Non-line-of-sight building layout repositioning method and device in array interferometric synthetic aperture radar three-dimensional imaging, storage medium and program product

Abstract

The invention discloses a non-line-of-sight building layout repositioning method and device in array interferometric synthetic aperture radar three-dimensional imaging, a storage medium and a program product, is applied to the technical field of array interferometric synthetic aperture radar three-dimensional imaging, and aims to solve the problem that non-line-of-sight building layout repositioning cannot be realized when building layout parameters are unknown in the prior art. The method comprises the following steps: firstly, acquiring an initial three-dimensional point cloud of an observation scene by using an array interferometric synthetic aperture radar system; then performing curved surface reconstruction on the point cloud by applying an Alpha Shape algorithm, and constructing a boundary model for reflection path analysis; identifying a non-line-of-sight point and an associated reflection point thereof by analyzing whether a connecting line of the point cloud and the radar antenna intersects with the curved surface model; and finally, correcting and repositioning the non-line-of-sight building layout by adopting a density clustering and plane fitting algorithm and combining a mirror symmetry geometrical relationship. According to the method, prior structure information is not needed, the multi-path reflection path can be effectively identified, the real space position of the building in the non-line-of-sight area can be recovered, the three-dimensional imaging precision is remarkably improved, and powerful support is provided for SAR application such as urban perception and structure reconstruction.

Description

Technical Field

[0001] The present invention belongs to the technical field of array interferometric synthetic aperture radar three-dimensional imaging, and in particular relates to a non-line-of-sight building layout imaging technology. Background Art

[0002] Array interferometric synthetic aperture radar (ISAR) 3D imaging technology primarily uses radar to observe urban targets from different angles. Combined with multi-baseline interferometric synthesis processing, it achieves high-precision 3D reconstruction of buildings and terrain in complex urban environments. This technology has important applications in urban planning, building structure monitoring, and post-disaster assessment. When radar signals propagate in urban environments, they travel along multiple paths to the target and back to the receiver. The signal returning to the receiver after the radar directly illuminates the target is called the direct path signal. Due to the dense density of tall buildings and smooth walls in cities, radar electromagnetic waves can reflect one or more times off the building's exterior, generating multipath signals that reflect off the wall or ground before returning to the receiver. These signals may also reflect off other building surfaces after passing the target and then bounce back to the receiver, ultimately causing multipath ghosting in the image. Multipath ghosting can severely interfere with the quality of 3D point clouds and reduce the reliability of the imaging results. By properly identifying and correcting multipath ghosting, targets can be reconstructed even in occluded areas lacking a direct path, thereby enhancing the integrity and accuracy of urban 3D modeling. Therefore, the utilization of multipath signals in urban environment perception is crucial for 3D imaging.

[0003] In the field of three-dimensional imaging, research on multipath exploitation methods for urban environment perception is still relatively limited and in its infancy. The Institute of Space Information Innovation of the Chinese Academy of Sciences proposed a non-line-of-sight (NLOS) target imaging method based on multipath signal analysis (Y. Lin, X. Qiu, Y. Luo, et al., “Viability of multipath exploitation in urban canyon: Range profiles analysis in TomoSAR imaging,” IGARSS, 2024.). This method analyzes the propagation characteristics of multipath signals in NLOS areas and demonstrates the feasibility of using multipath signals for NLOS target imaging. By analyzing the number of reflections and paths of multipath signals, this method can detect and image NLOS targets hidden in urban canyons. Building on this foundation, the Institute of Space Information Innovation of the Chinese Academy of Sciences further proposed a multipath-based method for relocalizing non-line-of-sight (NLOS) targets in 3D synthetic aperture radar (SAR) imaging of built-up areas (Lin Yuqing, Qiu Xiaolan, Peng Lingxiao, et al. "Non-line-of-sight (NLOS) target relocalization method in 3D SAR imaging of built-up areas based on a multipath model" [J]. Journal of Radars, 2024). This method leverages the geometric relationships and reflection characteristics of multipath signals. By building a low-altitude spherical wave model and a multipath model, combined with building plane fitting and geometric analysis, it restores the virtual shadows of NLOS targets to their true positions, thereby achieving accurate 3D imaging and relocalization of NLOS targets in urban canyons. This method only works in scenarios where the building layout is known; it cannot achieve NLOS building layout relocalization when the building layout parameters are unknown. Therefore, studying NLOS building layout relocalization methods in array interferometric SAR 3D imaging is of great value in the field of urban environmental perception. Summary of the Invention

