps of different levels of the area to be spliced ​​in the existing map, accurate road structure information is retained and the accuracy of block division is improved; and through the scale change algorithm, the scale of the road block can be adaptively adjusted, so that the standardized base map is more in line with the actual scene, and image distortion caused by scale change during image splicing is reduced.

[0088] S102: sequentially obtaining a road block image in each of the aerial images of a corresponding level according to the standardized base map and in combination with the aerial images;

[0089] Furthermore, the road block image can be obtained by the following method: obtaining an aerial image taken vertically by an aerial photography device at each of the center point coordinates; and obtaining the road block image in each of the aerial images according to the road nodes of the map road block corresponding to the center point coordinates and the connectivity information between the road nodes.

[0090] Furthermore, the road block image in each of the aerial images is obtained in sequence according to the road nodes of the map road block corresponding to the center point coordinates and the connectivity information between the road nodes, including: calculating the actual distance from the center point coordinates to each of the road nodes in sequence according to the center point coordinates and each of the corresponding road nodes in the map road block; obtaining the shooting height when the image is taken at the center point coordinates, and combining the shooting height and the actual distance to obtain the mapping node of each of the road nodes in the aerial image in sequence; dividing the road block image in the aerial image according to all the mapping nodes and the connectivity information between the road nodes.

[0091] Preferably, the aerial image can be obtained by, but not limited to, the following method: controlling the drone to the center point coordinates and performing vertical shooting at a preset shooting height.

[0092] Preferably, the mapping nodes in the aerial image can be obtained by the following steps:

[0093] S1021: extracting the center point GPS information of the road block and the road node GPS information of the corresponding road block, and respectively calculating the actual distance between the center point GPS position and the GPS position of each road node of the corresponding road block;

[0094] S1022: The pixel center point of the image is matched with the GPS position of the center point of the block, and the actual distance of S1021 is converted into pixel distance through the flight altitude of the drone and the internal parameters of the camera;

[0095] S1023: Find the mapping node of the corresponding block road node in the aerial image through the pixel center point position combined with the pixel distance.

[0096] Compared with the prior art, the above embodiment has the following beneficial effects: by calculating the actual distance from the center point coordinates to each road node, combined with information such as the shooting height, the mapping node of each road node in the aerial image can be obtained, thereby achieving accurate mapping of the road nodes, and improving the continuity of the road structure in the stitched image and the consistency with the actual road structure; combined with the high resolution of the aerial image, more detailed and clear road block images can be obtained, providing high-quality support data for subsequent image stitching, and improving the accuracy of subsequent image stitching.

[0097] S103: sequentially filling the road block images into the standardized base map of the corresponding levels to obtain mosaic images of different levels;

[0098] Furthermore, the step of sequentially filling the road block images into the standardized base map of the corresponding level to obtain mosaic images of different levels includes: sequentially combining the scale information of the existing map and the shooting height when the image is shot at the center point coordinates, and using a scale transformation algorithm to convert the road block images into road block images of standard size; sequentially filling the road block images of standard size into the road blocks corresponding to the center point coordinates of the standardized base map of the corresponding level, until the road blocks in the standardized base maps of different levels are filled.

[0099] S104: According to a predetermined hierarchical order, the mosaic image of the current level is superimposed on the mosaic image of the next level in sequence to obtain a complete mosaic image.

[0100] Preferably, steps S103 to S104 can be implemented by the following steps:

[0101] Step 1: splice the first-level road block images contained in the aerial image into the corresponding first-level road standardized base map through a scale transformation algorithm according to the aerial photography height and the highway map scale information;

[0102] Step 2: splice the secondary road block images contained in the aerial image into the corresponding secondary road standardized base map according to the aerial photography height and the highway map scale information through a scale transformation algorithm, and overlay the primary road standardized base map onto the secondary road standardized base map;

[0103] Step 3: splice the third-level road block images contained in the aerial image into the corresponding third-level road standardized base map through a scale transformation algorithm according to the aerial photography height and the highway map scale information, and superimpose the second-level road standardized base map after the superposition step in step 2 onto the third-level road standardized base map;

[0104] Step 4: The internal road blocks contained in the aerial image are spliced ​​into the corresponding internal road standardized base map through a scale transformation algorithm according to the aerial photography height and the highway map scale information, and the third-level road standardized base map after the overlay step in step 3 is overlaid on the internal road standardized base map to finally form a spliced ​​image.

