Method and system for optimizing viewpoint network

Abstract

The invention discloses a viewpoint network optimization method and system, and the method comprises the steps: obtaining a quantization scanning region and an edge set array of a scene target, carrying out the viewpoint network construction of the quantization scanning region, and obtaining a first viewpoint network; performing connected viewpoint supplementation on the first viewpoint network to obtain a second viewpoint network; according to a preset first overlapping degree threshold value, carrying out overlapping connection point supplementation on the second viewpoint network to obtain a third viewpoint network; according to the edge set array, redundant viewpoint filtering is carried out on the third viewpoint network, and an optimized viewpoint network corresponding to the scene target is obtained. According to the method, the robustness of the viewpoint network and the comprehensiveness of the viewpoint scanning area can be effectively improved, and the integrity of scene target detail information collected by the viewpoint network can be improved. The invention relates to the technical field of sensor deployment.

Description

Technical Field

[0001] The present invention relates to the technical field of sensor deployment, and in particular to an optimization method and system for a viewpoint network. Background Art

[0002] With the wide application of lidar and camera sensors in the three-dimensional reconstruction of object surfaces, the viewpoint planning problem (VPP) has become an important research direction. The core of the VPP problem is to construct a viewpoint network with optimal sensor positions and postures so that the object surface can achieve a predetermined reconstruction goal in three-dimensional reconstruction.

[0003] Currently, traditional viewpoint planning methods usually construct a viewpoint network composed of multiple sensor positions and postures for scene targets based on geometric analysis. This method fails to fully consider the complexity of the environment, and it is prone to blind spots in the line of sight or un-covered viewpoint scanning areas, resulting in poor integrity of the detailed information of the captured scene targets. In addition, this method often only focuses on the relationship between local viewpoints, lacks a sufficient global perspective, and the robustness of the viewpoint network is not satisfactory.

[0004] Therefore, the problems existing in the prior art still need to be solved and optimized urgently. Summary of the Invention

[0005] An object of the present invention is to solve at least to some extent one of the technical problems existing in the related art.

[0006] To this end, an object of an embodiment of the present invention is to provide an optimization method and system for a viewpoint network. Among them, the method can effectively improve the robustness of the viewpoint network and the comprehensiveness of the viewpoint scanning area, which is beneficial to improving the integrity of the detailed information of the scene targets collected by the viewpoint network.

[0007] To achieve the above technical object, the technical solutions adopted in the embodiments of the present application include:

[0008] In a first aspect, an embodiment of the present application provides an optimization method for a viewpoint network, including:

[0009] Obtain a quantized scanning area and an edge set array of a scene target, and construct a viewpoint network for the quantized scanning area to obtain a first viewpoint network;

[0010] Perform connected viewpoint filling on the first viewpoint network to obtain a second viewpoint network. The second viewpoint network includes a number of second viewpoints and a number of first viewpoints originally in the first viewpoint network. The newly filled target viewpoints in the second viewpoint network have direct connection relationships with at least two non-target viewpoints in the second viewpoint network. The target viewpoint is any one of the second viewpoints, and the non-target viewpoints are the first viewpoints or second viewpoints other than the target viewpoint;

[0011] According to a preset first overlap degree threshold, perform overlapping connection point filling on the second viewpoint network to obtain a third viewpoint network;

[0012] According to the edge set array, perform redundant viewpoint filtering on the third viewpoint network to obtain an optimized viewpoint network corresponding to the scene target.

[0013] In addition, according to the method of the above embodiments of the present application, the following additional technical features may also be included:

[0014] Further, in an embodiment of the present application, the constructing a first viewpoint network for the quantization scanning area includes:

[0015] Obtain a preset viewpoint adjacent pixel threshold;

[0016] Perform region imaging on the quantization scanning area to obtain a first region image;

[0017] Perform boundary transformation on the first region image to obtain a second region image;

[0018] Perform boundary inversion on the second region image to obtain a third region image;

[0019] Perform ridge line extraction and skeleton thinning on the third region image to obtain a thinned skeleton line image;

[0020] According to the viewpoint adjacent pixel threshold, perform viewpoint extraction on the skeleton line intersection points in the thinned skeleton line image to obtain the first viewpoint network.

[0021] Further, in an embodiment of the present application, the performing connected viewpoint filling on the first viewpoint network to obtain a second viewpoint network includes:

[0022] Use the current first viewpoint network as a fourth viewpoint network and obtain the first connection paths of each pair of fourth viewpoints in the fourth viewpoint network;

[0023] Perform farthest point filling on the fourth viewpoint network according to all the first connection paths to obtain a fifth viewpoint network and a farthest point new identifier of the fifth viewpoint network, where the farthest point new identifier is used to indicate whether the fifth viewpoint network is the fourth viewpoint network after adding the farthest point;

[0024] If the farthest point new identifier indicates that the fifth viewpoint network is the fourth viewpoint network after adding the farthest point, update the current first viewpoint network according to the fifth viewpoint network, and then return to execute the step of using the current first viewpoint network as the fourth viewpoint network and obtaining the first connection path of each fourth viewpoint pair in the fourth viewpoint network; or, if the farthest point new identifier indicates that the fifth viewpoint network is not the fourth viewpoint network after adding the farthest point, determine the fifth viewpoint network as the second viewpoint network.

[0025] Further, in an embodiment of the present application, the performing farthest point filling on the fourth viewpoint network according to all the first connection paths to obtain a fifth viewpoint network and a farthest point new identifier of the fifth viewpoint network includes:

[0026] Perform viewpoint pair screening on the fourth viewpoint network according to all the first connection paths to obtain a number of fifth viewpoint pairs and a second connection path corresponding to each fifth viewpoint pair, where the fifth viewpoint pair is a fourth viewpoint pair with a direct connection relationship;

[0027] Perform farthest point search on the corresponding fifth viewpoint pairs according to all the second connection paths to obtain a number of target farthest points;

[0028] Perform first viewpoint filling and updating on the fourth viewpoint network according to all the target farthest points to obtain the fifth viewpoint network and the farthest point new identifier.

[0029] Further, in an embodiment of the present application, performing farthest point search on the corresponding fifth viewpoint pairs according to the second connection path to obtain target farthest points includes:

[0030] Obtain the current fifth viewpoint pair and the distance threshold of the current fifth viewpoint pair;

[0031] Perform farthest point search on the second connection path of the current fifth viewpoint pair to obtain an intermediate farthest point, where the intermediate farthest point is the path pixel point with the largest viewpoint distance among all the path pixel points of the second connection path, and the viewpoint distance is the minimum interval distance between the path pixel point and the straight line where the current fifth viewpoint pair is located;

[0032] Compare the distance threshold with the viewpoint distance of the intermediate farthest point to obtain a distance comparison result;

[0033] If the distance comparison result is that the viewing distance of the middle farthest point is greater than or equal to the distance threshold, then determine the middle farthest point as the target farthest point.

[0034] Further, in an embodiment of the present application, the step of adding overlapping connection points to the second viewpoint network according to a preset first overlap degree threshold to obtain a third viewpoint network includes:

[0035] Take the current second viewpoint network as the sixth viewpoint network, and obtain the first viewpoint overlap degree of each sixth viewpoint pair in the sixth viewpoint network;

[0036] According to the first overlap degree threshold and all the first viewpoint overlap degrees, perform middle point filling on the sixth viewpoint network to obtain a seventh viewpoint network and a middle point new addition identifier of the seventh viewpoint network, where the middle point new addition identifier is used to indicate whether the seventh viewpoint network is the sixth viewpoint network after adding middle points;

[0037] If the middle point new addition identifier indicates that the seventh viewpoint network is the sixth viewpoint network after adding middle points, then update the current second viewpoint network according to the seventh viewpoint network, and then return to execute the step of taking the current second viewpoint network as the sixth viewpoint network and obtaining the first viewpoint overlap degree of each sixth viewpoint pair in the sixth viewpoint network; or, if the middle point new addition identifier indicates that the seventh viewpoint network is not the sixth viewpoint network after adding middle points, then determine the seventh viewpoint network as the third viewpoint network.