[0004] To solve the above technical problems, the present invention proposes a method, device, storage medium and program product for relocating the layout of non-line-of-sight buildings in array interferometric synthetic aperture radar three-dimensional imaging. Without the need for prior building structure information, it can realize the identification and correction of non-line-of-sight building shadows caused by multipath propagation in urban environments.

[0005] One of the technical solutions adopted by the present invention is: a method for relocating non-line-of-sight building layout in array interferometric synthetic aperture radar three-dimensional imaging, comprising:

[0006] S1. Observe the target area from multiple perspectives using an array interferometric synthetic aperture radar system. Using the tomographic synthetic aperture radar method, coherently combine multiple SAR images acquired from multiple observation angles, extract the vertical scattering distribution of each pixel, and generate an initial three-dimensional point cloud of the observation scene.

[0007] S2, randomly downsampling the initial 3D point cloud obtained in step S1, and then reconstructing the downsampling result using the Alpha Shape algorithm to generate a 3D surface model reflecting the outer contour of the building structure;

[0008] S3. Distinguish the propagation path type by analyzing whether the line connecting each point in the initial three-dimensional point cloud and the radar antenna intersects with the three-dimensional surface model generated in step S2, and obtain line-of-sight points, non-line-of-sight points, and reflection points associated with the non-line-of-sight points;

[0009] S4, correcting the non-line-of-sight point obtained in step S3;

[0010] S5. Superimpose the line-of-sight points obtained in step S3 and the corrected non-line-of-sight points obtained in step S4 to obtain a final three-dimensional point cloud.

[0011] In step S1, an array interferometric synthetic aperture radar system is used to perform multi-perspective observations of the target area. Using a tomographic synthetic aperture radar method, multiple SAR images acquired from different observation angles are coherently combined to extract the vertical scattering distribution of each pixel point, generating an initial three-dimensional point cloud of the observed scene. This three-dimensional point cloud reflects the approximate spatial distribution of scatterers and can describe the spatial characteristics of building structures within the line-of-sight area. However, due to the presence of multipath propagation effects, the layout of buildings in non-line-of-sight areas often appears in the point cloud as offset, distorted, or ghosted, failing to reflect their true spatial location. Therefore, geometric repositioning is required in conjunction with subsequent steps to accurately reconstruct non-line-of-sight building structures.

[0012] The sampling operation in step S2 can be described as:

[0013] P down ={p i |p i ∈P,ξ i <η}

[0014] Among them, P is the initial point cloud set, P down is the point cloud set after downsampling, p i is a point in P, ξ i is a uniform random variable in the interval [0,1], and η is the sampling rate.

[0015] The surface reconstruction in step S2 is as follows: the Alpha Shape algorithm is used to reconstruct the surface of the down-sampling result to generate a three-dimensional surface model reflecting the outer contour of the building structure, which is denoted as S r , providing spatial boundary information for subsequent reflection path analysis and non-line-of-sight target relocation.

[0016] Step S3 mainly performs non-line-of-sight building layout and associated reflection point identification. The specific process is as follows:

[0017] The 3D point cloud obtained in step 1 contains both line-of-sight and non-line-of-sight points. To identify non-line-of-sight building layouts, the point cloud must be classified and the reflection point associated with each non-line-of-sight point determined. Since multipath propagation primarily involves reflections on surfaces (this method only considers first-order reflections), the propagation path type can be distinguished by analyzing whether the line connecting each point to the radar antenna intersects the surface model.

[0018] Assume the position of the radar antenna is Represents a set of real numbers, where x, y, and z represent the spatial three-dimensional geometric coordinates of the antenna. Select any point P in the initial three-dimensional point cloud and consider the following two cases:

[0019] Case 1 (line of sight path): If line segment AP and surface S r If there is no intersection, then point P is the sight point.

