Method and device for simulating radar detection effect based on GPU in simulation system

Abstract

The application discloses a method and device for simulating radar detection effect based on GPU in a simulation system, comprising: based on an image collector, collecting a scene image of a scene to be detected after the scene is irradiated by a point light source; transmitting the scene image to a sub-graphic processor matched with the graphic processor of the image collector, so that the sub-graphic processor determines an actual illumination area in the scene image which is not blocked by a visible obstacle in the scene to be detected as a radar detection result. It can be seen that the application can simulate a scene in which a laser radar ray is blocked by an obstacle by using the characteristic that the light source stops illuminating at the blocked place, and convert a complex radar signal originally processed by a CPU into a simple image signal processed by a GPU only, so that the consumption of CPU computing power in radar detection is reduced, and the overall performance of radar detection is improved.

Description

Technical Field

[0001] This invention relates to the field of simulation technology, and in particular to a method and apparatus for simulating radar detection effects based on a GPU in a simulation system. Background Technology

[0002] In the fields of robot ranging, navigation, and mapping, radar detection is an indispensable technology. Current mainstream simulation software typically uses ray tracing to simulate the high-frequency emission of radar to explore surrounding visible obstacles, thus achieving map scanning. After scanning, the results need to be represented based on the detection points (i.e., the endpoints of the ray tracings) as the radar detection results. Currently, there are two main ways to represent the detection results: the first method involves drawing multiple dense ray tracings based on the detection points to represent the radar detection results (e.g.,...). Figure 1 As shown in the figure, however, drawing rays requires real-time calculation and adjustment of ray length, which not only consumes a lot of CPU computing power and reduces the overall performance of radar detection, but also leads to cluttered and unintuitive scanning results; the second method connects the detection points and draws them into a polygonal patch to represent the radar detection results (as shown in the figure). Figure 2 As shown in the figure, however, drawing polygonal patches requires real-time mesh generation, which involves a large amount of computation and consumes significant CPU resources. Therefore, reducing CPU resource consumption in radar detection to improve overall radar detection performance is crucial. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method and apparatus for simulating radar detection effect based on GPU in a simulation system, which can reduce the CPU computing power consumption in radar detection and improve the overall performance of radar detection.

[0004] To address the aforementioned technical problems, the first aspect of this invention discloses a method for simulating radar detection effects in a simulation system based on a GPU, the method comprising:

[0005] Based on an image acquisition device pre-configured for the scene to be detected, a scene image of the scene to be detected after being illuminated by a point light source is acquired, wherein the point light source is placed in the scene to be detected;

[0006] The scene image is transmitted to a sub-graphics processor that matches the image acquisition unit in the graphics processor, so that the sub-graphics processor determines the actual illuminated area in the scene image that is not obscured by visible obstacles in the scene to be detected, and uses it as the radar detection result corresponding to the scene to be detected.

[0007] As an optional implementation, in the first aspect of the present invention, the acquisition of a scene image of the scene to be detected after being illuminated by a point light source, based on an image acquisition device pre-configured for the scene to be detected, includes:

[0008] Based on the image acquisition device pre-configured for the scene to be detected, the orthogonal projection of the illumination plane where the point light source is located onto the target plane after the scene to be detected is illuminated by the point light source is acquired, which is used as the scene image of the scene to be detected after being illuminated by the point light source.

[0009] The target plane includes the simulated imaging plane corresponding to the image acquisition device, the illumination plane is parallel to the target plane, the scene image includes the cross-sectional projection of each visible obstacle in the scene to be detected, the visible obstacle is an obstacle located within the acquisition field of view of the image acquisition device, and the cross-sectional projection of each visible obstacle includes the projection of the cross-section of the visible obstacle after it has been cut by the illumination plane.

[0010] As an optional implementation, in a first aspect of the invention, the sub-graphics processor determines the actual illuminated area in the scene image that is not obscured by visible obstacles in the scene to be detected, as the radar detection result corresponding to the scene to be detected, including:

[0011] The sub-graphics processor determines the illumination range of the point light source in the scene image based on the position of the point light source in the scene to be detected and the light parameters of the point light source;

[0012] The sub-graphics processor determines the area outside the cross-sectional projections corresponding to all the visible obstacles within the illumination range as the initial illumination area in the scene image;

[0013] The sub-graphics processor corrects the initial illumination area based on a pre-determined illumination angle calculation algorithm, the position of the projection point of the point light source in the scene image, and the position of the cross-section projection of each visible obstacle in the scene image, to obtain the actual illumination area in the scene image that is not blocked by the visible obstacles in the scene to be detected, which is used as the radar detection result corresponding to the scene to be detected.

[0014] As an optional implementation, in the first aspect of the present invention, the sub-graphics processor corrects the initial illumination area based on a pre-determined illumination shadow angle calculation algorithm, the position of the projection point of the point light source in the scene image, and the position of the cross-section projection of each visible obstacle in the scene image, to obtain the actual illumination area in the scene image that is not obscured by the visible obstacles in the scene to be detected, as the radar detection result corresponding to the scene to be detected, including:

[0015] The sub-graphics processor determines at least one target obstacle from all the visible obstacles based on the position of the cross-sectional projection of each visible obstacle in the scene image, wherein the cross-sectional projection of each target obstacle overlaps with the illumination range.

[0016] The sub-graphics processor determines whether there is at least one shadowed area in the initial lighting area based on a pre-determined lighting shadowed angle calculation algorithm, the position of the projection point of the point light source in the scene image, and the position of the projection of the cross section corresponding to each target obstacle in the scene image. The shadowed area includes the area that does not receive light because the light from the point light source is blocked by the target obstacle.

[0017] When the judgment result is yes, the sub-graphics processor removes all the shadowed corner areas from the initial illumination area to obtain the actual illumination area in the scene image that is not obscured by the visible obstacles in the scene to be detected, which is used as the radar detection result corresponding to the scene to be detected.

[0018] As an optional implementation, in the first aspect of the invention, before acquiring a scene image of the scene to be detected after it has been illuminated by a point light source, based on an image acquisition device pre-configured for the scene to be detected, the method further includes:

[0019] Determine the real-time position of the point light source placed in the scene to be detected;

[0020] Based on the real-time position, the scene to be detected is segmented to obtain a segmentation plane parallel to a pre-determined target plane, which serves as the illumination plane of the point light source in the scene to be detected. The point light source is located in the illumination plane, and the target plane includes a simulated imaging plane corresponding to an image acquisition device pre-configured for the scene to be detected.

[0021] As an optional implementation, in the first aspect of the present invention, the method further includes:

[0022] The radar detection results are overlaid with the scene to be detected to obtain a visualized radar detection result, which is then output to a display for viewing.

[0023] The step of overlaying the radar detection result with the scene to be detected to obtain a visualized radar detection result includes:

[0024] The radar detection results are projected onto a target layer pre-configured for the scene to be detected to obtain a visualized radar detection result. The target layer is parallel to the illumination plane, and the target layer includes a transparent layer pre-configured for the scene to be detected or a bottom layer pre-configured for the scene to be detected.