[0038] Further, in an embodiment of the present application, the step of performing middle point filling on the sixth viewpoint network according to the first overlap degree threshold and all the first viewpoint overlap degrees to obtain a seventh viewpoint network and a middle point new addition identifier of the seventh viewpoint network includes:

[0039] According to the first overlap degree threshold, perform overlap degree comparison on all the first viewpoint overlap degrees to obtain an overlap degree comparison result corresponding to each first viewpoint overlap degree;

[0040] According to all the overlap degree comparison results, perform viewpoint pair screening on the sixth viewpoint network to obtain a number of seventh viewpoint pairs, where the seventh viewpoint pair is the sixth viewpoint pair whose corresponding first viewpoint overlap degree is less than the first overlap degree threshold;

[0041] Perform middle point selection on all the seventh viewpoint pairs to obtain the target middle point of each seventh viewpoint pair;

[0042] Based on all the target intermediate points, perform second viewpoint filling and updating on the sixth viewpoint network to obtain the seventh viewpoint network and the newly added intermediate point identifiers.

[0043] Further, in an embodiment of the present application, the redundant viewpoint filtering of the third viewpoint network according to the edge set array to obtain the optimized viewpoint network corresponding to the scene target includes:

[0044] Perform uniform sampling on the edge set array to obtain a segmented edge set, and the segmented edge set includes several scene segmentation edges of the scene target;

[0045] According to the segmented edge set, perform overlapping information extraction on the third viewpoint network to obtain a scan information table, and the scan information table includes several scan information arrays. Each scan information array corresponds to a third viewpoint in the third viewpoint network, and the scan information array includes several scan array elements. Each scan array element is used to represent whether the viewpoint scan edge of the third viewpoint overlaps with the corresponding scene segmentation edge;

[0046] Perform heuristic optimization on the scan information table to obtain the optimized viewpoint network.

[0047] Further, in an embodiment of the present application, the performing heuristic optimization on the scan information table to obtain the optimized viewpoint network includes:

[0048] Obtain a preset second overlap degree threshold and an edge number threshold;

[0049] Take the current scan information table as the first information table and obtain a preferred viewpoint set corresponding to the first information table;

[0050] Perform preferred viewpoint information screening on the first information table to obtain a preferred viewpoint array and preferred viewpoints corresponding to the preferred viewpoint array. The preferred viewpoint array is the scan information array with the largest number of overlapping relationships in the first information table;

[0051] According to the preferred viewpoint array, perform row and column information update on the first information table to obtain a second information table and several information viewpoints corresponding to the second information table;

[0052] According to the second overlap degree threshold and the preferred viewpoints, perform overlap degree screening on all the information viewpoints to obtain several candidate viewpoints. Each candidate viewpoint is an information viewpoint with a second viewpoint overlap degree greater than or equal to the second overlap degree threshold, and the second viewpoint overlap degree is the viewpoint overlap degree between the information viewpoint and the preferred viewpoint;

[0053] According to the edge number threshold, compare the arrays of scanning information of all the candidate viewpoints to obtain an edge number comparison result, where the edge number comparison result is used to represent whether there is at least one candidate viewpoint whose total supplementary edge number is greater than the edge number threshold, and the total supplementary edge number is the total number of scene segmentation edges overlapping with the viewpoint scanning edges of the candidate viewpoint;

[0054] If the edge number comparison result is that there is no at least one candidate viewpoint whose total supplementary edge number is greater than the edge number threshold, then update the preferred viewpoint set according to the preferred viewpoint to obtain an updated preferred viewpoint set, and obtain the optimized viewpoint network according to the updated preferred viewpoint set; or, if the edge number comparison result is that there is no at least one candidate viewpoint whose total supplementary edge number is greater than the edge number threshold, then update the preferred viewpoint set according to the preferred viewpoint and the candidate viewpoint with the largest total supplementary edge number, and update the current scanning information table according to the candidate viewpoint with the largest total supplementary edge number and the second information table, and then return to execute the step of using the current scanning information table as the first information table and obtaining the preferred viewpoint set corresponding to the first information table.

[0055] In a second aspect, an embodiment of the present application provides an optimization system for a viewpoint network, including:

[0056] A first processing unit, configured to obtain an array of edge sets of a quantization scanning area and a scene target, and construct a viewpoint network for the quantization scanning area to obtain a first viewpoint network;

[0057] A second processing unit, configured to perform connected viewpoint filling on the first viewpoint network to obtain a second viewpoint network, where the second viewpoint network includes several second viewpoints and several first viewpoints originally in the first viewpoint network, and the newly filled target viewpoints in the second viewpoint network are at least directly connected to two non-target viewpoints in the second viewpoint network, the target viewpoint is any one of the second viewpoints, and the non-target viewpoint is the first viewpoint or a second viewpoint other than the target viewpoint;

[0058] A third processing unit, configured to perform overlapping connection point filling on the second viewpoint network according to a preset first overlap degree threshold to obtain a third viewpoint network;

[0059] A fourth processing unit, configured to filter redundant viewpoints from the third viewpoint network according to the edge set array to obtain an optimized viewpoint network corresponding to the scene target.

[0060] In a third aspect, an embodiment of the present application further provides an electronic device, including:

[0061] At least one processor;

[0062] At least one memory for storing at least one program;

[0063] When the at least one program is executed by the at least one processor, the at least one processor implements the above-mentioned method.

[0064] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, in which a program executable by a processor is stored, and the program executable by the processor is used to implement the above-mentioned method when executed by the processor.

[0065] The advantages and beneficial effects of the present application will be partially given in the following description, partially will become obvious from the following description, or will be learned through the practice of the present application:

[0066] An optimization method and system for a viewpoint network disclosed in an embodiment of the present application, wherein the method obtains a quantized scan region and an edge set array of a scene target, constructs a viewpoint network for the quantized scan region to obtain a first viewpoint network; performs connected viewpoint filling on the first viewpoint network to obtain a second viewpoint network, the second viewpoint network includes a plurality of second viewpoints and a plurality of first viewpoints originally in the first viewpoint network, and the newly filled target viewpoints in the second viewpoint network are directly connected to at least two non-target viewpoints in the second viewpoint network, the target viewpoint is any one of the second viewpoints, and the non-target viewpoint is the first viewpoint or a second viewpoint other than the target viewpoint; according to a preset first overlap degree threshold, perform overlapping connection point filling on the second viewpoint network to obtain a third viewpoint network; according to the edge set array, perform redundant viewpoint filtering on the third viewpoint network to obtain an optimized viewpoint network corresponding to the scene target. By constructing a viewpoint network for the quantized scan region and performing connected viewpoint filling on the first viewpoint network, the method can fully consider the global connectivity between viewpoints in the viewpoint network, provide a sufficient global perspective for the viewpoint network, effectively improve the robustness of the viewpoint network and the comprehensiveness of the viewpoint scanning region, and is beneficial to improving the integrity of the detailed information of the scene target subsequently collected by the viewpoint network. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following introduces the accompanying drawings related to the technical solutions in the embodiments of the present application or the prior art. It should be understood that the accompanying drawings in the following introduction are only for conveniently and clearly expressing some embodiments of the technical solutions in the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative efforts.

[0068] Figure 1Schematic flowchart of an optimization method for a viewpoint network provided by an embodiment of the present application;

[0069] Figure 2 Schematic diagram of a quantization scanning area provided by an embodiment of the present application;

[0070] Figure 3 Schematic diagram of a viewpoint scanning edge of a sixth viewpoint pair provided by an embodiment of the present application

[0071] Figure 4 Schematic structural framework diagram of an optimization system for a viewpoint network provided by an embodiment of the present application;

[0072] Figure 5 Schematic structural diagram of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0073] The embodiments of the present application will be described in detail below. Examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals indicate the same or similar elements or elements with the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and should not be construed as a limitation to the present application. For the step numbers in the following embodiments, they are only set for the convenience of explanation and illustration, and no limitation is imposed on the order between the steps. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0074] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0075] Currently, traditional viewpoint planning methods usually construct a viewpoint network composed of the positions and postures of multiple sensors for scene targets based on geometric analysis. This method fails to fully consider the complexity of the environment, and it is prone to line-of-sight blind spots or uncovered viewpoint scanning areas, resulting in poor integrity of the detailed information of the captured scene targets; in addition, this method often only focuses on the relationship between local viewpoints, lacks a sufficient global perspective, and the robustness of the viewpoint network is not satisfactory.