[0020] Case 2 (non-line-of-sight path): If line segment AP and surface S r The existence of an intersection point R indicates that the radar signal reaches point P after being reflected from point P. Then point P is the line-of-sight point, and point R is its associated reflection point. The position of R can be determined by the following expression: R∈S r And A+t(P NLOS -A)=R, where t∈(0,1).

[0021] By traversing the entire point cloud set and performing the above judgment on each point, all non-line-of-sight points and the associated reflection point sets can be identified.

[0022] The non-line-of-sight building layout correction implementation process in step S4 is as follows:

[0023] S41 , clustering and grouping the reflection points identified in step S3 using a density clustering algorithm (such as DBSCAN) to form a number of reflection point clusters.

[0024] The mathematical expression of the DBSCAN algorithm is as follows:

[0025] set up is a set of reflection points, defining points The neighborhood of is:

[0026]

[0027] Where d(·,·) is the Euclidean distance and ∈ is the minimum neighborhood radius. The conditions for being a core point are:

[0028]

[0029] Here, |·| represents the cardinality of the set, and MinPts is the preset minimum neighborhood point threshold.

[0030] S42. Apply the RANSAC algorithm to each cluster to perform plane fitting and extract the corresponding reflection plane. The plane equation obtained by fitting can be expressed as:

[0031] π k :a k x+b k y+c k z+d k =0

[0032] Among them, (a k ,b k ,c k ) is the plane normal vector And the normalization conditions are met:

[0033]

[0034] ‖·‖2 represents the L2 norm;

[0035] S43, using the fitting plane as the mirror symmetry axis, project the corresponding non-line-of-sight point to the mirror position. and the reflection plane π k , its mirror point The calculation can be expressed as follows:

[0036]

[0037] Where D is To plane π k The directed distance of:

[0038]

[0039] By performing the above mirror transformation on all non-line-of-sight points, the spatial correction and repositioning of the non-line-of-sight building layout can be achieved, thereby recovering its true geometric position without relying on prior structural information.

[0040] The second technical solution adopted by the present invention is: a computer device including a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of a non-line-of-sight building layout repositioning method in array interferometric synthetic aperture radar three-dimensional imaging.

[0041] The third technical solution adopted by the present invention is: a computer scale storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of a non-line-of-sight building layout repositioning method in array interferometric synthetic aperture radar three-dimensional imaging are implemented.

[0042] The third technical solution adopted by the present invention is: a computer program product, including a computer program, which, when executed by a processor, implements the steps of a non-line-of-sight building layout repositioning method in array interferometric synthetic aperture radar three-dimensional imaging.

[0043] Beneficial effects of the present invention: The present invention proposes a method for relocating the layout of non-line-of-sight buildings in array interferometric synthetic aperture radar three-dimensional imaging. First, an array interferometric synthetic aperture radar system is used to obtain an initial three-dimensional point cloud of the observation scene; then, an Alpha Shape algorithm is applied to reconstruct the point cloud to construct a boundary model for reflection path analysis; then, by analyzing whether the line connecting the point cloud and the radar antenna intersects with the surface model, non-line-of-sight points and their associated reflection points are identified; finally, a density clustering and plane fitting algorithm is used, combined with a mirror-symmetric geometric relationship, to correct and relocate the layout of non-line-of-sight buildings.

[0044] The method of the present invention does not require prior structural information, can effectively identify multipath reflection paths, restore the real spatial position of buildings in non-line-of-sight areas, significantly improve the accuracy of three-dimensional imaging, and provide strong support for SAR applications such as urban perception and structural reconstruction. This method can accurately identify and correct non-line-of-sight building shadows generated by multipath propagation without any prior building structure information. By constructing a three-dimensional point cloud and combining it with reflection point analysis, this method can effectively distinguish between line-of-sight points and non-line-of-sight points, and use mirror-symmetric geometric relationships to relocate the offset non-line-of-sight building structures to their true positions. Compared with existing multipath correction methods that rely on prior geometric models, the present invention has stronger adaptability and can obtain high-precision and high-reliability three-dimensional reconstruction results in complex urban environments, providing strong support for SAR systems in practical applications such as urban perception and building monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 Processing flow chart for non-line-of-sight building layout relocation method.

[0046] Figure 2 Schematic diagram of the simulation scene.