[0025] As an optional implementation, in the first aspect of the present invention, the method further includes:

[0026] Determine the effect display information matching the scene to be detected, wherein the effect display information includes effect display elements for displaying the radar detection results, wherein the effect display elements include static display elements and / or dynamic display elements;

[0027] Based on the effect display information, the radar detection results are plotted to update the radar detection results.

[0028] A second aspect of the present invention discloses a device for simulating radar detection effects based on a GPU in a simulation system, the device comprising:

[0029] The acquisition module is used to acquire scene images of the scene to be detected after being illuminated by a point light source, based on an image acquisition device pre-configured for the scene to be detected, wherein the point light source is placed in the scene to be detected;

[0030] The transmission module is used to transmit the scene image to a sub-graphics processor that matches the image acquisition unit in the graphics processor, so that the sub-graphics processor can determine the actual illuminated area in the scene image that is not obscured by visible obstacles in the scene to be detected, and use it as the radar detection result corresponding to the scene to be detected.

[0031] As an optional implementation, in a second aspect of the invention, the acquisition module acquires a scene image of the scene to be detected after it has been illuminated by a point light source, based on an image acquisition device pre-configured for the scene to be detected. The specific method includes:

[0032] Based on the image acquisition device pre-configured for the scene to be detected, the orthogonal projection of the illumination plane where the point light source is located onto the target plane after the scene to be detected is illuminated by the point light source is acquired, which is used as the scene image of the scene to be detected after being illuminated by the point light source.

[0033] The target plane includes the simulated imaging plane corresponding to the image acquisition device, the illumination plane is parallel to the target plane, the scene image includes the cross-sectional projection of each visible obstacle in the scene to be detected, the visible obstacle is an obstacle located within the acquisition field of view of the image acquisition device, and the cross-sectional projection of each visible obstacle includes the projection of the cross-section of the visible obstacle after it has been cut by the illumination plane.

[0034] As an optional implementation, in a second aspect of the invention, the sub-graphics processor determines the actual illuminated area in the scene image that is not obscured by visible obstacles in the scene to be detected, as a specific method for obtaining the radar detection result corresponding to the scene to be detected, including:

[0035] Based on the position of the point light source in the scene to be detected and the light parameters of the point light source, the illumination range of the point light source in the scene image is determined;

[0036] The area outside the cross-sectional projections corresponding to all the visible obstacles within the illumination range is defined as the initial illumination area in the scene image;

[0037] Based on the predetermined illumination shadow angle calculation algorithm, the position of the projection point of the point light source in the scene image, and the position of the projection of the cross section corresponding to each visible obstacle in the scene image, the initial illumination area is corrected to obtain the actual illumination area in the scene image that is not blocked by the visible obstacles in the scene to be detected, which is used as the radar detection result corresponding to the scene to be detected.

[0038] As an optional implementation, in a second aspect of the present invention, the sub-graphics processor corrects the initial illumination area based on a pre-determined illumination shadow angle calculation algorithm, the position of the projection point of the point light source in the scene image, and the position of the cross-section projection of each visible obstacle in the scene image, to obtain the actual illumination area in the scene image that is not obscured by the visible obstacles in the scene to be detected. This, as a specific method for obtaining the radar detection result corresponding to the scene to be detected, includes:

[0039] Based on the position of the cross-sectional projection corresponding to each of the visible obstacles in the scene image, at least one target obstacle is determined from all the visible obstacles, wherein the cross-sectional projection corresponding to each target obstacle has an overlapping area with the illumination range;

[0040] Based on the predetermined lighting angle calculation algorithm, the position of the projection point of the point light source in the scene image, and the position of the projection of the cross section corresponding to each target obstacle in the scene image, it is determined whether there is at least one shadow angle region in the initial lighting area. The shadow angle region includes the area that does not receive light because the light from the point light source is blocked by the target obstacle.

[0041] When the judgment result is yes, all the shadowed corner areas are removed from the initial illumination area to obtain the actual illumination area in the scene image that is not blocked by the visible obstacles in the scene to be detected, which is used as the radar detection result corresponding to the scene to be detected.

[0042] As an optional implementation, in a second aspect of the invention, the apparatus further includes:

[0043] The first determining module is used to determine the real-time position of the point light source placed in the scene to be detected in the scene to be detected before the acquisition module acquires the scene image of the scene to be detected after being illuminated by the point light source based on the image acquisition device pre-configured for the scene to be detected.

[0044] The cutting module is used to cut the scene to be detected based on the real-time position to obtain a cutting plane parallel to a pre-determined target plane, which serves as the illumination plane of the point light source in the scene to be detected. The point light source is located in the illumination plane, and the target plane includes a simulated imaging plane corresponding to the image acquisition device pre-configured for the scene to be detected.

[0045] As an optional implementation, in a second aspect of the invention, the apparatus further includes:

[0046] The overlay module is used to overlay the radar detection result with the scene to be detected to obtain a visualized radar detection result, which is output to the display terminal for viewing.

[0047] The specific method by which the overlay module overlays the radar detection results with the scene to be detected to obtain visualized radar detection results includes:

[0048] The radar detection results are projected onto a target layer pre-configured for the scene to be detected to obtain a visualized radar detection result. The target layer is parallel to the illumination plane, and the target layer includes a transparent layer pre-configured for the scene to be detected or a bottom layer pre-configured for the scene to be detected.

[0049] As an optional implementation, in a second aspect of the invention, the apparatus further includes:

[0050] The second determining module is used to determine the effect display information of the scene to be detected, wherein the effect display information includes effect display elements for displaying the radar detection results, wherein the effect display elements include static display elements and / or dynamic display elements;

[0051] A drawing module is used to draw the radar detection results based on the effect display information in order to update the radar detection results.

[0052] A third aspect of the present invention discloses another device for simulating radar detection effects based on a GPU in a simulation system, the device comprising:

[0053] Memory containing executable program code;

[0054] A processor coupled to the memory;

[0055] The processor calls the executable program code stored in the memory to execute the method for simulating radar detection effects based on GPU in the simulation system disclosed in the first aspect of the present invention.

[0056] The fourth aspect of the present invention discloses a computer storage medium storing computer instructions, which, when invoked, are used to execute the method for simulating radar detection effects based on GPU in the simulation system disclosed in the first aspect of the present invention.