[0105] In summary, it can be seen that compared with the prior art, the aerial image stitching method provided by the embodiment of the present invention has the following beneficial effects: a scale transformation algorithm is used to convert the road block image into a road block image of a standard size, ensuring that the road block image has a consistent scale in standardized base maps at different levels, thereby improving the coherence of image stitching; by adopting a scale transformation algorithm and precise filling, the geometric shape of each area in the image can be effectively maintained, ensuring that the road block image can maintain its original shape in the base map filled in, and the final stitching image is not distorted or stretched, thereby improving the quality of the stitching image; the stitching image is divided into different levels, and finally multiple levels are superimposed together, thereby further improving the clarity and accuracy of the stitching image while retaining the image detail information.

[0106] Embodiment 2

[0107] Please refer to Figure 2 , is an aerial image stitching device provided by an embodiment of the present invention, and the aerial image stitching device includes: a standard base map acquisition module 21, a road block image acquisition module 22, an image filling module 23 and an image overlay module 24.

[0108] In one embodiment, the standard base map acquisition module 21 is used to obtain highway map information of the area to be spliced ​​from an existing map, and generate standardized base maps of several levels according to the highway map information; the standardized base map includes: the center point coordinates of each map road block of the corresponding level, a number of road nodes of each map road block of the corresponding level, and connectivity information between the road nodes; the road block image acquisition module 22 is used to sequentially obtain the road block image in each aerial image of the corresponding level according to the standardized base map and in combination with the aerial image; the image filling module 23 is used to sequentially fill the road block image into the standardized base map of the corresponding level to obtain spliced ​​images of different levels; the image overlay module 24 is used to sequentially overlay the spliced ​​image of the current level onto the spliced ​​image of the next level according to a predetermined level order to obtain a complete spliced ​​image.

[0109] Reference Figure 3, which is a standard base map acquisition module in an aerial image stitching device provided by an embodiment of the present invention, the standard base map acquisition module includes: a road refinement map extraction unit 211, a map road block division unit 212 and a standardized base map generation unit 213; the map road block division unit 212 includes: a map road node extraction subunit 2121 and a map road block division subunit 2122; the standardized base map generation unit 213 includes: a transformation scale calculation subunit 2131 and an image scale transformation subunit 2132.

[0110] In one embodiment, the road refinement map extraction unit 211 is used to extract the road refinement maps of each level of the area to be spliced ​​from the existing map; the map road block division unit 212 is used to divide the area to be spliced ​​according to the road refinement maps of each level to obtain the map road block division map of the corresponding level; the standardized base map generation unit 213 is used to obtain the standardized base map of the corresponding level through a scale change algorithm based on the map road block division map of the corresponding level, combined with the scale information of the existing map and the preset shooting height of the aerial image.

[0111] In one embodiment, the map road node extraction subunit 2121 is used to extract the road node network of each level from the road refinement map of each level to obtain the connectivity information between the road nodes; the road node network is composed of a plurality of road nodes; the map road block division subunit 2122 is used to divide the area to be spliced ​​into a plurality of closed map road blocks of corresponding levels according to the road node network; there are no empty areas between the map road blocks of the same level.

[0112] In one embodiment, the map road node extraction subunit 2121 is further used to set the pixel points in the road refinement map that have more than two different adjacent pixel similarity points as the road nodes.

[0113] In one embodiment, the transformation scale calculation subunit 2131 is used to calculate the scale transformation information based on the scale information of the existing map and the preset shooting height of the aerial image; the image scale transformation subunit 2132 is used to convert the map road block division map into the standardized base map of the corresponding level through a scale change algorithm based on the scale transformation information.

[0114] Reference Figure 4, which is a road block image acquisition module in an aerial image stitching device provided by an embodiment of the present invention, the road block image acquisition module includes: an aerial image acquisition unit 221 and a road block image acquisition unit 222; the road block image acquisition unit 222 includes: a distance calculation subunit 2221, a node mapping subunit 2222 and an image division subunit 2223.

[0115] In one embodiment, the aerial image acquisition unit 221 is used to acquire aerial images taken vertically by the aerial photography device at each of the center point coordinates; the road block image acquisition unit 222 is used to acquire the road block image in each of the aerial images in turn according to the road nodes of the map road block corresponding to the center point coordinates and the connectivity information between the road nodes.

[0116] In one embodiment, the distance calculation subunit 2221 is used to calculate the actual distance from the center point coordinates to each road node according to the center point coordinates and each road node of the corresponding map road block in turn; the node mapping subunit 2222 is used to obtain the shooting height when the image is taken at the center point coordinates, and combine the shooting height and the actual distance to obtain the mapping node of each road node in the aerial image in turn; the image division subunit 2223 is used to divide the road block image in the aerial image according to the connectivity information between all the mapping nodes and the road nodes.

[0117] Reference Figure 5 , is an image filling module in an aerial image stitching device provided by an embodiment of the present invention, and the image filling module includes: an image standardization unit 231 and an image filling unit 232.