[0076] In addition, the method based on geometric analysis often fails to reconstruct scene targets with as few viewpoints as possible. The constructed viewpoint network uses a relatively large number of viewpoints, resulting in a high hardware cost for the viewpoint network. Moreover, the subsequent acquisition time of the three-dimensional point cloud is relatively long, and the computational resources required for three-dimensional reconstruction are relatively large, leading to poor efficiency of subsequent three-dimensional reconstruction. Additionally, for the purpose of enabling the line of sight to capture every detail of the scene target as much as possible, the viewpoints of the constructed viewpoint network are often unevenly distributed in space, with insufficient or unbalanced overlapping areas in viewpoint planning, and the robustness of the viewpoint network is poor.

[0077] In view of this, embodiments of the present invention provide an optimization method and system for a viewpoint network. Specifically, the method constructs a viewpoint network for a quantized scanning area by extracting the skeleton of the quantized scanning area. The joints of the skeleton represent the convergence points of the observation information of the viewpoints. Then, a first viewpoint network is constructed based on the skeleton, which can provide a sufficient global perspective for the viewpoint network and effectively improve the comprehensiveness of the viewpoint scanning area.

[0078] Furthermore, the method fills in connected viewpoints for the first viewpoint network by selectively adding the farthest points on the connected paths between the viewpoints of the first viewpoint network. This can fully consider the global connectivity between the viewpoints of the viewpoint network, effectively improve the robustness of the viewpoint network and the comprehensiveness of the viewpoint scanning area, and is conducive to improving the integrity of the detailed information of the scene target subsequently collected by the viewpoint network. At the same time, since the newly added farthest points are added on the connected paths between the viewpoints, that is, the newly added farthest points are placed on the skeleton and joints, it can effectively ensure that the entire scene target space is covered with the fewest viewpoints, effectively reduce the number of viewpoints used by the viewpoint network, and is conducive to reducing the hardware cost of the viewpoint network, reducing the subsequent acquisition time of the three-dimensional point cloud and the computational resources required for three-dimensional reconstruction.

[0079] In addition, the method fills in overlapping connection points for the second viewpoint network by selectively adding intermediate points on the connected paths between the viewpoints of the second viewpoint network. This can further consider the viewpoint planning overlapping area between the viewpoints of the viewpoint network on the basis of fully considering the global connectivity between the viewpoints of the viewpoint network, which is conducive to improving the robustness of the viewpoint network.

[0080] Moreover, the method filters redundant viewpoints from the third viewpoint network. By fully considering the global connectivity and overlapping area of each viewpoint of the viewpoint network, the redundant viewpoints of the viewpoint network can be removed, and the obtained viewpoint network can be further optimized to cover the entire scene target space with the fewest viewpoints, which is conducive to reducing the hardware cost of the viewpoint network, reducing the subsequent acquisition time of the three-dimensional point cloud and the computational resources required for three-dimensional reconstruction.

[0081] Refer to Figure 1, in the embodiments of the present application, an optimization method for a viewpoint network includes:

[0082] Step 110: Obtain a quantized scan region and an edge set array of a scene target, and construct a viewpoint network for the quantized scan region to obtain a first viewpoint network;

[0083] In the embodiments of the present application, the scene target may be a scene environment to be scanned (such as a spatial environment inside a house) or an object to be scanned, and the quantized scan region may be a quantized representation of the scan region obtained after a scan sensor (such as a LiDAR sensor) scans the scene target.

[0084] Exemplarily, referring to Figure 2 , for a certain scan sensor, its scan range parameters include a minimum scan radius r min and a maximum scan radius r max , and these scan range parameters are used to characterize the object distance and viewing angle that the scan sensor can capture. In plane geometry, set the position of the scan sensor as the center O, and use r min and r max to draw two circles respectively, which intersect with the target line segment (line segment AB) in the scene target. The intersection points with r min are C and D, and the intersection points with r max are E and F. The quantized scan region in line segment AB that can be scanned by this scan sensor is:

[0085] S valid = intersect(AB, EF) - intersect(AB, CD)

[0086] where, S valid is the quantized scan region in line segment AB that can be scanned by this scan sensor, that is, the effective region of the viewpoint scan edge of this scan sensor. This viewpoint scan edge includes line segment EC and line segment DB; intersect(·) is the intersection symbol.

[0087] It can be understood that the edge set array may be the edge position information of the scene target, etc. In the embodiments of the present application, taking the scene target as a house scene as an example, its edge set array can obtain the wall line information of the house scene from the two-dimensional plane information file (such as DXF file, etc.) of the house scene. This wall line information includes side lengths, position coordinates, etc., and an edge set array is generated accordingly.

[0088] It should be noted that after obtaining the quantized scan region, an initial viewpoint network corresponding to the scene target can be constructed based on the quantized scan region of the scan sensor, and the constructed initial viewpoint network is determined as the first viewpoint network.

[0089] In some embodiments, step 110, constructing a view point network for the quantized scanning region to obtain a first view point network, includes:

[0090] A1. Obtain a preset adjacent pixel threshold for view points;

[0091] A2. Image the quantized scanning region to obtain a first region image;

[0092] A3. Perform boundary transformation on the first region image to obtain a second region image;

[0093] A4. Perform boundary inversion on the second region image to obtain a third region image;

[0094] A5. Extract ridge lines and thin the skeleton from the third region image to obtain a thinned skeleton line image;

[0095] A6. Extract view points from the skeleton line intersection points in the thinned skeleton line image according to the adjacent pixel threshold for view points to obtain the first view point network.

[0096] In the embodiments of the present application, the adjacent pixel threshold for view points is used to characterize the threshold of adjacent pixel points on the skeleton line graph. This threshold can specifically be the minimum threshold, and its specific array can be set according to actual situations. In the embodiments of the present application, the adjacent pixel threshold for view points is taken as 3 as an example. Step A2 can be to perform image preprocessing such as graying, normalizing, and binarizing on the quantized scanning region in sequence, so as to obtain a binary image of the quantized scanning region, and determine this binary image as the first region image.

[0097] It can be understood that step A3 can be to add a boundary to the first region image and perform distance transformation (such as Euclidean distance transformation) on the first region image after adding the boundary, so as to obtain a second region image; then, the boundary originally added to the second region image can be removed to maintain the original size of the image, and the pixel values of each pixel point in the second region image can be flipped, so as to obtain a third region image, and this third region image can be used to extract the skeleton line.

[0098] It should be noted that step A5 may be to perform ridge filtering on the third region image to obtain the topological skeleton line of the quantization scanning region, and then refine the topological skeleton line to obtain the refined skeleton line image; for the skeleton line in the refined skeleton line image, the point with the most concentrated joint representation information, so step A6 may be to screen the intersection points in the skeleton line based on the adjacent pixel threshold of the viewpoint, specifically, the pixel points with no less than 3 adjacent points on the refined skeleton line image can be determined as intersection points, so as to obtain a number of intersection points, and all the intersection points form an initial viewpoint network (i.e., the first viewpoint network). Each viewpoint in the first viewpoint network corresponds to the position of a scanning sensor, and the first viewpoint network composed of the initial skeleton intersection points can ensure full coverage of the scene target in the quantization scanning region.

[0099] Step 120: Perform connected viewpoint filling on the first viewpoint network to obtain a second viewpoint network. The second viewpoint network includes several second viewpoints and several first viewpoints originally in the first viewpoint network. The newly filled target viewpoints in the second viewpoint network have a direct connection relationship with at least two non-target viewpoints in the second viewpoint network. The target viewpoint is any one of the second viewpoints, and the non-target viewpoint is the first viewpoint or a second viewpoint other than the target viewpoint.

[0100] In the embodiment of the present application, connected viewpoints can be added based on several first viewpoints in the first viewpoint network to obtain a second viewpoint network. Among them, the newly added second viewpoints are at least connected to two other viewpoints in the second viewpoint network except themselves.