[0057] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0058] In this embodiment of the invention, an image acquisition device pre-configured for the scene to be detected acquires an image of the scene after it is illuminated by a point light source, wherein the point light source is placed in the scene to be detected. The scene image is transmitted to a sub-graphics processor matched to the image acquisition device in the graphics processor, so that the sub-graphics processor determines the actual illuminated area in the scene image that is not obscured by visible obstacles in the scene to be detected, which is used as the radar detection result corresponding to the scene to be detected. It can be seen that implementing the present invention can acquire a scene image of the scene to be detected after it is illuminated by a light source through an image acquisition device, and process it in the graphics processor, i.e., the GPU, to obtain the actual illuminated area of ​​the point light source as the radar detection result. This can utilize the characteristic that the light source stops illuminating at the obstructed location to simulate the scene where the lidar beam is blocked by an obstacle, transforming the complex radar signal that originally needed to be processed by the CPU into a simple image signal that only needs to be processed by the GPU, thereby reducing the situation where the CPU workload is saturated and the GPU is idle, reducing the CPU computing power consumption in radar detection and improving the overall performance of radar detection. In addition, since the actual illuminated area is directly extracted from the scene image as the radar detection result, the intuitiveness and aesthetics of the radar detection result can also be improved. Attached Figure Description

[0059] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0060] Figure 1 This is a schematic diagram of a traditional radar detection method for plotting radar detection results;

[0061] Figure 2 This is a schematic diagram illustrating another method for plotting radar detection results using traditional radar detection techniques.

[0062] Figure 3 This is a flowchart illustrating a method for simulating radar detection effects based on a GPU in a simulation system, as disclosed in an embodiment of the present invention.

[0063] Figure 4 This is a schematic diagram of a scenario in a simulation system based on GPU to simulate radar detection effects, as disclosed in an embodiment of the present invention.

[0064] Figure 5 This is a schematic diagram of a scenario in another simulation system disclosed in this invention, which uses a GPU to simulate radar detection effects.

[0065] Figure 6 This is a schematic diagram of a scenario in another simulation system disclosed in this invention, which uses a GPU to simulate radar detection effects.

[0066] Figure 7 This is a flowchart illustrating another method for simulating radar detection effects based on a GPU in a simulation system disclosed in an embodiment of the present invention;

[0067] Figure 8 This is a schematic diagram of a scenario in another simulation system disclosed in this invention, which uses a GPU to simulate radar detection effects.

[0068] Figure 9 This is a schematic diagram of a scenario in another simulation system disclosed in this invention, which uses a GPU to simulate radar detection effects.

[0069] Figure 10 This is a schematic diagram of the structure of a device for simulating radar detection effects based on a GPU in a simulation system disclosed in an embodiment of the present invention;

[0070] Figure 11 This is a schematic diagram of the structure of a device for simulating radar detection effects based on a GPU in another simulation system disclosed in an embodiment of the present invention;

[0071] Figure 12 This is a schematic diagram of the structure of a device for simulating radar detection effects based on GPU in another simulation system disclosed in an embodiment of the present invention. Detailed Implementation

[0072] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0073] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or end that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or ends.

[0074] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0075] This invention discloses a method and apparatus for simulating radar detection effects based on a GPU in a simulation system. It can acquire scene images of the scene to be detected after being illuminated by a light source using an image acquisition device, and process these images using a graphics processing unit (GPU) to obtain the actual illuminated area of ​​the point light source as the radar detection result. This utilizes the characteristic that the light source stops illuminating at obstructed locations to simulate the scene where lidar rays are blocked by obstacles. It transforms the complex radar signal that originally required CPU processing into a simple image signal that only requires GPU processing, thereby reducing the CPU workload saturation and GPU idleness, reducing CPU computing power consumption in radar detection and improving the overall performance of radar detection. Furthermore, since the actual illuminated area is directly extracted from the scene image as the radar detection result, it also improves the intuitiveness and aesthetics of the radar detection results. Detailed descriptions follow.

[0076] Example 1

[0077] Please see Figure 3 , Figure 3 This is a flowchart illustrating a method for simulating radar detection effects based on a GPU in a simulation system, as disclosed in an embodiment of the present invention. Figure 3 The GPU-based method for simulating radar detection effects in the described simulation system can be applied to simulation software to simulate LiDAR detection effects, or to actual LiDAR detection, and can also be used to acquire planar self-scanning images. This invention does not limit the scope of the application. Figure 3 As shown, the method for simulating radar detection effects based on GPU in this simulation system can include the following operations:

[0078] 101. Based on the image acquisition device pre-configured for the scene to be detected, acquire scene images of the scene to be detected after being illuminated by a point light source.

[0079] In this embodiment of the invention, a point light source is placed in the scene to be detected.

[0080] As an optional implementation method, such as Figure 4 As shown, based on an image acquisition device pre-configured for the scene to be detected, acquiring scene images of the scene after it has been illuminated by a point light source can include:

[0081] Based on the image acquisition device pre-configured for the scene to be detected, the orthogonal projection of the illumination plane where the point light source is located onto the target plane after the scene to be detected is illuminated by the point light source is acquired, which is used as the scene image after the scene to be detected is illuminated by the point light source.

[0082] Optionally, the target plane may include the simulated imaging plane corresponding to the image acquisition device, the illumination plane is parallel to the target plane, and the scene image may include the cross-sectional projection corresponding to each visible obstacle in the scene to be detected. The visible obstacle is the obstacle located within the acquisition field of view of the image acquisition device, and the cross-sectional projection corresponding to each visible obstacle includes the projection of the cross-section of the visible obstacle after it is cut by the illumination plane.

[0083] As can be seen, implementing this optional implementation method can project the illumination plane where the point light source in the scene to be detected is located onto the simulation imaging plane as an orthogonal projection as the scene image. This can convert the three-dimensional radar detection scene into a two-dimensional radar detection scene, further simplifying the radar detection method, reducing the computing power required for radar detection, and by using orthogonal projection as the scene image, the ratio between the scene images collected at different angles and heights and the scene to be detected can be kept consistent, improving the uniformity of the display size of the real-time updated radar detection results.

[0084] In this embodiment of the invention, optionally, the image acquisition device moves with the point light source as it moves in the scene to be detected, and the movement parameters of the image acquisition device match the movement parameters of the point light source. The movement parameters may include movement distance and movement direction. Further optionally, the image acquisition device moves with the point light source in the direction of the illumination plane. That is, if the reference plane of the scene to be detected is a horizontal plane, and the illumination plane and the target plane are also horizontal planes, then the image acquisition device only moves when the horizontal component of the point light source's movement distance is not zero. If the point light source only moves in the vertical direction, the image acquisition device does not move accordingly. Optionally, the image acquisition device includes a camera.

[0085] like Figure 4 As shown, after loading the scene to be detected, a light is placed in the scene as a point light source to represent the radar. Since a common image acquisition device (i.e., the main camera in the figure) is usually used to render the entire simulation scene in a normal simulation scene, this embodiment, in addition to using a common image acquisition device to render the entire scene to be detected normally, also creates a new image acquisition device (i.e., the secondary camera in the figure). The new image acquisition device is an orthogonal camera that projects vertically onto the scene to be detected, and the new image acquisition device can only be used to acquire the cross-sectional projection of visible obstacles in the scene to be detected (i.e., the three buildings in the figure). For example, if the scene to be detected is placed horizontally, the image acquisition device can project vertically downwards or upwards and acquire the horizontal cross-section of the visible obstacles.

[0086] It should be noted that in other embodiments, the perspective projection of the illumination plane where the point light source is located onto the target plane after the scene to be detected is illuminated by the point light source can also be collected as a scene image.