[0118] In one embodiment, the image standardization unit 231 is used to sequentially combine the scale information of the existing map and the shooting height when the image is shot at the center point coordinates, and use a scale transformation algorithm to convert the road block image into a road block image of standard size; the image filling unit 232 is used to sequentially fill the road block image of standard size into the road block corresponding to the center point coordinates of the corresponding level standardized base map until the road blocks in the standardized base maps of different levels are filled.

[0119] Embodiment 3

[0120] Based on the above-mentioned embodiment of the aerial image stitching method, another embodiment of the present invention provides an aerial image stitching terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the aerial image stitching method of any embodiment of the present invention is implemented.

[0121] Exemplarily, in this embodiment, the computer program may be divided into one or more modules, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the aerial image stitching device.

[0122] The aerial image stitching device may be a computing device such as a desktop computer, a notebook, a palm computer, a cloud server, etc. The aerial image stitching terminal device may include, but is not limited to, a processor and a memory.

[0123] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the aerial image stitching device, and uses various interfaces and lines to connect the various parts of the entire aerial image stitching device. The memory may be used to store the computer program and / or module, and the processor implements various functions of the aerial image stitching device by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, at least one application required for a function, etc.; the data storage area may store data created according to the use of the mobile phone, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Med ia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0124] Embodiment 4

[0125] Based on the above-mentioned embodiment of the aerial image stitching method, another embodiment of the present invention provides a storage medium, wherein the storage medium includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute the aerial image stitching method of any embodiment of the present invention.

[0126] In this embodiment, the storage medium is a computer-readable storage medium, and the computer program includes computer program code, which may be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0127] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for stitching aerial images, characterized in that: include: Obtaining highway map information of the area to be spliced ​​from the existing map, and generating standardized base maps of several levels according to the highway map information; the standardized base maps include: the coordinates of the center point of each map road block of the corresponding level, several road nodes of each map road block of the corresponding level, and connectivity information between the road nodes; According to the standardized base map and in combination with the aerial images, a road block image in each of the aerial images of the corresponding level is obtained; Sequentially filling the road block images into the standardized base map of the corresponding levels to obtain mosaic images of different levels; According to the predetermined hierarchical order, the mosaic image of the current level is superimposed on the mosaic image of the next level in turn to obtain a complete mosaic image; The method of obtaining road map information of the area to be spliced ​​from the existing map and generating standardized base maps of several levels according to the road map information includes: Extracting road refinement maps of each level in the area to be spliced ​​from the existing map, and setting pixel points in the road refinement map that have two or more different adjacent pixel similarity points as the road nodes; Divide the area to be spliced ​​according to the road refinement map of each level to obtain a map road block division map of the corresponding level; According to the road block division map of the corresponding level, combined with the scale information of the existing map and the preset shooting height of the aerial image, the standardized base map of the corresponding level is obtained through the scale change algorithm.

2. The aerial image stitching method according to claim 1, characterized in that: The step of sequentially obtaining a road block image in each of the aerial images of a corresponding level based on the standardized base map and in combination with the aerial images includes: Acquire an aerial image taken vertically by an aerial photography device at each of the center point coordinates; The road block image in each of the aerial images is obtained in sequence according to the road nodes of the map road block corresponding to the center point coordinates and the connectivity information between the road nodes.

3. The aerial image stitching method according to claim 2, characterized in that: The step of sequentially obtaining a road block image in each of the aerial images according to the road nodes of the map road block corresponding to the center point coordinates and the connectivity information between the road nodes includes: Calculating the actual distance from the center point coordinate to each road node according to the center point coordinate and each road node of the corresponding map road block in sequence; Acquire the shooting height when shooting the image at the center point coordinates, and sequentially obtain the mapping node of each road node in the aerial image by combining the shooting height and the actual distance; According to the connectivity information between all the mapping nodes and the road nodes, the road block image in the aerial image is divided.

4. The aerial image stitching method according to claim 1, characterized in that: The step of sequentially filling the road block images into the standardized base map of the corresponding levels to obtain mosaic images of different levels includes: Combining the scale information of the existing map and the shooting height when the image is shot at the center point coordinates, and applying a scale transformation algorithm, the road block image is converted into a road block image of a standard size; The road block images of the standard size are sequentially filled into the road blocks corresponding to the center point coordinates of the standardized base map of the corresponding level until the road blocks in the standardized base maps of different levels are filled.

5. The aerial image stitching method according to claim 1, characterized in that: The step of dividing the area to be spliced ​​according to the road refinement map of each level to obtain a map road block division map of a corresponding level includes: Extracting the road node network of each level from the road refinement graph of each level, and obtaining the connectivity information between the road nodes; the road node network is composed of a plurality of road nodes; According to the road node network, the area to be spliced ​​is divided into a number of closed map road blocks of corresponding levels; there are no empty areas between the map road blocks of the same level.