[0101] In some embodiments, step 120: Perform connected viewpoint filling on the first viewpoint network to obtain a second viewpoint network, includes:

[0102] B1: Take the current first viewpoint network as the fourth viewpoint network and obtain the first connection path for each pair of fourth viewpoints in the fourth viewpoint network.

[0103] In the embodiment of the present application, for the connected viewpoint filling in a certain cycle, first, the current latest first viewpoint network can be taken as the fourth viewpoint network, and the fourth viewpoint pairs formed by each viewpoint in the fourth viewpoint network and other viewpoints are obtained, and then based on the fourth viewpoint pairs, the first connection path between two fourth viewpoints is determined.

[0104] It should be noted that in the fourth viewpoint network of the current loop iteration, the fourth viewpoint can be the original first viewpoint of the first viewpoint network or a newly added connected viewpoint in a previous loop iteration. Additionally, since there are a certain number of unconnected viewpoint pairs in the first viewpoint network, among all the fourth viewpoint pairs obtained in step B1, there may also be some unconnected fourth viewpoint pairs. For unconnected fourth viewpoint pairs, the corresponding first connected path can be empty.

[0105] B2. According to all the first connected paths, perform farthest point filling on the fourth viewpoint network to obtain a fifth viewpoint network and a farthest point new addition identifier of the fifth viewpoint network. The farthest point new addition identifier is used to indicate whether the fifth viewpoint network is the fourth viewpoint network after adding the farthest point.

[0106] In some embodiments, step B2. According to all the first connected paths, perform farthest point filling on the fourth viewpoint network to obtain a fifth viewpoint network and a farthest point new addition identifier of the fifth viewpoint network, includes:

[0107] B21. According to all the first connected paths, perform viewpoint pair screening on the fourth viewpoint network to obtain a plurality of fifth viewpoint pairs and a second connected path corresponding to each fifth viewpoint pair. The fifth viewpoint pair is a fourth viewpoint pair with a direct connection relationship.

[0108] In the embodiments of the present application, based on all the first connected paths, the farthest point can be filled for the corresponding fourth viewpoint pairs in the fourth viewpoint network, thereby obtaining a fifth viewpoint network and a farthest point new addition identifier. Specifically, if at least one farthest point is successfully filled in the fourth viewpoint network, a farthest point new addition identifier indicating that the fifth viewpoint network is the fourth viewpoint network after adding the farthest point can be obtained; or, if at least one farthest point is not successfully filled in the four-viewpoint network, a farthest point new addition identifier indicating that the fifth viewpoint network is not the fourth viewpoint network after adding the farthest point can be obtained, that is, the fifth viewpoint network is the same as the fourth viewpoint network.

[0109] It can be understood that step B21 can be to delete the fourth viewpoint pairs with an empty first connected path among all the fourth viewpoint pairs and retain the fourth viewpoint pairs with a non-empty first connected path, thereby obtaining a plurality of fifth viewpoint pairs with a connection relationship and a second connected path corresponding to each fifth viewpoint.

[0110] It should be noted that for a fifth viewpoint pair with a direct connection relationship, there are no other viewpoints on the connection path between the two fifth viewpoints of the fifth viewpoint pair, and the connection path between the two fifth viewpoints can be a certain section of the skeleton line in the refined skeleton line image.

[0111] B22. Perform a farthest point search on the corresponding fifth viewpoint pair according to all the second connection paths to obtain a number of target farthest points;

[0112] Further, the step B22, performing a farthest point search on the corresponding fifth viewpoint pair according to the second connection path to obtain a target farthest point, includes:

[0113] B221. Obtain the current fifth viewpoint pair and the distance threshold of the current fifth viewpoint pair;

[0114] B222. Perform a farthest point search on the second connection path of the current fifth viewpoint pair to obtain an intermediate farthest point, where the intermediate farthest point is the path pixel point with the largest viewpoint distance among all the path pixel points of the second connection path, and the viewpoint distance is the minimum distance between the path pixel point and the line where the current fifth viewpoint pair is located;

[0115] B223. Compare the distance threshold with the viewpoint distance of the intermediate farthest point to obtain a distance comparison result;

[0116] B224. If the distance comparison result is that the viewpoint distance of the intermediate farthest point is greater than or equal to the distance threshold, then determine the intermediate farthest point as the target farthest point.

[0117] In the embodiments of the present application, according to all the second connection paths, each fifth viewpoint pair can be traversed in a loop to obtain a number of target farthest points. The number of the target farthest points is not necessarily the same as the number of the fifth viewpoint pairs. Specifically, if there is a qualified farthest point for each fifth viewpoint pair, then the number of the obtained target farthest points is the same as the number of the fifth viewpoint pairs; otherwise, if there is at least one fifth viewpoint pair whose farthest point does not meet the conditions, then the number of the obtained target farthest points is less than the number of the fifth viewpoint pairs.

[0118] It can be understood that for a certain loop traversal process, first, the fifth viewpoint pair corresponding to the current loop traversal process and the distance threshold can be obtained. There are already various specific ways to obtain the distance threshold. For example, it can be a preset value, or it can be obtained based on querying a distance threshold value table. The embodiments of the present application will not elaborate on this here. In addition, in the embodiments of the present application, all the fifth viewpoint pairs can use the same distance threshold, or the distance thresholds corresponding to all the fifth viewpoint pairs are not completely the same.

[0119] It should be noted that step B222 may be to find a path pixel point in the second connected path of the fifth viewpoint pair, so that the viewpoint distance from the path pixel point to the straight line where the corresponding two fifth viewpoints are located is the largest. Specifically, step B222 may first be to determine the corresponding viewpoint straight line based on the two fifth viewpoints, and then find a path pixel point with the largest minimum interval distance from the viewpoint straight line in the corresponding second connected path, and determine this path pixel point as the intermediate farthest point. In addition, the minimum interval distance between the path pixel point and the viewpoint straight line can be obtained by connecting the path pixel point and the viewpoint straight line in a perpendicular manner on the viewpoint straight line, and this application will not elaborate here.

[0120] It is worth mentioning that step B223 may be to compare the magnitude relationship between the distance threshold and the viewpoint distance to obtain a distance comparison result. Specifically, if the distance comparison result is that the viewpoint distance of the intermediate farthest point is greater than or equal to the distance threshold, it means that this intermediate farthest point meets the conditions. At this time, this intermediate farthest point can be determined as the target farthest point corresponding to the current fifth viewpoint pair, and the current fifth viewpoint pair is updated, and step B221 is returned to execute to find the farthest point for the next fifth viewpoint pair; or, if the distance comparison result is that the viewpoint distance of the intermediate farthest point is less than the distance threshold, it means that this intermediate farthest point does not meet the conditions. At this time, this intermediate farthest point can be deleted, and the current fifth viewpoint pair is updated, and step B221 is returned to execute to find the farthest point for the next fifth viewpoint pair. The same applies to the fifth viewpoint pairs in the remaining loop traversal process, and it can be simply deduced by analogy.

[0121] B23. According to all the target farthest points, perform the first viewpoint filling and updating on the fourth viewpoint network to obtain the fifth viewpoint network and the farthest point new addition identifier.

[0122] In the embodiment of this application, after all the fifth viewpoint pairs have been looped through, all the generated target farthest points can be filled into the fourth viewpoint network to obtain the fifth viewpoint network. Specifically, if the total number of all the generated target farthest points is 0, the obtained fifth viewpoint network is the same as the fourth viewpoint network, and the generated farthest point new addition identifier is that the fifth viewpoint network is not the fourth viewpoint network after adding the farthest point; or, if the total number of all the generated target farthest points is greater than or equal to 1, the generated farthest point new addition identifier is that the fifth viewpoint network is the fourth viewpoint network after adding the farthest point.

[0123] B3. If the farthest point new addition identifier is that the fifth viewpoint network is the fourth viewpoint network after adding the farthest point, then update the current first viewpoint network according to the fifth viewpoint network, and then return to execute the step of using the current first viewpoint network as the fourth viewpoint network and obtaining the first connected path of each fourth viewpoint pair in the fourth viewpoint network.