[0087] 102. The scene image is transmitted to the sub-graphics processor that matches the image acquisition unit in the graphics processor, so that the sub-graphics processor can determine the actual illuminated area in the scene image that is not obscured by visible obstacles in the scene to be detected, and use it as the radar detection result corresponding to the scene to be detected.

[0088] In this embodiment of the invention, the graphics processor may include a GPU, and the sub-graphics processor may include a rendering pipeline pre-configured for the image acquisition device within the GPU.

[0089] As an optional implementation method, such as Figure 5 As shown, the sub-graphics processor determines the actual illuminated area in the scene image that is not obscured by visible obstacles in the scene to be detected, which serves as the radar detection result corresponding to the scene to be detected. This can include:

[0090] The sub-graphics processor determines the illumination range of the point light source in the scene image based on the position of the point light source in the scene to be detected and the light parameters of the point light source;

[0091] The sub-graphics processor determines the area outside the illuminating range, except for the cross-sectional projections corresponding to all visible obstacles, as the initial illumination area in the scene image;

[0092] The sub-graphics processor corrects the initial illumination area based on the pre-determined illumination angle calculation algorithm, the position of the projection point of the point light source in the scene image, and the position of the cross-section projection of each visible obstacle in the scene image. This results in the actual illumination area in the scene image that is not obscured by the visible obstacles in the scene to be detected, which is then used as the radar detection result for the scene to be detected.

[0093] In this embodiment of the invention, optionally, when the point light source moves in the scene to be detected, the corresponding radar detection results are also updated in real time.

[0094] In this optional implementation, the light parameters of the point light source may include the light length, and the illumination range of the point light source in the scene may include a circular area defined by the position of the projection point of the point light source in the scene image as the center and the light length as the radius. The position of the projection point of the point light source in the scene image may be determined by the position of the point light source in the scene to be detected.

[0095] As can be seen, implementing this optional implementation method can remove the cross-sectional projection of obstacles from the scene image that should originally be covered by the point light source, and further correct it according to the relevant algorithm to obtain the actual illumination range of the point light source as the radar detection result. This can transform the complex radar detection boundary confirmation process into a simple image cropping process, improve the efficiency and accuracy of radar detection, and further reduce the computing power required for radar detection.

[0096] In this optional implementation, optionally, such as Figure 5 As shown, the sub-graphics processor corrects the initial illumination area based on a pre-determined illumination angle calculation algorithm, the position of the projection point of the point light source in the scene image, and the position of the projection of the cross-section corresponding to each visible obstacle in the scene image. This corrects the actual illumination area in the scene image that is not obscured by visible obstacles in the scene to be detected, and serves as the radar detection result for the scene to be detected. This result may include:

[0097] The sub-graphics processor determines at least one target obstacle from all visible obstacles based on the position of the cross-sectional projection of each visible obstacle in the scene image, wherein the cross-sectional projection of each target obstacle overlaps with the illumination range.

[0098] The sub-graphics processor determines whether there is at least one shadowed area in the initial lighting area based on the pre-determined lighting shadowed angle calculation algorithm, the position of the projection point of the point light source in the scene image, and the position of the cross section projection of each target obstacle in the scene image. The shadowed area includes the area that does not receive light because the light from the point light source is blocked by the target obstacle.

[0099] When the judgment result is yes, the sub-graphics processor removes all shadowed corner areas from the initial illumination area to obtain the actual illumination area in the scene image that is not obscured by visible obstacles in the scene to be detected, which is used as the radar detection result corresponding to the scene to be detected.

[0100] As can be seen, implementing this optional implementation method can also remove the shadowed corner area blocked by obstacles from the initial illumination area according to the illumination shadowed corner calculation algorithm, and obtain the actual illumination range of the point light source as the radar detection result. This reduces the situation where the radar detection result retains the shadowed area due to only removing the cross-sectional projection of the obstacle. Furthermore, it can transform the complex radar detection boundary confirmation process into a simple graphic calculation process, improve the efficiency and accuracy of radar detection, and further reduce the computing power required for radar detection.

[0101] In this optional implementation, further options include, for example... Figure 5 As shown, the sub-graphics processor determines whether there is at least one shadowed region in the initial lighting area based on a pre-determined lighting shadow angle calculation algorithm, the position of the projection point of the point light source in the scene image, and the position of the cross-section projection of each target obstacle in the scene image. This can include:

[0102] For each target obstacle, the sub-graphics processor determines two tangents from the projection point to the cross-sectional projection of each target obstacle based on the pre-determined illumination shadow angle calculation algorithm, the position of the projection point of the point light source in the scene image, and the position of the cross-sectional projection of the target obstacle in the scene image. These two tangents are then used as the two target tangents for the target obstacle.

[0103] The sub-graphics processor extends the tangent line of each target corresponding to each obstacle from the tangent point to the edge of the initial lighting area;

[0104] For each target obstacle, the sub-graphics processor determines the area enclosed by the extended portion of the two target tangents corresponding to the target obstacle and the edge of the target obstacle, which is the occlusion area corresponding to the target obstacle;

[0105] For each target obstacle, the sub-graphics processor determines whether the occlusion area corresponding to the target obstacle overlaps with the initial illumination area. If the determination result is yes, it is determined that the initial illumination area has at least one shadowed corner area.

[0106] As can be seen, implementing this optional implementation method can also determine the area formed by the projection point of the point light source to the tangent of the cross-section projection of each obstacle and the edge of the cross-section projection of the obstacle, and determine whether there is a hidden corner area in each area. This can simplify the method of determining the hidden corner area and improve the efficiency and accuracy of determining the hidden corner area.

[0107] As can be seen, implementing the embodiments of the present invention can acquire scene images of the scene to be detected after being illuminated by a light source through an image acquisition device, and process them in a graphics processor (GPU) to obtain the actual illuminated area of ​​the point light source as the radar detection result. This can utilize the characteristic that the light source stops illuminating at the blocked location to simulate the scene where the lidar beam is blocked by an obstacle, transforming the complex radar signal that originally needed to be processed by the CPU into a simple image signal that only needs to be processed by the GPU. This reduces the situation where the CPU workload is saturated and the GPU is idle, reduces the CPU computing power consumption in radar detection, and improves the overall performance of radar detection. In addition, since the actual illuminated area is directly extracted from the scene image as the radar detection result, the intuitiveness and aesthetics of the radar detection result can also be improved.

[0108] In an optional embodiment, before acquiring a scene image of the scene to be detected after it has been illuminated by a point light source, based on an image acquisition device pre-configured for the scene to be detected, the method may further include:

[0109] Determine the real-time position of the point light source placed in the scene to be inspected;

[0110] Based on the real-time position, the scene to be detected is segmented to obtain a segmentation plane parallel to a pre-determined target plane, which serves as the illumination plane of the point light source in the scene to be detected. The point light source is located in the illumination plane, and the target plane may include the simulated imaging plane corresponding to the image acquisition device pre-configured for the scene to be detected.

