A method, device and storage medium for recovering geometric information of rigid objects

Abstract

A method, device and storage medium for recovering geometric information of rigid objects, the method comprising the steps of: obtaining an original environment point cloud map; generating a target scene point cloud according to the original environment point cloud map; fitting the target scene point cloud into a three-dimensional space surface; projecting the target scene point cloud onto the three-dimensional space surface; interpolating the three-dimensional space surface; triangulating the three-dimensional space surface and obtaining triangulated triangles; and creating a rigid object geometric information depth map according to the triangulated triangles. The present application provides a method, device and storage medium for recovering geometric information of rigid objects, which use methods such as surface fitting, point cloud projection and interpolation to create a high-resolution point cloud map. The depth map generated based on the above point cloud map is more accurate than the depth map generated directly, and is closer to the real scene in terms of hierarchical distribution.

Description

Technical Field

[0001] The present invention belongs to the technical field of geometric information recovery of rigid objects, and in particular relates to a method, a device and a storage medium for geometric information recovery of rigid objects. Background Art

[0002] The geometric information of an object refers to its size and position in three-dimensional Euclidean space. The geometric information of objects has a wide range of applications in computer vision and mobile vehicles, such as semantic segmentation, three-dimensional modeling and reconstruction, detection, navigation, etc. Unmanned autonomous mobile vehicles have recently been widely used in industrial safety inspection, disinfection and epidemic prevention, warehousing and logistics, smart factories, agricultural harvesting and other fields. Unmanned autonomous mobile vehicles need to install different sensors to collect different types of information around them, and realize autonomous driving from the departure point to the destination based on the perception and observation of the surrounding environment. Cameras and LiDAR are two typical sensors. Cameras are used to collect visual information of the physical world, while LiDAR is used to obtain information about the distance, azimuth, height, speed, posture, shape, etc. of targets in physical space.

[0003] A depth map is a grayscale image that captures the distance information from scene objects to the viewpoint based on the brightness intensity. In practice, it is difficult to generate dense depth maps directly from sensor data due to the limited range of sensor data acquisition. Typically, depth maps can be recovered using computer vision information or LiDAR point clouds. Vision-based depth map recovery uses visual data collected by a camera to achieve 3D reconstruction using a multi-view stereo (MVS) approach. Visual 3D reconstruction is often not ideal because obvious features and textures are always missing in visual scenes. LiDAR point cloud-based depth map recovery can directly utilize the distance information provided by LiDAR measurements without the need for rich visual information. However, due to the limited measurement range and accuracy of LiDAR sensing and the sparse point cloud data obtained, this method makes it difficult to achieve dense and high-resolution 3D reconstruction.

[0004] At the same time, the implementation cost of LiDAR technology is very expensive. For example, the measurement accuracy of the 16-line LiDAR model Velodyne VLP-16 is ±3cm, and the angular resolutions in the vertical and horizontal directions are 2° and 0.1° respectively; while the measurement accuracy of the 64-line LiDAR model Velodyne HDL-64E is ±2cm, and the angular resolutions in the vertical and horizontal directions are 0.4° and 0.08° respectively. Due to the limited sensing range of LiDAR, environmental interference and economic considerations, what can be provided is often a sparse point cloud, which in turn leads to the generation of a sparse depth map. Although the measurement accuracy and resolution of the above-mentioned 64-line LiDAR are slightly better than those of the above-mentioned 16-line LiDAR, the price is more than 600,000 yuan, which is more than 20 times the price of the above-mentioned 16-line LiDAR (more than 30,000 yuan), and the cost performance is relatively low. Summary of the invention

[0005] In view of the above problems, the present invention provides a method, device and storage medium for recovering geometric information of a rigid object, which overcome the above problems or at least partially solve the above problems.

[0006] In order to solve the above technical problems, the present invention provides a method for recovering geometric information of a rigid object, the method comprising the steps of:

[0007] Get the original environment point cloud map;

[0008] Generate a target scene point cloud according to the original environment point cloud map;

[0009] Fitting the target scene point cloud into a three-dimensional space surface;

[0010] Projecting the target scene point cloud onto the three-dimensional space surface;

[0011] Performing an interpolation operation on the three-dimensional space surface;

[0012] Triangulating the three-dimensional space surface and obtaining triangulated triangles;

[0013] A depth map of rigid object geometry information is created from the triangulated triangles.