[0124] Or, B4. If the newly added identifier of the farthest point indicates that the fifth viewpoint network is not the fourth viewpoint network after adding the farthest point, then determine the fifth viewpoint network as the second viewpoint network.

[0125] In the embodiment of the present application, if the newly added identifier of the farthest point indicates that the fifth viewpoint network is the fourth viewpoint network after adding the farthest point, it means that there may still be the farthest points in the fifth viewpoint network whose viewpoint distances are greater than or equal to the distance threshold and can be added. At this time, the fifth viewpoint network can be used as the latest first viewpoint network, and then return to execute B1 to continue adding the farthest point; or, if the newly added identifier of the farthest point indicates that the fifth viewpoint network is not the fourth viewpoint network after adding the farthest point, it means that the fifth viewpoint network is the same as the fourth viewpoint network, and there are no farthest points in the fifth viewpoint network whose viewpoint distances are greater than or equal to the distance threshold and can be added. At this time, the current fifth viewpoint network can be determined as the second viewpoint network.

[0126] Step 130. According to a preset first overlap degree threshold, perform overlap connection point filling on the second viewpoint network to obtain a third viewpoint network;

[0127] In the embodiment of the present application, according to the first overlap degree threshold, overlap connection points can be added on the basis of several second viewpoints in the first viewpoint network, so as to obtain a third viewpoint network, which can fully consider the viewpoint planning overlap area between each viewpoint of the viewpoint network.

[0128] In some embodiments, the step 130. According to a preset first overlap degree threshold, perform overlap connection point filling on the second viewpoint network to obtain a third viewpoint network, includes:

[0129] C1. Use the current second viewpoint network as the sixth viewpoint network, and obtain the first viewpoint overlap degree of each sixth viewpoint pair in the sixth viewpoint network;

[0130] In the embodiment of the present application, for the overlap connection point filling of a certain cycle round, first, the current latest second viewpoint network can be used as the sixth viewpoint network, and each viewpoint in the sixth viewpoint network and other viewpoints are used to form a sixth viewpoint pair, and then based on the sixth viewpoint pair, the first viewpoint overlap degree of each sixth viewpoint pair is determined.

[0131] Exemplarily, referring to Figure 3 , for the sixth viewpoint a and the sixth viewpoint b in the sixth viewpoint pair, the viewpoint scan edge observed by the sixth viewpoint a can be expressed as L a , and the viewpoint scan edge observed by the sixth viewpoint b can be expressed as L b , then the common edge segment of the sixth viewpoint a and the sixth viewpoint b can be expressed as:

[0132] L ab = L a ∩L b

[0133] where L ab is the common edge segment of the sixth viewpoints a and b, and this common edge segment is Figure 3 the line segments GH and IJ in; ∩ is the intersection symbol.

[0134] The first viewpoint overlap degree of the sixth viewpoint pair ab corresponding to the sixth viewpoints a and b can be expressed as:

[0135]

[0136] where O ab is the first viewpoint overlap degree of the sixth viewpoint pair ab.

[0137] C2. According to the first overlap degree threshold and all the first viewpoint overlap degrees, perform middle point filling on the sixth viewpoint network to obtain a seventh viewpoint network and a middle point new identifier of the seventh viewpoint network, and the middle point new identifier is used to indicate whether the seventh viewpoint network is the sixth viewpoint network after adding middle points;

[0138] Further, the step C2. According to the first overlap degree threshold and all the first viewpoint overlap degrees, perform middle point filling on the sixth viewpoint network to obtain a seventh viewpoint network and a middle point new identifier of the seventh viewpoint network, includes:

[0139] C21. According to the first overlap degree threshold, perform overlap degree comparison on all the first viewpoint overlap degrees to obtain an overlap degree comparison result corresponding to each first viewpoint overlap degree;

[0140] C22. According to all the overlap degree comparison results, perform viewpoint pair screening on the sixth viewpoint network to obtain a number of seventh viewpoint pairs, and the seventh viewpoint pairs are the sixth viewpoint pairs whose corresponding first viewpoint overlap degrees are less than the first overlap degree threshold;

[0141] C23. Select middle points for all the seventh viewpoint pairs to obtain a target middle point for each seventh viewpoint pair;

[0142] C24. According to all the target middle points, perform second viewpoint filling and updating on the sixth viewpoint network to obtain the seventh viewpoint network and the middle point new identifier.

[0143] In an embodiment of the present application, intermediate points can be added to corresponding sixth viewpoint pairs in the sixth viewpoint network based on the first overlap degree threshold and the first viewpoint overlap degree of each sixth viewpoint pair, so as to obtain a seventh viewpoint network and an intermediate point new identifier. The content of this intermediate point new identifier is similar to that of the aforementioned farthest point new identifier and can be simply analogized.

[0144] It can be understood that step C21 can be to compare the size relationship between each first viewpoint overlap degree and the first overlap degree threshold, so as to obtain an overlap degree comparison result corresponding to each first viewpoint overlap degree; step C22 can be to screen the corresponding sixth viewpoint pairs in the sixth viewpoint network based on each overlap degree comparison result, so as to obtain several seventh viewpoint pairs. Specifically, for a certain overlap degree comparison result, if the overlap degree comparison result is that the first viewpoint overlap degree is greater than or equal to the first overlap degree threshold, it means that the overlap degree of the sixth viewpoint pair corresponding to this overlap degree comparison result is sufficient, that is, there is no shortage or imbalance in the viewpoint planning overlap area of the two sixth viewpoints in this sixth viewpoint pair; or, if the overlap degree comparison result is that the first viewpoint overlap degree is less than the first overlap degree threshold, it means that the overlap degree of the sixth viewpoint pair corresponding to this overlap degree comparison result is insufficient, that is, there is a shortage or imbalance in the viewpoint planning overlap area of the two sixth viewpoints in this sixth viewpoint pair. At this time, this sixth viewpoint pair can be determined as a seventh viewpoint pair, and the same applies to the other overlap degree comparison results, which can be simply analogized.

[0145] It should be noted that for a certain seventh viewpoint pair, step C23 can be to select an intermediate point on the connection path between the two seventh viewpoints of this seventh viewpoint pair and determine the selected intermediate point as the target intermediate point of this seventh viewpoint pair, where the intermediate point is the path pixel point located in the middle of the connection path of this seventh viewpoint pair, and the same applies to the other seventh viewpoint pairs, which can be simply analogized. Then, after obtaining the target intermediate point of each seventh viewpoint pair, all the obtained target intermediate points can be filled into the sixth viewpoint network to obtain the seventh viewpoint network.

[0146] It is worth mentioning that the number of viewpoint pairs of the seventh viewpoint pairs obtained in step C22 can be 0 or greater than or equal to 1. If the number of viewpoint pairs of the seventh viewpoint pairs obtained in step C22 is 0, then the number of target intermediate points obtained in step C23 is also 0. At this time, the seventh viewpoint network obtained in step C24 is the same as the sixth viewpoint network, and the corresponding intermediate point new identifier is that the seventh viewpoint network is not the sixth viewpoint network after adding intermediate points; or, if the number of viewpoint pairs of the seventh viewpoint pairs obtained in step C22 is not 0, at this time, the seventh viewpoint network obtained in step C24 is the sixth viewpoint network after adding intermediate points, and the corresponding intermediate point new identifier is that the seventh viewpoint network is the sixth viewpoint network after adding intermediate points.

[0147] C3. If the newly added identifier of the intermediate point indicates that the seventh viewpoint network is the sixth viewpoint network after adding the intermediate point, then update the current second viewpoint network according to the seventh viewpoint network, and then return to execute the step of using the current second viewpoint network as the sixth viewpoint network and obtaining the first viewpoint overlap degree of each sixth viewpoint pair in the sixth viewpoint network;

[0148] Or, C4. If the newly added identifier of the intermediate point indicates that the seventh viewpoint network is not the sixth viewpoint network after adding the intermediate point, then determine the seventh viewpoint network as the third viewpoint network.

[0149] In the embodiments of the present application, the content of steps C3 to C4 is similar to the content of the foregoing steps B3 to B4, and can be simply analogized. Therefore, the present application will not elaborate here.