[0111] For example, when the simulated imaging plane corresponding to the image acquisition device is located in the horizontal direction, the vertical height of the point light source in the scene to be detected can be determined, and the scene to be detected can be horizontally cut at this vertical height to obtain the illumination plane.

[0112] In this embodiment of the invention, optionally, the operation of slicing the scene to be detected can also be handled by the rendering pipeline.

[0113] As can be seen, implementing this optional embodiment can cut the scene to be detected according to the position of the point light source before acquiring the scene image, and use the cut plane as the illumination plane. This can simulate the difference in radar detection results at different positions, such as different heights. For example, radar signals can pass through the doorway but not through the door lintel. Therefore, cutting the scene to be detected can improve the accuracy of the obtained cross-sectional projection of the obstacle, thereby further improving the accuracy of radar detection.

[0114] In yet another optional embodiment, the method may further include:

[0115] The sub-graphics processor performs color processing on the initial lighting area to change the color of the initial lighting area to a preset color, such as red. The preset color is different from the colors of other areas in the scene image except for the initial lighting area.

[0116] And, such as Figure 6 As shown, the sub-graphics processor removes all shadowed corner areas from the initial illumination region to obtain the actual illumination area in the scene image that is not obscured by visible obstacles in the scene to be detected. This area serves as the radar detection result corresponding to the scene to be detected and may include:

[0117] The sub-graphics processor changes the color of all shadowed areas in the initial illumination area from a preset color to another color, and determines the remaining areas in the initial illumination area that are the preset colors as the actual illuminated areas in the scene image that are not obscured by visible obstacles in the scene to be detected, and uses them as the radar detection results corresponding to the scene to be detected.

[0118] As can be seen, implementing this optional embodiment can also color the initial illumination area, which facilitates the cropping of the initial illumination area, improves the intuitiveness of the actual illumination area determination process, and determines the area whose color finally meets the preset conditions as the actual illumination area, thereby improving the efficiency and accuracy of the actual illumination area determination.

[0119] Example 2

[0120] Please see Figure 7 , Figure 7 This is a flowchart illustrating another method for simulating radar detection effects based on a GPU in a simulation system disclosed in an embodiment of the present invention. Figure 7 The GPU-based method for simulating radar detection effects in the described simulation system can be applied to simulation software to simulate LiDAR detection effects, or to actual LiDAR detection, and can also be used to acquire planar self-scanning images. This invention does not limit the scope of the application. Figure 7 As shown, the method for simulating radar detection effects based on GPU in this simulation system can include the following operations:

[0121] 201. Based on the image acquisition device pre-configured for the scene to be detected, acquire scene images of the scene to be detected after being illuminated by a point light source.

[0122] 202. The scene image is transmitted to the sub-graphics processor that matches the image acquisition unit in the graphics processor, so that the sub-graphics processor can determine the actual illuminated area in the scene image that is not obscured by visible obstacles in the scene to be detected, and use it as the radar detection result corresponding to the scene to be detected.

[0123] 203. Overlay the radar detection results with the scene to be detected to obtain a visualized radar detection result.

[0124] In this embodiment of the invention, optionally, the visualized radar detection results are output to a display terminal for viewing. Optionally, the display terminal may include one or more display devices such as a mobile phone, a computer screen, or the display screen of a radar detector. This embodiment of the invention does not limit the display device.

[0125] As an optional implementation method, such as Figure 8 As shown, the radar detection results are overlaid with the scene to be detected to obtain visualized radar detection results, which may include:

[0126] The radar detection results are projected onto a target layer pre-configured for the scene to be detected to obtain a visualized radar detection result. The target layer is parallel to the illumination plane and includes a transparent layer or a bottom layer pre-configured for the scene to be detected.

[0127] As can be seen, by projecting the radar detection results onto the transparent layer and the bottom layer, this optional embodiment can reduce the situation where the radar detection results obscure the scene information to be detected.

[0128] In this optional implementation, projecting the radar detection results onto a target layer pre-configured for the scene to be detected to obtain visualized radar detection results may include:

[0129] When the radar detection result matches the illumination range of the point light source, that is, when it is a circular area, the radar detection result is projected onto the target layer pre-configured for the scene to be detected, based on the position of the projection point of the point light source in the radar detection result and the position of the point light source in the scene to be detected, to obtain the visualized radar detection result. The position of the projection point on the target layer after superposition matches the position of the point light source on the illumination plane.

[0130] When the radar detection results do not match the illumination range of the point light source, i.e., when it is a polygonal region, the radar detection results are projected onto a target layer pre-configured for the scene to be detected, based on the position of the projection point of the point light source in the radar detection results, the position of the point light source in the scene to be detected, the position of the cutout of the cross-sectional projection corresponding to each visible obstacle in the radar detection results, and the position of each visible obstacle in the scene to be detected. This results in a visualized radar detection result. The position of the projection point on the target layer after superposition matches the position of the point light source on the illumination plane, and the position of the cutout of the cross-sectional projection corresponding to each visible obstacle on the target layer matches the position of the visible obstacle on the illumination plane.

[0131] As can be seen, implementing this optional implementation method can also directly superimpose the radar detection results onto the scene to be detected based on the position of the point light source when the radar detection results show that there are no obstacles, and superimpose the radar detection results onto the scene to be detected based on the position of the light source and the position of the obstacle when the radar detection results show that there are obstacles, thereby improving the flexibility and accuracy of superimposing the radar detection results onto the scene to be detected.

[0132] As can be seen, implementing this embodiment of the invention involves acquiring scene images of the scene to be detected after being illuminated by a light source using an image acquisition device, and then processing these images in a graphics processing unit (GPU) to obtain the actual illuminated area of ​​the point light source as the radar detection result. This utilizes the characteristic that the light source stops illuminating at the blocked location to simulate the scene where the lidar beam is blocked by an obstacle. This transforms the complex radar signal that originally required CPU processing into a simple image signal that only requires GPU processing, thereby reducing the situation where the CPU is saturated while the GPU is idle. This reduces the CPU computing power consumption in radar detection and improves the overall performance of radar detection. In addition, since the actual illuminated area is directly extracted from the scene image as the radar detection result, the intuitiveness and aesthetics of the radar detection result are also improved. Furthermore, the radar monitoring result can be superimposed on the scene to be detected, allowing users to intuitively view the current radar detection range and indirectly observe the internal structure of obstacles in the scene to be detected. Users can also verify the radar detection result based on the scene to be detected, further improving the accuracy of the radar detection result.

[0133] In an optional embodiment, such as Figure 9 As shown, the method may further include:

[0134] Determine the effect display information for matching the scene to be detected, wherein the effect display information includes effect display elements for displaying radar detection results, wherein the effect display elements include static display elements and / or dynamic display elements;

[0135] Based on the displayed information, the radar detection results are plotted and updated.