[0014] Preferably, the step of obtaining the original environment point cloud image comprises the following steps:

[0015] Obtain the original environment to be detected;

[0016] Arranging a laser radar in the original environment;

[0017] Using the laser radar to detect the original environment;

[0018] Obtain the original environment point cloud map generated by the laser radar.

[0019] Preferably, generating a target scene point cloud according to the original environment point cloud map comprises the steps of:

[0020] Acquire all point clouds in the original environment point cloud map;

[0021] Obtaining location information of all the point clouds;

[0022] Determine the target scene and the irrelevant scene according to all the position information;

[0023] Eliminate all point cloud data of the irrelevant scenes;

[0024] Acquire the remaining point cloud data of all the target scenes;

[0025] The target scene point cloud is generated according to all the remaining point cloud data.

[0026] Preferably, fitting the target scene point cloud into a three-dimensional space surface comprises the steps of:

[0027] Obtaining a background point cloud corresponding to the background in the target scene point cloud;

[0028] Obtaining a foreground point cloud corresponding to the foreground in the target scene point cloud;

[0029] Obtaining a rigid object point cloud corresponding to the rigid object in the foreground point cloud;

[0030] Fitting the background point cloud into a plane;

[0031] The rigid object point cloud is fitted into a quadratic surface.

[0032] Preferably, projecting the target scene point cloud onto the three-dimensional space surface comprises the steps of:

[0033] Acquire a background point cloud in the target scene point cloud and a rigid object point cloud in the foreground;

[0034] Acquire a plane and a quadratic surface in the three-dimensional space surface;

[0035] Projecting the background point cloud onto the plane;

[0036] The rigid object point cloud is projected onto the quadratic surface.

[0037] Preferably, projecting the rigid object point cloud onto the quadratic surface comprises the steps of:

[0038] Acquire a first rigid object point cloud corresponding to a first rigid object in the rigid object point cloud;

[0039] Acquire a second rigid object point cloud corresponding to a second rigid object in the rigid object point cloud;

[0040] Acquire a first quadratic surface corresponding to the first rigid object in the quadratic surface;

[0041] Obtaining a second quadratic surface corresponding to the second rigid object in the quadratic surface;

[0042] Projecting the first rigid object point cloud onto the first quadratic surface;

[0043] The second rigid object point cloud is projected onto the second quadratic surface.

[0044] Preferably, the interpolation operation on the three-dimensional space surface comprises the steps of:

[0045] Obtaining real scene data information in the original environment to be detected;

[0046] Generate a real scene point cloud according to the real scene data information;

[0047] Interpolation is performed on the three-dimensional space surface according to the real scene point cloud.

[0048] The present invention also provides a device for recovering geometric information of a rigid object, the device comprising the steps of:

[0049] An acquisition module is used to obtain the original environment point cloud map;

[0050] A target scene point cloud generation module, used to generate a target scene point cloud according to the original environment point cloud map;

[0051] A three-dimensional space surface fitting module, used for fitting the target scene point cloud into a three-dimensional space surface;

[0052] A projection module, used for projecting the target scene point cloud onto the three-dimensional space surface;

[0053] An interpolation module, used for performing an interpolation operation on the three-dimensional space surface;

[0054] A triangulation module, used for triangulating the three-dimensional space surface and obtaining triangulation triangles;

[0055] A depth map creation module is used to create a depth map of rigid object geometric information based on the triangulated triangles.

[0056] The present invention further provides an electronic device, comprising:

[0057] at least one processor; and,

[0058] a memory communicatively connected to the at least one processor; wherein,

[0059] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform any of the aforementioned rigid object geometric information recovery methods.

[0060] The present invention also provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable the computer to execute any of the aforementioned rigid object geometric information recovery methods.

[0061] One or more technical solutions in the embodiments of the present invention have at least the following technical effects or advantages: The method, device and storage medium for recovering geometric information of rigid objects provided by the present application use surface fitting, point cloud projection, interpolation and other methods to create a high-resolution point cloud map. The depth map generated based on the above point cloud map is more accurate than the depth map generated directly, and is closer to the real scene in terms of hierarchical distribution. The above method enables low-priced LiDAR sensors to have performance close to that of high-priced LiDAR sensors, improves the functional performance of LiDAR sensors, and saves unnecessary expenses. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0063] Figure 1 It is a flowchart of a method for recovering geometric information of a rigid object provided by an embodiment of the present invention;

[0064] Figure 2 It is a structural schematic diagram of a rigid object geometric information recovery device provided by an embodiment of the present invention;

[0065] Figure 3 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention;