[0150] Step 140. Filter redundant viewpoints from the third viewpoint network according to the edge set array to obtain an optimized viewpoint network corresponding to the scene target.

[0151] In the embodiments of the present application, redundant viewpoints in the third viewpoint network can be filtered according to the edge position information of the scene target in the edge set array, etc., so as to obtain an optimized viewpoint network. This optimized viewpoint network can not only fully consider the global connectivity of each viewpoint in the viewpoint network and the overlapping area, but also use the minimum number of viewpoints to cover the space of the entire scene target.

[0152] In some embodiments, the step 140. Filter redundant viewpoints from the third viewpoint network according to the edge set array to obtain an optimized viewpoint network corresponding to the scene target, includes:

[0153] D1. Uniformly sample the edge set array to obtain a segmented edge set, and the segmented edge set includes several scene segmentation edges of the scene target;

[0154] D2. Extract overlapping information from the third viewpoint network according to the segmented edge set to obtain a scan information table. The scan information table includes several scan information arrays. Each scan information array corresponds to a third viewpoint in the third viewpoint network. The scan information array includes several scan array elements, and each scan array element is used to represent whether the viewpoint scan edge of the third viewpoint overlaps with the corresponding scene segmentation edge;

[0155] In the embodiments of the present application, step D1 can be to uniformly sample the edge set array. By using a specific length as the segmentation standard, the edge set array is segmented to obtain a segmented edge set. This segmented edge set includes several scene segmentation edges, and the segmentation length of each scene segmentation edge is the same.

[0156] It can be understood that for a certain third viewpoint in the third viewpoint network, step D2 can extract overlapping information for each scene segmentation edge in the segmented edge set based on the viewpoint scanning edge corresponding to the third viewpoint. Specifically, it can be determined whether the viewpoint scanning edge overlaps with each scene segmentation edge, so as to obtain a number of first scene segmentation edges and a number of second scene segmentation edges. The first scene segmentation edge is the scene segmentation edge that overlaps with the viewpoint scanning edge, and the second scene segmentation edge is the scene segmentation edge that does not overlap with the viewpoint scanning edge; then, by generating corresponding first scanning array elements for each first scene segmentation edge respectively, and generating corresponding second scanning array elements for each second scene segmentation edge, the scanning information array corresponding to the third viewpoint can be obtained by integrating all the first scanning array elements and all the second scanning array elements. The same applies to the other third viewpoints. After obtaining the scanning information arrays of all the third viewpoints, the scanning information table can be obtained by integrating all the scanning information arrays.

[0157] It should be noted that the specific representation method of the scanning array elements in the embodiments of the present application is not specifically limited. For example, the first scanning array element can be the element "1", while the second scanning array element can be the element "0"; or, the first scanning array element can be the element "true", while the second scanning array element can be the element "false". The examples in the present application are only for illustration.

[0158] D3. Perform heuristic optimization on the scanning information table to obtain the optimized viewpoint network

[0159] Further, the step D3 of performing heuristic optimization on the scanning information table to obtain the optimized viewpoint network includes:

[0160] D31. Obtain a preset second overlap degree threshold and edge number threshold;

[0161] D32. Use the current scanning information table as the first information table, and obtain a preferred viewpoint set corresponding to the first information table;

[0162] D33. Perform preferred viewpoint information screening on the first information table to obtain a preferred viewpoint array and the preferred viewpoints corresponding to the preferred viewpoint array. The preferred viewpoint array is the scanning information array with the largest number of overlapping relationship quantities in the first information table;

[0163] D34. Update the row and column information of the first information table according to the preferred viewpoint array to obtain a second information table and several information viewpoints corresponding to the second information table;

[0164] D35. Screen the information viewpoints according to the second overlap threshold and the preferred viewpoint to obtain a number of candidate viewpoints. Each candidate viewpoint is an information viewpoint with a second viewpoint overlap greater than or equal to the second overlap threshold, where the second viewpoint overlap is the viewpoint overlap between the information viewpoint and the preferred viewpoint;

[0165] D36. Compare the number of edges of the scan information arrays of all the candidate viewpoints according to the edge number threshold to obtain an edge number comparison result, which is used to indicate whether there is at least one candidate viewpoint with a total supplementary edge number greater than the edge number threshold. The total supplementary edge number is the total number of edges of the scene segmentation edges that overlap with the viewpoint scan edges of the candidate viewpoint;

[0166] D37. If the edge number comparison result indicates that there is no at least one candidate viewpoint with a total supplementary edge number greater than the edge number threshold, update the preferred viewpoint set according to the preferred viewpoint to obtain an updated preferred viewpoint set, and obtain the optimized viewpoint network according to the updated preferred viewpoint set;

[0167] Alternatively, D38. If the edge number comparison result indicates that there is no at least one candidate viewpoint with a total supplementary edge number greater than the edge number threshold, update the preferred viewpoint set according to the preferred viewpoint and the candidate viewpoint with the largest total supplementary edge number, and update the current scan information table according to the candidate viewpoint with the largest total supplementary edge number and the second information table. Then, return to execute the step of using the current scan information table as the first information table and obtaining the preferred viewpoint set corresponding to the first information table.

[0168] In the embodiments of the present application, for a certain optimization process in heuristic optimization, first, the current scan information table can be used as the first information table, and the preferred viewpoint set corresponding to the first information table can be obtained. Specifically, if this optimization process is the first optimization process, the current scan information table can be the scan information table obtained based on step D2, and the preferred viewpoint set is an empty viewpoint set; or, if this optimization process is the second or more optimization processes, the current scan information table can be the updated scan information table obtained in step D38 in the previous optimization process, and the preferred viewpoint set can be the updated preferred viewpoint set obtained in step D38 in the previous optimization process.

[0169] It can be understood that the preferred viewpoint information screening in step D33 can be to perform statistical screening on the scan array elements in each scan information array. Specifically, it can be to count the total number of scan array elements (i.e., the aforementioned first scan array elements) representing the overlapping relationship in each scan information array, and then determine the scan information array with the largest total number of scan array elements representing the overlapping relationship as the preferred viewpoint array, and the preferred viewpoint is the third viewpoint corresponding to the preferred viewpoint array.

[0170] It should be noted that the row and column information update in step D34 can first be to delete the preferred viewpoint array in the first information table, that is, to delete the row where the preferred viewpoint array is located in the first information table; at the same time, obtain the columns where all the scan array elements representing the overlapping relationship in the preferred viewpoint array are located, and then delete the columns where the corresponding scan array elements are located in the first information table, so as to determine the remaining scan information arrays as the second information table, and determine the third viewpoint corresponding to each scan information array as the information viewpoint.

[0171] It is worth mentioning that the overlap degree screening in step D35 can first be to calculate the second viewpoint overlap degree between the preferred viewpoint and each information viewpoint respectively based on the preferred viewpoint, so as to obtain the second viewpoint overlap degree corresponding to each information viewpoint; then, compare the size relationship between the second overlap degree threshold and each second viewpoint overlap degree, and determine the information viewpoints with the second viewpoint overlap degree greater than or equal to the second overlap degree threshold as the candidate viewpoints. Step D36 can be to compare the number of edges with the scan information array corresponding to each candidate viewpoint based on the edge number threshold, so as to obtain the edge number comparison result.

[0172] For the scan information array of a certain candidate viewpoint, the total supplementary edge number of the scan information array can be the total number of scan array elements representing the overlapping relationship in the scan information array. Specifically, if the edge number comparison result is that there is no total supplementary edge number of at least one candidate viewpoint greater than the edge number threshold, then the preferred viewpoint can be added to the preferred viewpoint set to obtain the updated preferred viewpoint set, and the preferred viewpoint set can be used as the optimized viewpoint network with the minimized number of viewpoints, sufficient viewpoint overlap degree and strong network observation integrity; or, if the edge number comparison result is that there is at least one candidate viewpoint with a total supplementary edge number greater than the edge number threshold, then the preferred viewpoint and several candidate viewpoints with the largest total supplementary edge number among all candidate viewpoints can be added to the preferred viewpoint set, and the current scan information table can be updated according to the second information table and the candidate viewpoints with the total step edge number.