[0136] In this optional embodiment, the effect display information may optionally include one or more of the color, size, shape, etc., of the effect display element. The effect display information may be a preset default option or a user-defined setting; this embodiment of the invention does not impose any limitations. Preferably, the effect display element may be a ring-shaped display element, such as a ring-shaped static element or a ring-shaped dynamic effect element. The center of the ring-shaped display element is the projection point of a point light source, which can vividly simulate a radar signal transmission scene. Furthermore, the radius difference between any two adjacent rings in the ring-shaped display element may be equal or unequal. For example, for each ring, the radius difference between it and the adjacent outer ring is greater than the radius difference between it and the adjacent inner ring. The area between any two adjacent rings may be filled with the same color or different colors; for example, it may be filled with a gradient color; this embodiment of the invention does not impose any limitations.

[0137] In this optional embodiment, if the effect display element is a ring display element and the radar detection result is a polygonal area, after drawing the ring display element and the radar detection result, it is necessary to further remove the part of the element area in the ring display element that matches the cross-sectional projection corresponding to the visible obstacle, thereby forming a signal discontinuity effect of the radar.

[0138] As can be seen, implementing this optional embodiment can draw static display elements and dynamic display elements in the radar detection results, thereby further improving the intuitiveness and aesthetics of the radar detection results. In addition, it meets the user's need to customize the drawing of radar detection effect display elements, improves the flexibility of radar detection result drawing, and further enhances the user experience.

[0139] Example 3

[0140] Please see Figure 10 , Figure 10 This is a schematic diagram of the structure of a device for simulating radar detection effects based on a GPU in a simulation system disclosed in an embodiment of the present invention. Figure 10 The GPU-based device for simulating radar detection effects in the described simulation system can be applied to simulation software to simulate LiDAR detection effects, or it can be applied to actual LiDAR detection, or it can be used to acquire planar self-scanning images. This invention does not limit the scope of the application. Figure 10 As shown, the device in this simulation system that simulates radar detection effects based on a GPU may include:

[0141] The acquisition module 301 is used to acquire scene images of the scene to be detected after being illuminated by a point light source, based on an image acquisition device pre-configured for the scene to be detected, wherein the point light source is placed in the scene to be detected;

[0142] The transmission module 302 is used to transmit the scene image to the sub-graphics processor that matches the image acquisition unit in the graphics processor, so that the sub-graphics processor can determine the actual illumination area in the scene image that is not blocked by visible obstacles in the scene to be detected, and use it as the radar detection result corresponding to the scene to be detected.

[0143] It is evident that implementation Figure 10 The described device can acquire scene images of the scene to be detected after being illuminated by a light source through an image acquisition device, and process them in a graphics processing unit (GPU) to obtain the actual illuminated area of ​​the point light source as the radar detection result. This can utilize the characteristic that the light source stops illuminating when it is blocked to simulate the scene where the lidar beam is blocked by an obstacle. It transforms the complex radar signal that originally needed to be processed by the CPU into a simple image signal that only needs to be processed by the GPU, thereby reducing the situation where the CPU is saturated and the GPU is idle, reducing the CPU computing power consumption in radar detection and improving the overall performance of radar detection. In addition, since the actual illuminated area is directly extracted from the scene image as the radar detection result, the intuitiveness and aesthetics of the radar detection result can also be improved.

[0144] In an optional embodiment, such as Figure 10 As shown, the acquisition module 301 acquires scene images of the scene to be detected after being illuminated by a point light source, based on an image acquisition device pre-configured for the scene to be detected. The specific methods may include:

[0145] Based on the image acquisition device pre-configured for the scene to be detected, the orthogonal projection of the illumination plane where the point light source is located onto the target plane after the scene to be detected is illuminated by the point light source is acquired, which is used as the scene image after the scene to be detected is illuminated by the point light source.

[0146] The target plane includes the simulated imaging plane corresponding to the image acquisition device. The illumination plane is parallel to the target plane. The scene image includes the cross-sectional projection of each visible obstacle in the scene to be detected. The visible obstacle is the obstacle located within the acquisition field of view of the image acquisition device. The cross-sectional projection of each visible obstacle includes the projection of the cross-section of the visible obstacle after it has been cut by the illumination plane.

[0147] It is evident that implementation Figure 10The described device can also project the illumination plane where the point light source in the scene to be detected is located onto the simulated imaging plane as an orthogonal projection as a scene image. This can convert the three-dimensional radar detection scene into a two-dimensional radar detection scene, further simplifying the radar detection method, reducing the computing power required for radar detection, and by using orthogonal projection as a scene image, the ratio between scene images acquired at different angles and heights and the scene to be detected can be kept consistent, improving the uniformity of the display size of the real-time updated radar detection results.

[0148] In another alternative embodiment, such as Figure 10 As shown, the sub-graphics processor determines the actual illuminated area in the scene image that is not obscured by visible obstacles in the scene to be detected, and uses this as the specific method for obtaining the radar detection result corresponding to the scene to be detected. This method may include:

[0149] Based on the position of the point light source in the scene to be detected and the light parameters of the point light source, the illumination range of the point light source in the scene image is determined.

[0150] The area outside the illuminating range, except for the cross-sectional projections corresponding to all visible obstacles, is defined as the initial illumination area in the scene image;

[0151] Based on the predetermined illumination angle calculation algorithm, the position of the projection point of the point light source in the scene image, and the position of the cross-section projection of each visible obstacle in the scene image, the initial illumination area is corrected to obtain the actual illumination area in the scene image that is not blocked by the visible obstacles in the scene to be detected, which is used as the radar detection result for the scene to be detected.

[0152] It is evident that implementation Figure 10 The described device can also remove the cross-sectional projection of obstacles from the scene image that should originally be covered by the point light source and further correct it according to relevant algorithms to obtain the actual illumination range of the point light source as the radar detection result. This can transform the complex radar detection boundary confirmation process into a simple image cropping process, improve the efficiency and accuracy of radar detection, and further reduce the computing power required for radar detection.

[0153] In yet another alternative embodiment, such as Figure 10 As shown, the sub-graphics processor corrects the initial illumination area based on a pre-determined illumination angle calculation algorithm, the position of the projection point of the point light source in the scene image, and the position of the projection of the cross section corresponding to each visible obstacle in the scene image. This results in the actual illumination area in the scene image that is not obscured by visible obstacles in the scene to be detected. The specific method for using this as the radar detection result for the scene to be detected can include:

[0154] Based on the position of the cross-sectional projection of each visible obstacle in the scene image, at least one target obstacle is determined from all visible obstacles, wherein the cross-sectional projection of each target obstacle overlaps with the illumination range.

[0155] Based on the predetermined lighting angle calculation algorithm, the position of the projection point of the point light source in the scene image, and the position of the cross section projection of each target obstacle in the scene image, it is determined whether there is at least one shadow angle region in the initial lighting area. The shadow angle region includes the area that does not receive light because the light from the point light source is blocked by the target obstacle.

[0156] When the judgment result is yes, all shadowed corner areas are removed from the initial illumination area to obtain the actual illumination area in the scene image that is not obscured by visible obstacles in the scene to be detected, which is used as the radar detection result corresponding to the scene to be detected.