[0066] Figure 4 is a schematic diagram of the structure of a non-transitory computer-readable storage medium provided by an embodiment of the present invention;

[0067] Figure 5 is the original environment map provided by the embodiment of the present invention;

[0068] Figure 6 Yes Figure 5The sparse point cloud obtained by detecting the original environment in ;

[0069] Figure 7 It is directly Figure 6 Processing the generated depth map;

[0070] Figure 8 It is a depth map generated by a method for recovering geometric information of a rigid object provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0071] The present invention will be described in detail below in conjunction with specific implementations and examples, and the advantages and various effects of the present invention will be more clearly presented. It should be understood by those skilled in the art that these specific implementations and examples are used to illustrate the present invention, rather than to limit the present invention.

[0072] Throughout the specification, unless otherwise specifically stated, the terms used herein should be understood as meanings commonly used in the art. Therefore, unless otherwise defined, all technical and scientific terms used herein have the same meanings as those generally understood by those skilled in the art to which the present invention belongs. In the event of a conflict, the present specification takes precedence.

[0073] Unless otherwise specified, various raw materials, reagents, instruments and equipment used in the present invention can be purchased from the market or prepared by existing methods.

[0074] like Figure 1 In an embodiment of the present application, the present invention provides a method for recovering geometric information of a rigid object, the method comprising the steps of:

[0075] S1: Get the original environment point cloud map;

[0076] In an embodiment of the present application, the step of obtaining the original environment point cloud image includes the following steps:

[0077] Obtain the original environment to be detected;

[0078] Arranging a laser radar in the original environment;

[0079] Using the laser radar to detect the original environment;

[0080] Obtain the original environment point cloud map generated by the laser radar.

[0081] In the embodiment of the present application, when obtaining the original environment point cloud map, first find the original environment to be detected, such as Figure 5The original environment is shown, and then a laser radar is arranged in the original environment; the original environment is then detected using the laser radar, and a point cloud map of the original environment can be generated after the laser radar detection, and then the point cloud map of the original environment generated by the laser radar is obtained.

[0082] In the embodiment of the present application, the model of the laser radar is Velodyne VLP-16.

[0083] S2: Generate a target scene point cloud according to the original environment point cloud map;

[0084] In an embodiment of the present application, generating a target scene point cloud according to the original environment point cloud map comprises the following steps:

[0085] Acquire all point clouds in the original environment point cloud map;

[0086] Obtaining location information of all the point clouds;

[0087] Determine the target scene and the irrelevant scene according to all the position information;

[0088] Eliminate all point cloud data of the irrelevant scenes;

[0089] Acquire the remaining point cloud data of all the target scenes;

[0090] The target scene point cloud is generated according to all the remaining point cloud data.

[0091] In the embodiment of the present application, there may be many irrelevant objects in the original environment, so the point cloud data of irrelevant scenes will also exist in the obtained original environment point cloud map. At this time, it is necessary to delete the point cloud data of these irrelevant scenes to generate the target scene point cloud. Specifically, all point clouds and the position information of all point clouds in the original environment point cloud map are obtained, and the target scene and irrelevant scenes are determined based on all the position information. For example, the scene corresponding to the position within the research range is the target scene, and the scene corresponding to the position outside this range is the irrelevant scene. Then, the point cloud data of all the irrelevant scenes are eliminated, and the remaining point cloud data of all the target scenes can be obtained, and the target scene point cloud is generated based on all the remaining point cloud data.

[0092] In the embodiments of the present application, Figure 5 The original environment shown can be obtained after processing in steps S1 and S2. Figure 6 The target scene point cloud is shown.

[0093] S3: fitting the target scene point cloud into a three-dimensional space surface;

[0094] In an embodiment of the present application, fitting the target scene point cloud into a three-dimensional space surface comprises the steps of:

[0095] Obtaining a background point cloud corresponding to the background in the target scene point cloud;

[0096] Obtaining a foreground point cloud corresponding to the foreground in the target scene point cloud;

[0097] Obtaining a rigid object point cloud corresponding to the rigid object in the foreground point cloud;

[0098] Fitting the background point cloud into a plane;

[0099] The rigid object point cloud is fitted into a quadratic surface.

[0100] In the embodiment of the present application, when fitting the target scene point cloud into a three-dimensional space surface, it is necessary to fit different objects in the target scene point cloud into corresponding faces in the three-dimensional space surface. Specifically, obtain the background point cloud corresponding to the background in the target scene point cloud, such as Figure 5 , and obtain the rigid object point cloud corresponding to the rigid object in the foreground point cloud, such as Figure 5 The cup and box in the image are then fitted into a plane and the rigid object point cloud is fitted into a quadratic surface.