[0173] It should be added that the preset edge number threshold can be set according to the actual situation, such as 0, 1, etc.; also, for the update of the scan information table, it can first be based on the column where the scan array element representing the overlap relationship corresponding to the candidate view point with the largest total supplementary edge number is located, delete the column where the corresponding scan array element in the second information table is located, so as to obtain the updated second information table, and then based on the updated second information table, replace and update the current scan information table.

[0174] It should be noted that since the total supplementary edge numbers of different candidate view points may be the same, there may be multiple candidate view points with the largest total supplementary edge number. For several candidate view points with the largest total supplementary edge number, in the first feasible implementation manner, these candidate view points with the largest total supplementary edge number can be added to the preferred view point set.

[0175] Alternatively, in the second feasible implementation manner, first, the preferred view points can be added to the preferred view point set to obtain an intermediate preferred view point set, then the view point set overlap degree between each candidate view point with the largest total supplementary edge number and the intermediate preferred view point set is obtained, and then the candidate view point corresponding to the largest view point set overlap degree is added to the intermediate preferred view point set, so as to complete the update of the preferred view point set. For a certain candidate view point with the largest total supplementary edge number, the view point set overlap degree can first be to obtain the intermediate view point overlap degree between the candidate view point and each preferred view point in the intermediate preferred view point set, and then the largest intermediate view point overlap degree among all the intermediate view point overlap degrees is determined as the view point set overlap degree corresponding to the intermediate preferred view point set. The examples of this application are only for illustration.

[0176] Next, a view point network optimization system proposed according to an embodiment of the present application will be described in detail with reference to the accompanying drawings.

[0177] Referring to Figure 4 , a view point network optimization system proposed in an embodiment of the present application includes:

[0178] A first processing unit 101, configured to obtain an edge set array of a quantization scan region and a scene target, and construct a view point network for the quantization scan region to obtain a first view point network;

[0179] A second processing unit 102, configured to perform connected view point filling on the first view point network to obtain a second view point network, where the second view point network includes several second view points and several first view points originally in the first view point network, and the newly filled target view points in the second view point network have a direct connection relationship with at least two non-target view points in the second view point network, the target view point is any one of the second view points, and the non-target view point is the first view point or a second view point other than the target view point;

[0180] A third processing unit 103, configured to fill in overlapping connection points for the second viewpoint network according to a preset first overlap threshold to obtain a third viewpoint network;

[0181] A fourth processing unit 104, configured to filter redundant viewpoints from the third viewpoint network according to the edge set array to obtain an optimized viewpoint network corresponding to the scene target.

[0182] It can be understood that the content in the above method embodiments is applicable to the system embodiments. The functions specifically implemented by the system embodiments are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those in the above method embodiments.

[0183] Referring Figure 5 , an electronic device is further provided in an embodiment of the present application, including:

[0184] At least one processor 201;

[0185] At least one memory 202, configured to store at least one program;

[0186] When the at least one program is executed by the at least one processor 201, the at least one processor 201 implements the above method embodiment.

[0187] Similarly, it can be understood that the content in the above method embodiments is applicable to the device embodiments. The functions specifically implemented by the device embodiments are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those in the above method embodiments.

[0188] An embodiment of the present application further provides a computer-readable storage medium, in which a program executable by a processor 201 is stored, and the program executable by the processor 201 is used to implement the above method embodiment when executed by the processor 201.

[0189] Similarly, the content in the above method embodiments is applicable to the computer-readable storage medium embodiments. The functions specifically implemented by the computer-readable storage medium embodiments are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those in the above method embodiments.

[0190] In some alternative embodiments, the functions / operations recited in the block diagrams may not occur in the order presented in the operational illustrations. For example, depending on the functions / operations involved, two blocks shown in succession may actually be executed substantially simultaneously or the blocks can sometimes be executed in reverse order. Further, the embodiments presented and described in the flowcharts of the present application are provided by way of example in order to provide a more thorough understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed and in which sub-operations described as part of a larger operation are performed independently.

[0191] In addition, although the present application has been described in the context of functional modules, it should be understood that, unless otherwise stated to the contrary, one or more of the functions and / or features may be integrated in a single physical device and / or software module, or one or more functions and / or features may be implemented in separate physical devices or software modules. It should also be understood that a detailed discussion of the actual implementation of each module is not necessary for an understanding of the present application. Rather, given the attributes, functions, and internal relationships of the various functional modules in the devices disclosed herein, the actual implementation of the modules will be understood within the ordinary skill of an engineer. Thus, those of ordinary skill in the art will be able to implement the present application as set forth in the claims without undue experimentation. It should also be understood that the particular concepts disclosed are merely illustrative and are not intended to limit the scope of the present application, the scope of which is determined by the full scope of the appended claims and their equivalents.

[0192] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such understanding, the technical solution of the present application, in essence or the part that contributes to the prior art or part of the technical solution, can be embodied in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method of the embodiments of the present application. The foregoing storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc.

[0193] The logic and / or steps represented in the flowchart or otherwise described herein can, for example, be considered as a definitional sequence of executable instructions for implementing logical functions, which can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device. As used in this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in connection with the instruction execution system, apparatus, or device.

[0194] More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection portion having one or more wirings (electronic device), a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which a program can be printed, as the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or otherwise processing as appropriate, and then storing it in a computer memory.

[0195] It should be understood that various parts of the present application can be implemented by hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or a combination of the following techniques known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application-specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), and the like.

[0196] In the foregoing description of this specification, the descriptions with reference to the terms "one embodiment / example", "another embodiment / example", or "certain embodiments / examples", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0197] Although embodiments of the present application have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present application. The scope of the present application is defined by the claims and their equivalents.

[0198] The above is a specific description of the preferred embodiments of the present application, but the present application is not limited to the embodiments. Those skilled in the art can make various equivalent deformations or substitutions without violating the spirit of the present application, and these equivalent deformations or substitutions are all included within the scope defined by the claims of the present application.

Claims

1. A method for optimizing a viewpoint network, characterized in that: include: Acquire an edge set array of a quantized scan area and a scene target, and construct a viewpoint network for the quantized scan area to obtain a first viewpoint network; The first viewpoint network is supplemented with connected viewpoints to obtain a second viewpoint network, wherein the second viewpoint network includes a plurality of second viewpoints and a plurality of first viewpoints originally in the first viewpoint network, the newly supplemented target viewpoint in the second viewpoint network has a direct connection relationship with at least two non-target viewpoints in the second viewpoint network, the target viewpoint is any one of the second viewpoints, and the non-target viewpoint is the first viewpoint or a second viewpoint other than the target viewpoint; According to a preset first overlap threshold, filling the second viewpoint network with overlapping connection points to obtain a third viewpoint network; According to the edge set array, redundant viewpoints are filtered on the third viewpoint network to obtain an optimized viewpoint network corresponding to the scene target.

2. The method according to claim 1, characterized in that: The constructing of a viewpoint network for the quantized scan area to obtain a first viewpoint network includes: Obtaining a preset viewpoint adjacent pixel threshold; Performing regional imaging on the quantified scanning area to obtain a first regional image; Performing boundary transformation on the first region image to obtain a second region image; Performing boundary inversion on the second region image to obtain a third region image; Performing ridge line extraction and skeleton refinement on the third region image to obtain a refined skeleton line image; According to the viewpoint adjacent pixel threshold, viewpoint extraction is performed on the skeleton line intersection points in the refined skeleton line image to obtain the first viewpoint network.

3. The method according to claim 1, characterized in that The method of performing connected viewpoint addition on the first viewpoint network to obtain a second viewpoint network includes: Taking the current first viewpoint network as the fourth viewpoint network, and obtaining a first connected path of each fourth viewpoint pair in the fourth viewpoint network; According to all the first connected paths, the fourth viewpoint network is supplemented with the farthest point to obtain a fifth viewpoint network and a farthest point newly added identifier of the fifth viewpoint network, wherein the farthest point newly added identifier is used to indicate whether the fifth viewpoint network is the fourth viewpoint network after the farthest point is added; If the fifth viewpoint network newly identified as the farthest point is the fourth viewpoint network after adding the farthest point, the current first viewpoint network is updated according to the fifth viewpoint network, and then the step of returning to execute the current first viewpoint network as the fourth viewpoint network and obtaining the first connectivity path of each fourth viewpoint pair in the fourth viewpoint network is performed; or, if the fifth viewpoint network newly identified as the farthest point is not the fourth viewpoint network after adding the farthest point, the fifth viewpoint network is determined as the second viewpoint network.