[0047] Figure 3 Schematic diagram of antenna location distribution.

[0048] Figure 4 is the side view of the initial 3D point cloud.

[0049] Figure 5 It is the top view of the initial 3D point cloud.

[0050] Figure 6 It is a three-dimensional surface model.

[0051] Figure 7 is a non-line-of-sight point set.

[0052] Figure 8 is the reflection point set.

[0053] Figure 9 is the reflection surface fitting result.

[0054] Figure 10 It is the 3D point cloud after relocation. DETAILED DESCRIPTION

[0055] To facilitate those skilled in the art to understand the technical content of the present invention, the present invention is further explained below with reference to the accompanying drawings.

[0056] This embodiment is as follows Figure 2 Taking the dimensions and CAD model of the L-shaped building shown as an example, the drone-mounted radar track's starting coordinates are (-28m, -300m, 300m), and the track's ending coordinates are (88m, -300m, 300m). The azimuth sampling interval is 0.08m, and the radar's angle of incidence is 45°. The radar's transmitted signal is a stepped frequency signal with a center frequency of 16GHz and a bandwidth of 1600MHz, with a frequency step of 0.8MHz. The radar array mode is 1 transmit, 16 receive, and the antenna positions are as follows: Figure 3 In this simulation scenario, the electromagnetic scattering modeling method based on SBR-PO is used to generate radar echo data.

[0057] like Figure 1 As shown, the method of the present invention comprises the following processing steps:

[0058] Step 1: Acquisition of 3D point cloud

[0059] The 1-transmit 16-receive mode is equivalent to 16 antenna phase centers, which can obtain 16 single-view SAR images. After completing the image registration, de-skewing and other pre-processing, the orthogonal matching pursuit algorithm is used to perform sparse reconstruction in the height direction to obtain the initial 3D point cloud. The reconstruction result is as follows: Figure 4 and Figure 5 shown. Figure 5 In the figure, the area enclosed by the dotted circle behind the L-shaped building is the NLOSsurface of the building, i.e., the non-line-of-sight surface. Its position needs to be corrected to achieve non-line-of-sight building layout repositioning.

[0060] Step 2: Surface reconstruction

[0061] In order to reduce the computational complexity of subsequent steps, the initial three-dimensional point cloud obtained in step 1 is randomly downsampled according to a set ratio. The selection of the sampling ratio needs to strike a balance between ensuring the density of the point cloud and the efficiency of the algorithm. If the sampling ratio is too large, the number of triangular facets will increase and the computational burden will be increased. If the ratio is too small, the point cloud will be sparse, affecting the quality of surface reconstruction. In this embodiment, the sampling ratio is set to 0.1. Subsequently, the Alpha Shape algorithm is used to reconstruct the surface of the downsampled point cloud. The Alpha sphere radius is set to 3, and the hole filling threshold is set to 1. The surface reconstruction result is as follows: Figure 6 shown.

[0062] Step 3: Non-line-of-sight building layout and associated reflection point identification

[0063] Traverse the point cloud dataset and determine the propagation path type based on whether the line between each point and the radar antenna intersects the surface model. If there is no intersection between the line and the surface, the point is a line-of-sight point; if there is an intersection, it is a non-line-of-sight point, and the intersection is the associated reflection point of the point. The final non-line-of-sight point set and reflection point set are as follows: Figure 7 and Figure 8 shown.

[0064] Step 4: Non-line-of-sight building layout correction

[0065] The reflection points identified in step 3 are clustered and outliers are removed using the Density-Based Spatial Clustering of Applications with Noise (DBSCAN). The neighborhood radius ∈ is set to 5 and the minimum number of neighborhood points MinPts is set to 10. Here, only one reflection point cluster is obtained. Subsequently, the Random Sample Consensus (RANSAC) algorithm is applied to the cluster for plane fitting. The fitting results are shown in the figure below. Figure 9 As shown, the unnormalized reflection plane equation is -0.0216x-16.3012y+581.2563=0. Finally, the non-line-of-sight point set is mirror-symmetrical about the reflection plane to achieve the relocation of the non-line-of-sight building layout. The relocation result is shown in Figure 10 As shown in the figure, the repositioning error is controlled within 0.5m.