[0157] It is evident that implementation Figure 10 The described device can also remove the shadowed corner area blocked by obstacles from the initial illumination area according to the illumination shadowed corner calculation algorithm, and obtain the actual illumination range of the point light source as the radar detection result. This reduces the situation where the radar detection result retains the shadowed area due to only removing the cross-sectional projection of the obstacle. Furthermore, it can transform the complex radar detection boundary confirmation process into a simple graphic calculation process, which improves the efficiency and accuracy of radar detection and further reduces the computing power required for radar detection.

[0158] In yet another alternative embodiment, such as Figure 10 As shown, the specific method by which the sub-graphics processor determines whether there is at least one shadowed corner region in the initial lighting area based on the pre-determined lighting shadow angle calculation algorithm, the position of the projection point of the point light source in the scene image, and the position of the cross-section projection of each target obstacle in the scene image, may include:

[0159] For each target obstacle, based on the pre-determined illumination shadow angle calculation algorithm, the position of the projection point of the point light source in the scene image, and the position of the cross-sectional projection of the target obstacle in the scene image, two tangents from the projection point to the cross-sectional projection of each target obstacle are determined, which are then used as the two target tangents for the target obstacle.

[0160] Extend the tangent line of each target corresponding to each obstacle from the tangent point to the edge of the initial lighting area;

[0161] For each target obstacle, the area enclosed by the line segments extending from the two target tangents corresponding to the target obstacle and the edge of the target obstacle is defined as the occlusion area corresponding to the target obstacle.

[0162] For each target obstacle, determine whether the occlusion area corresponding to the target obstacle overlaps with the initial illumination area. If the determination result is yes, it is determined that the initial illumination area has at least one shadowed corner area.

[0163] It is evident that implementation Figure 10 The described device can also determine the area formed by the projection point of the point light source to the tangent of the cross-section projection of each obstacle and the edge of the cross-section projection of the obstacle, and determine whether there is a hidden corner area in each area. This simplifies the method of determining the hidden corner area and improves the efficiency and accuracy of determining the hidden corner area.

[0164] In yet another alternative embodiment, such as Figure 11 As shown, the device may further include:

[0165] The first determining module 303 is used to determine the real-time position of the point light source placed in the scene to be detected before the acquisition module 301 acquires the scene image after the scene to be detected is illuminated by the point light source based on the image acquisition device pre-configured for the scene to be detected.

[0166] The cutting module 304 is used to cut the scene to be detected based on the real-time position to obtain a cutting plane parallel to a pre-determined target plane, which serves as the illumination plane of the point light source in the scene to be detected. The point light source is located in the illumination plane, and the target plane may include the simulated imaging plane corresponding to the image acquisition device pre-configured for the scene to be detected.

[0167] It is evident that implementation Figure 11 The described device can cut the scene to be detected according to the position of the point light source before acquiring scene images, and use the cut plane as the illumination plane. This can simulate the difference in radar detection results at different positions, such as different heights. For example, radar signals can pass through the doorway but not through the lintel above the doorway. Therefore, cutting the scene to be detected can improve the accuracy of the cross-sectional projection of the acquired obstacle, thereby further improving the accuracy of radar detection.

[0168] In yet another alternative embodiment, such as Figure 11 As shown, the device may further include:

[0169] The overlay module 305 is used to overlay the radar detection results with the scene to be detected to obtain a visualized radar detection result, which is then output to a display for viewing.

[0170] The overlay module 305 overlays the radar detection results with the scene to be detected to obtain visualized radar detection results. Specific methods for this can include:

[0171] The radar detection results are projected onto a target layer pre-configured for the scene to be detected to obtain a visualized radar detection result. The target layer is parallel to the illumination plane and includes a transparent layer or a bottom layer pre-configured for the scene to be detected.

[0172] It is evident that implementation Figure 11 The described device can also overlay radar monitoring results with the scene to be detected, allowing users to intuitively view the current radar detection range and indirectly observe the internal structure of obstacles in the scene to be detected. It also allows users to verify the radar detection results based on the scene to be detected, further improving the accuracy of the radar detection results. In addition, by projecting the radar detection results onto a transparent layer and a bottom layer, the obstruction of the scene information by the radar detection results can be reduced.

[0173] In yet another alternative embodiment, such as Figure 11 As shown, the overlay module 305 projects the radar detection results onto a target layer pre-configured for the scene to be detected, and obtains the visualized radar detection results in the following ways:

[0174] When the radar detection result matches the illumination range of the point light source, that is, when it is a circular area, the radar detection result is projected onto the target layer pre-configured for the scene to be detected based on the position of the projection point of the point light source in the radar detection result and the position of the point light source in the scene to be detected, so as to obtain the visualized radar detection result. The position of the projection point on the reference plane of the scene to be detected after superposition overlaps with the position of the light source on the reference plane.

[0175] When the radar detection results do not match the illumination range of the point light source, i.e., when it is a polygonal region, the radar detection results are projected onto a target layer pre-configured for the scene to be detected, based on the position of the projection point of the point light source in the radar detection results, the position of the point light source in the scene to be detected, the position of the cutout of the cross-sectional projection corresponding to each visible obstacle in the radar detection results, and the position of each visible obstacle in the scene to be detected. This results in a visualized radar detection result. After superposition, the position of the projection point on the reference plane of the scene to be detected overlaps with the position of the light source on the reference plane. After superposition, the position of the cutout of the cross-sectional projection corresponding to each visible obstacle on the reference plane overlaps with the position of the visible obstacle on the reference plane.

[0176] It is evident that implementation Figure 11The described device can also overlay the radar detection results onto the scene to be detected directly based on the position of the point light source when the radar detection results show that there are no obstacles, and overlay the radar detection results onto the scene to be detected based on the position of the electric light source and the position of the obstacle when the radar detection results show that there are obstacles, thereby improving the flexibility and accuracy of overlaying the radar detection results onto the scene to be detected.

[0177] In yet another alternative embodiment, such as Figure 11 As shown, the device may further include:

[0178] The second determining module 306 is used to determine the effect display information of the scene to be detected, wherein the effect display information includes effect display elements for displaying radar detection results, wherein the effect display elements include static display elements and / or dynamic display elements;

[0179] The drawing module 307 is used to draw the radar detection results based on the effect display information in order to update the radar detection results.

[0180] It is evident that implementation Figure 11 The described device can also render static and dynamic display elements in the radar detection results, thereby further improving the intuitiveness and aesthetics of the radar detection results.

[0181] Example 4

[0182] Please see Figure 12 , Figure 12 This is a schematic diagram of the structure of a device for simulating radar detection effects based on a GPU in another simulation system disclosed in an embodiment of the present invention. Figure 12 As shown, the device in this simulation system that simulates radar detection effects based on a GPU may include:

[0183] Memory 401 storing executable program code;

[0184] Processor 402 coupled to memory 401;

[0185] The processor 402 calls the executable program code stored in the memory 401 to execute the steps in the method for simulating radar detection effect based on GPU in the simulation system described in Embodiment 1 or Embodiment 2 of the present invention.