6. The aerial image stitching method according to claim 1, characterized in that: The method of obtaining a standardized base map of a corresponding level by using a scale change algorithm based on the road block division map of the corresponding level, combined with the scale information of the existing map and the preset shooting height of the aerial image, includes: Calculate scale transformation information based on the scale information of the existing map and the preset shooting height of the aerial image; According to the scale transformation information, the map road block division diagram is converted into the standardized base map of the corresponding level through a scale change algorithm.

7. An aerial image stitching device, characterized in that: The aerial image stitching device comprises: a standard base map acquisition module, a road block image acquisition module, an image filling module and an image overlay module; The standard base map acquisition module is used to obtain the road map information of the area to be spliced ​​from the existing map, and generate standardized base maps of several levels according to the road map information; the standardized base map includes: the center point coordinates of each map road block of the corresponding level, several road nodes of each map road block of the corresponding level, and the connectivity information between the road nodes; The road block image acquisition module is used to sequentially acquire the road block image in each of the aerial images of the corresponding level based on the standardized base map and in combination with the aerial images; The image filling module is used to sequentially fill the road block images into the standardized base map of the corresponding level to obtain mosaic images of different levels; The image superposition module is used to superimpose the mosaic image of the current level onto the mosaic image of the next level in sequence according to a predetermined level order to obtain a complete mosaic image; The standard base map acquisition module includes: a road refinement map extraction unit, a map road block division unit and a standardized base map generation unit; The road refinement map extraction unit is used to extract the road refinement maps of each level of the area to be spliced ​​from the existing map, and set the pixel points in the road refinement map that have two or more different adjacent pixel similarity points as the road nodes; The map road block division unit is used to divide the to-be-joined area according to the road refinement map of each level to obtain the map road block division map of the corresponding level; The standardized base map generation unit is used to obtain the standardized base map of the corresponding level through a scale change algorithm according to the map road block division map of the corresponding level, combined with the scale information of the existing map and the preset shooting height of the aerial image.

8. The aerial image stitching device according to claim 7, characterized in that: The road block image acquisition module includes: an aerial image acquisition unit and a road block image acquisition unit; Wherein, the aerial image acquisition unit is used to acquire the aerial image taken vertically by the aerial photography device at each of the center point coordinates; The road block image acquisition unit is used to sequentially acquire the road block image in each of the aerial images according to the road nodes of the map road block corresponding to the center point coordinates and the connectivity information between the road nodes.

9. The aerial image stitching device according to claim 8, characterized in that: The road block image acquisition unit includes: a distance calculation subunit, a node mapping subunit and an image division subunit; The distance calculation subunit is used to calculate the actual distance from the center point coordinate to each road node according to the center point coordinate and each road node of the corresponding map road block in sequence; The node mapping subunit is used to obtain the shooting height when the image is shot at the center point coordinates, and to sequentially obtain the mapping node of each road node in the aerial image by combining the shooting height and the actual distance; The image division subunit is used to divide the road block image in the aerial image according to the connectivity information between all the mapping nodes and the road nodes.

10. The aerial image stitching device according to claim 7, characterized in that: The image filling module includes: an image standardization unit and an image filling unit; The image standardization unit is used to sequentially combine the scale information of the existing map and the shooting height when the image is shot at the center point coordinates, and use a scale transformation algorithm to convert the road block image into a road block image of a standard size; The image filling unit is used to sequentially fill the road block images of the standard size into the road blocks corresponding to the center point coordinates of the standardized base map of the corresponding level until the road blocks in the standardized base maps of different levels are filled.

11. The aerial image stitching device according to claim 7, characterized in that: The map road block division unit includes: a map road node extraction subunit and a map road block division subunit; The map road node extraction subunit is used to extract the road node network of each level from the road refinement map of each level, and obtain the connectivity information between the road nodes; the road node network is composed of a number of road nodes; The map road block division subunit is used to divide the area to be spliced ​​into a number of closed map road blocks of corresponding levels according to the road node network; there are no empty areas between the map road blocks of the same level.

12. The aerial image stitching device according to claim 7, characterized in that: The standardized base map generation unit includes: a transformation scale calculation subunit and an image scale transformation subunit; The transformation scale calculation subunit is used to calculate the scale transformation information according to the scale information of the existing map and the preset shooting height of the aerial image; The image scale conversion subunit is used to convert the map road block division diagram into the standardized base map of the corresponding level through a scale change algorithm according to the scale conversion information.

13. A terminal device, characterized in that: The invention comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, an aerial image stitching method as claimed in any one of claims 1 to 6 is implemented.

14. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the aerial image stitching method according to any one of claims 1 to 6.

Images