[0066] The simulation results show that the non-line-of-sight building layout relocation method for array interferometric synthetic aperture radar three-dimensional imaging provided by the present invention can not only effectively utilize multipath, but also does not rely on prior layout information, verifying the correctness and effectiveness of the present invention.

[0067] This embodiment also provides a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of a method for relocating non-line-of-sight building layout in array interferometric synthetic aperture radar three-dimensional imaging.

[0068] This embodiment also provides a computer scale storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of a method for relocating non-line-of-sight building layout in array interferometric synthetic aperture radar three-dimensional imaging are implemented.

[0069] This embodiment also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of a method for relocating non-line-of-sight building layout in array interferometric synthetic aperture radar three-dimensional imaging.

[0070] Those skilled in the art will appreciate that the embodiments described herein are intended to aid the reader in understanding the principles of the present invention, and it should be understood that the scope of the present invention is not limited to such specific descriptions and embodiments. Various modifications and variations are readily apparent to those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are intended to be included within the scope of the claims.

Claims

1. A method for relocating non-line-of-sight building layout in array interferometric synthetic aperture radar three-dimensional imaging, characterized in that: include: S1. Observe the target area from multiple perspectives using an array interferometric synthetic aperture radar system. Using the tomographic synthetic aperture radar method, coherently combine multiple SAR images acquired from multiple observation angles, extract the vertical scattering distribution of each pixel, and generate an initial three-dimensional point cloud of the observation scene. S2, randomly downsampling the initial 3D point cloud obtained in step S1, and then reconstructing the downsampling result using the Alpha Shape algorithm to generate a 3D surface model reflecting the outer contour of the building structure; S3. Distinguish the propagation path type by analyzing whether the line connecting each point in the initial three-dimensional point cloud and the radar antenna intersects with the three-dimensional surface model generated in step S2, and obtain line-of-sight points, non-line-of-sight points, and reflection points associated with the non-line-of-sight points; S4. Correcting the non-line-of-sight point based on the reflection point associated with the non-line-of-sight point; S5. Superimpose the line-of-sight points obtained in step S3 and the corrected non-line-of-sight points obtained in step S4 to obtain a final three-dimensional point cloud.

2. The method for relocating non-line-of-sight building layout in array interferometric synthetic aperture radar three-dimensional imaging according to claim 1, characterized in that: The specific process of step S3 is: Assume the position of the radar antenna is A; select any point P in the initial 3D point cloud and consider the following two cases: Case 1: If line segment AP and surface S r If there is no intersection, then point P is the sight point; S r is the three-dimensional surface model obtained in step S2; Case 2: If line segment AP and surface S r The existence of intersection point R indicates that the radar signal reaches point P after being reflected from point P. Point P is a non-line-of-sight point, and point R is its associated reflection point. The position of R is determined by the following expression: R ∈ S r and A + t(P NLOS - A) = R where t∈(0,1); By traversing the entire initial three-dimensional point cloud set, all line-of-sight point sets, non-line-of-sight point sets, and reflection point sets associated with the non-line-of-sight points are identified.

3. The method for relocating non-line-of-sight building layout in array interferometric synthetic aperture radar three-dimensional imaging according to claim 2, characterized in that: Step S4 specifically includes the following sub-steps: S41, clustering and grouping the reflection points obtained in step S3 using a clustering algorithm to form a number of reflection point clusters; S42, performing plane fitting on each cluster to extract the corresponding reflection plane; S43. Using the fitting plane as the mirror symmetry axis, project the corresponding non-line-of-sight point to the mirror position, thereby obtaining a corrected non-line-of-sight point.

4. The method for relocating non-line-of-sight building layout in array interferometric synthetic aperture radar three-dimensional imaging according to claim 3, characterized in that: Step S41 specifically adopts a density clustering algorithm.

5. The method for relocating non-line-of-sight building layout in array interferometric synthetic aperture radar three-dimensional imaging according to claim 4, characterized in that: Step S42 specifically uses the RANSAC algorithm to perform plane fitting.

6. A computer device comprising a memory, a processor, and a computer program stored in the memory, wherein: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 5.

7. A computer scale storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

8. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images