[0186] Example 5

[0187] This invention discloses a computer storage medium storing computer instructions. When these computer instructions are invoked, they are used to execute steps in the method for simulating radar detection effects based on GPU in the simulation system described in Embodiment 1 or Embodiment 2 of this invention.

[0188] Example 6

[0189] This invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to perform the steps in the method for simulating radar detection effects based on a GPU in the simulation system described in Embodiment 1 or Embodiment 2.

[0190] The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0191] Through the detailed description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0192] Finally, it should be noted that the method and apparatus for simulating radar detection effect based on GPU in a simulation system disclosed in the embodiments of the present invention are merely preferred embodiments of the present invention and are only used to illustrate the technical solutions of the present invention, not to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for simulating radar detection effects based on GPU in a simulation system, characterized in that, The method includes: Based on an image acquisition device pre-configured for the scene to be detected, a scene image of the scene to be detected after being illuminated by a point light source is acquired, wherein the point light source is placed in the scene to be detected; The scene image is transmitted to a sub-graphics processor that matches the image acquisition unit in the graphics processor, so that the sub-graphics processor determines the actual illuminated area in the scene image that is not obscured by visible obstacles in the scene to be detected, and uses it as the radar detection result corresponding to the scene to be detected. The sub-graphics processor determines the actual illuminated area in the scene image that is not obscured by visible obstacles in the scene to be detected, as the radar detection result corresponding to the scene to be detected, including: The sub-graphics processor determines the illumination range of the point light source in the scene image based on the position of the point light source in the scene to be detected and the light parameters of the point light source; The sub-graphics processor determines the area outside the cross-sectional projections corresponding to all the visible obstacles within the illumination range as the initial illumination area in the scene image; The sub-graphics processor corrects the initial illumination area based on a pre-determined illumination angle calculation algorithm, the position of the projection point of the point light source in the scene image, and the position of the cross-section projection of each visible obstacle in the scene image, to obtain the actual illumination area in the scene image that is not blocked by the visible obstacles in the scene to be detected, which is used as the radar detection result corresponding to the scene to be detected.

2. The method for simulating radar detection effects based on GPU in the simulation system according to claim 1, characterized in that, The image acquisition device, pre-configured for the scene to be detected, acquires scene images of the scene to be detected after being illuminated by a point light source, including: Based on the image acquisition device pre-configured for the scene to be detected, the orthogonal projection of the illumination plane where the point light source is located onto the target plane after the scene to be detected is illuminated by the point light source is acquired, which is used as the scene image of the scene to be detected after being illuminated by the point light source. The target plane includes the simulated imaging plane corresponding to the image acquisition device, the illumination plane is parallel to the target plane, the scene image includes the cross-sectional projection of each visible obstacle in the scene to be detected, the visible obstacle is an obstacle located within the acquisition field of view of the image acquisition device, and the cross-sectional projection of each visible obstacle includes the projection of the cross-section of the visible obstacle after it has been cut by the illumination plane.

3. The method for simulating radar detection effects based on GPU in the simulation system according to claim 1, characterized in that, The sub-graphics processor, based on a pre-determined illumination angle calculation algorithm, the position of the projection point of the point light source in the scene image, and the position of the projection of the cross-section corresponding to each visible obstacle in the scene image, corrects the initial illumination area to obtain the actual illumination area in the scene image that is not obscured by the visible obstacles in the scene to be detected, which serves as the radar detection result corresponding to the scene to be detected, including: The sub-graphics processor determines at least one target obstacle from all the visible obstacles based on the position of the cross-sectional projection of each visible obstacle in the scene image, wherein the cross-sectional projection of each target obstacle overlaps with the illumination range. The sub-graphics processor determines whether there is at least one shadowed area in the initial lighting area based on a pre-determined lighting shadowed angle calculation algorithm, the position of the projection point of the point light source in the scene image, and the position of the projection of the cross section corresponding to each target obstacle in the scene image. The shadowed area includes the area that does not receive light because the light from the point light source is blocked by the target obstacle. When the judgment result is yes, the sub-graphics processor removes all the shadowed corner areas from the initial illumination area to obtain the actual illumination area in the scene image that is not obscured by the visible obstacles in the scene to be detected, which is used as the radar detection result corresponding to the scene to be detected.

4. The method for simulating radar detection effects based on GPU in the simulation system according to claim 1, characterized in that, Before acquiring a scene image of the scene to be detected after it has been illuminated by a point light source, based on an image acquisition device pre-configured for the scene to be detected, the method further includes: Determine the real-time position of the point light source placed in the scene to be detected; Based on the real-time position, the scene to be detected is segmented to obtain a segmentation plane parallel to a pre-determined target plane, which serves as the illumination plane of the point light source in the scene to be detected. The point light source is located in the illumination plane, and the target plane includes a simulated imaging plane corresponding to an image acquisition device pre-configured for the scene to be detected.

5. The method for simulating radar detection effects based on GPU in the simulation system according to claim 2 or 4, characterized in that, The method further includes: The radar detection results are overlaid with the scene to be detected to obtain a visualized radar detection result, which is then output to a display for viewing. The step of overlaying the radar detection result with the scene to be detected to obtain a visualized radar detection result includes: The radar detection results are projected onto a target layer pre-configured for the scene to be detected to obtain a visualized radar detection result. The target layer is parallel to the illumination plane, and the target layer includes a transparent layer pre-configured for the scene to be detected or a bottom layer pre-configured for the scene to be detected.

6. The method for simulating radar detection effects based on GPU in the simulation system according to any one of claims 1-3, characterized in that, The method further includes: Determine the effect display information matching the scene to be detected, wherein the effect display information includes effect display elements for displaying the radar detection results, wherein the effect display elements include static display elements and / or dynamic display elements; Based on the effect display information, the radar detection results are plotted to update the radar detection results.

7. A device for simulating radar detection effects based on a GPU in a simulation system, characterized in that, The apparatus is used to perform a method for simulating radar detection effects based on a GPU in a simulation system as described in any one of claims 1-6, and the apparatus comprises: The acquisition module is used to acquire scene images of the scene to be detected after being illuminated by a point light source, based on an image acquisition device pre-configured for the scene to be detected, wherein the point light source is placed in the scene to be detected; The transmission module is used to transmit the scene image to a sub-graphics processor that matches the image acquisition unit in the graphics processor, so that the sub-graphics processor can determine the actual illuminated area in the scene image that is not obscured by visible obstacles in the scene to be detected, and use it as the radar detection result corresponding to the scene to be detected.

8. A device for simulating radar detection effects based on a GPU in a simulation system, characterized in that, The device includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the method for simulating radar detection effects based on GPU in the simulation system as described in any one of claims 1-6.

9. A computer storage medium, characterized in that, The computer storage medium stores computer instructions, which, when invoked, are used to execute the method for simulating radar detection effects based on GPU in the simulation system as described in any one of claims 1-6.

Images