4. The method according to claim 3, characterized in that The method of performing the farthest point filling on the fourth viewpoint network according to all the first connected paths to obtain the fifth viewpoint network and the farthest point newly added identifier of the fifth viewpoint network includes: According to all the first connected paths, the fourth viewpoint network is screened for viewpoint pairs to obtain a plurality of fifth viewpoint pairs and a second connected path corresponding to each of the fifth viewpoint pairs, wherein the fifth viewpoint pairs are fourth viewpoint pairs having a direct connected relationship; According to all the second connected paths, searching for the farthest point of the corresponding fifth viewpoint pair to obtain a plurality of target farthest points; According to all the target farthest points, the fourth viewpoint network is updated with the first viewpoint to obtain the fifth viewpoint network and the farthest point newly added identifier.

5. The method according to claim 4, characterized in that According to the second connected path, searching for the farthest point of the corresponding fifth viewpoint pair to obtain the farthest point of the target includes: Obtaining a distance threshold between a current fifth viewpoint pair and the current fifth viewpoint pair; Searching for the farthest point on the second connected path of the current fifth viewpoint pair to obtain the middle farthest point, wherein the middle farthest point is a path pixel point with the largest viewpoint distance among all path pixel points of the second connected path, and the viewpoint distance is the minimum interval distance between the path pixel point and the straight line where the current fifth viewpoint pair is located; Comparing the distance threshold with the viewpoint distance of the farthest intermediate point to obtain a distance comparison result; If the distance comparison result is that the viewpoint distance of the middle farthest point is greater than or equal to the distance threshold, the middle farthest point is determined as the target farthest point.

6. The method according to claim 1, characterized in that The method of filling the second viewpoint network with overlapping connection points according to a preset first overlap threshold to obtain a third viewpoint network includes: Taking the current second viewpoint network as the sixth viewpoint network, and obtaining the first viewpoint overlap of each sixth viewpoint pair in the sixth viewpoint network; According to the first overlap threshold and all the first viewpoint overlaps, the sixth viewpoint network is supplemented with intermediate points to obtain a seventh viewpoint network and an additional intermediate point identifier of the seventh viewpoint network, wherein the additional intermediate point identifier is used to indicate whether the seventh viewpoint network is the sixth viewpoint network after the intermediate point is added; If the newly added mark of the intermediate point is that the seventh viewpoint network is the sixth viewpoint network after adding the intermediate point, the current second viewpoint network is updated according to the seventh viewpoint network, and then the step of returning to execute the current second viewpoint network as the sixth viewpoint network and obtaining the first viewpoint overlap of each sixth viewpoint pair in the sixth viewpoint network is returned; or, if the newly added mark of the intermediate point is that the seventh viewpoint network is not the sixth viewpoint network after adding the intermediate point, the seventh viewpoint network is determined as the third viewpoint network.

7. The method according to claim 6, characterized in that The method of filling the sixth viewpoint network with intermediate points according to the first overlap threshold and all the first viewpoint overlaps to obtain the seventh viewpoint network and the newly added identifier of the intermediate points of the seventh viewpoint network includes: performing overlap comparison on all the first viewpoint overlaps according to the first overlap threshold, to obtain an overlap comparison result corresponding to each first viewpoint overlap; According to all the overlap comparison results, screen the viewpoint pairs of the sixth viewpoint network to obtain a plurality of seventh viewpoint pairs, wherein the seventh viewpoint pairs are the sixth viewpoint pairs whose corresponding first viewpoint overlap is less than the first overlap threshold; Selecting middle points for all the seventh viewpoint pairs to obtain a target middle point for each of the seventh viewpoint pairs; According to all the target intermediate points, the sixth viewpoint network is updated with a second viewpoint to obtain the seventh viewpoint network and the newly added identifier of the intermediate point.

8. The method according to claim 1, characterized in that The step of filtering redundant viewpoints on the third viewpoint network according to the edge set array to obtain an optimized viewpoint network corresponding to the scene target includes: Uniformly sampling the edge set array to obtain a segmented edge set, wherein the segmented edge set includes a plurality of scene segmentation edges of the scene object; According to the segmented edge set, overlapping information is extracted from the third viewpoint network to obtain a scan information table, wherein the scan information table includes a plurality of scan information arrays, each of which corresponds to a third viewpoint in the third viewpoint network, and the scan information array includes a plurality of scan array elements, each of which is used to represent whether a viewpoint scan edge of the third viewpoint overlaps with a corresponding scene segmentation edge; The scanning information table is heuristically optimized to obtain the optimized viewpoint network.

9. The method according to claim 8, characterized in that The step of performing heuristic optimization on the scan information table to obtain the optimized viewpoint network includes: Obtaining a preset second overlap threshold and edge number threshold; Using the current scanning information table as the first information table, and acquiring a preferred viewpoint set corresponding to the first information table; Screening the first information table for preferred viewpoint information to obtain a preferred viewpoint array and preferred viewpoints corresponding to the preferred viewpoint array, wherein the preferred viewpoint array is a scanning information array with the largest number of overlapping relationships in the first information table; According to the preferred viewpoint array, updating the row and column information of the first information table to obtain a second information table and a plurality of information viewpoints corresponding to the second information table; According to the second overlap threshold and the preferred viewpoint, all the information viewpoints are screened for overlap to obtain a plurality of candidate viewpoints, each of the candidate viewpoints being an information viewpoint whose second viewpoint overlap is greater than or equal to the second overlap threshold, and the second viewpoint overlap is the viewpoint overlap between the information viewpoint and the preferred viewpoint; According to the edge number threshold, performing edge number comparison on the scan information arrays of all the candidate viewpoints to obtain an edge number comparison result, wherein the edge number comparison result is used to indicate whether there is a total supplementary edge number of at least one candidate viewpoint that is greater than the edge number threshold, and the total supplementary edge number is the total number of scene segmentation edges that overlap with the viewpoint scan edge of the candidate viewpoint; If the result of the edge number comparison is that there is no at least one candidate viewpoint whose total number of supplementary edges is greater than the edge number threshold, then according to the preferred viewpoint, the preferred viewpoint set is updated to obtain an updated preferred viewpoint set, and the optimized viewpoint network is obtained according to the updated preferred viewpoint set; or, if the result of the edge number comparison is that there is no at least one candidate viewpoint whose total number of supplementary edges is greater than the edge number threshold, then according to the preferred viewpoint and the candidate viewpoint with the largest total number of supplementary edges, the preferred viewpoint set is updated, and according to the candidate viewpoint with the largest total number of supplementary edges and the second information table, the current scanning information table is updated, and then the step of returning to execute the current scanning information table as the first information table and obtaining the preferred viewpoint set corresponding to the first information table.

10. A viewpoint network optimization system, characterized in that: include: A first processing unit is used to obtain an edge set array of a quantized scan area and a scene target, and to construct a viewpoint network for the quantized scan area to obtain a first viewpoint network; A second processing unit is used to perform connected viewpoint filling on the first viewpoint network to obtain a second viewpoint network, wherein the second viewpoint network includes a plurality of second viewpoints and a plurality of first viewpoints originally in the first viewpoint network, the newly added target viewpoint in the second viewpoint network has a direct connected relationship with at least two non-target viewpoints in the second viewpoint network, the target viewpoint is any one of the second viewpoints, and the non-target viewpoint is the first viewpoint or a second viewpoint other than the target viewpoint; A third processing unit, configured to fill the second viewpoint network with overlapping connection points according to a preset first overlap threshold, so as to obtain a third viewpoint network; The fourth processing unit is used to filter redundant viewpoints of the third viewpoint network according to the edge set array to obtain an optimized viewpoint network corresponding to the scene target.

Images