[0101] S4: Projecting the target scene point cloud onto the three-dimensional space surface;

[0102] In an embodiment of the present application, projecting the target scene point cloud onto the three-dimensional space surface comprises the steps of:

[0103] Acquire a background point cloud in the target scene point cloud and a rigid object point cloud in the foreground;

[0104] Acquire a plane and a quadratic surface in the three-dimensional space surface;

[0105] Projecting the background point cloud onto the plane;

[0106] The rigid object point cloud is projected onto the quadratic surface.

[0107] In an embodiment of the present application, when the target scene point cloud is projected onto the three-dimensional space surface, the background point cloud in the target scene point cloud and the rigid object point cloud in the foreground as well as the plane and quadratic surface in the three-dimensional space surface are first obtained, and then the background point cloud and the rigid object point cloud in the foreground are respectively projected onto the corresponding planes or surfaces, and the rigid object point cloud is projected onto the quadratic surface, that is, the corresponding objects are projected onto their corresponding surfaces so that the point clouds are smoothly distributed.

[0108] In an embodiment of the present application, projecting the rigid object point cloud onto the quadratic surface comprises the steps of:

[0109] Acquire a first rigid object point cloud corresponding to a first rigid object in the rigid object point cloud;

[0110] Acquire a second rigid object point cloud corresponding to a second rigid object in the rigid object point cloud;

[0111] Acquire a first quadratic surface corresponding to the first rigid object in the quadratic surface;

[0112] Obtaining a second quadratic surface corresponding to the second rigid object in the quadratic surface;

[0113] Projecting the first rigid object point cloud onto the first quadratic surface;

[0114] The second rigid object point cloud is projected onto the second quadratic surface.

[0115] In the embodiment of the present application, when the rigid object point cloud is projected onto the quadratic surface, the rigid object point cloud may correspond to point clouds of multiple rigid objects, and in this case, each rigid object point cloud needs to be projected onto the corresponding three-dimensional space surface. For example, there are a first rigid object and a second rigid object, the first rigid object corresponds to a first rigid object point cloud and a first quadratic surface, and the second rigid object corresponds to a second rigid object point cloud and a second quadratic surface. In this case, the first rigid object point cloud is projected onto the first quadratic surface, and the second rigid object point cloud is projected onto the second quadratic surface.

[0116] S5: performing an interpolation operation on the three-dimensional space surface;

[0117] In an embodiment of the present application, the interpolation operation on the three-dimensional space surface includes the steps of:

[0118] Obtaining real scene data information in the original environment to be detected;

[0119] Generate a real scene point cloud according to the real scene data information;

[0120] Interpolation is performed on the three-dimensional space surface according to the real scene point cloud.

[0121] In an embodiment of the present application, in order to enhance the similarity between the three-dimensional space surface and the real scene, it is necessary to obtain the real scene data information in the original environment to be detected, and generate a real scene point cloud based on the real scene data information, and then interpolate on the three-dimensional space surface based on the real scene point cloud.

[0122] S6: triangulate the three-dimensional space surface and obtain triangulation triangles;

[0123] In the embodiment of the present application, the Delaunay triangulation algorithm is used to triangulate the three-dimensional space surface and obtain triangulation triangles.

[0124] S7: Create a rigid object geometric information depth map based on the triangulated triangles.

[0125] In the embodiment of the present application, a Velodyne VLP-16 laser radar is used. Figure 5 The original environment shown can be obtained after processing in steps S1-S7. Figure 8 The depth map of the target scene is shown.

[0126] In the embodiments of the present application, Figure 6 The Velodyne VLP-16 16-line laser radar Figure 5 The sparse point cloud obtained by the original environment detection in Figure 7 For direct Figure 6 The depth map generated by processing is Figure 8 By comparison, it can be found that the method for recovering geometric information of rigid objects provided by the present application is more accurate than the directly generated depth map and is closer to the real scene in terms of hierarchical distribution.

[0127] like Figure 2 In an embodiment of the present application, the present invention further provides a device for recovering geometric information of a rigid object, the device comprising the steps of:

[0128] An acquisition module 10 is used to acquire an original environment point cloud map;

[0129] A target scene point cloud generation module 20 is used to generate a target scene point cloud according to the original environment point cloud map;

[0130] A three-dimensional space surface fitting module 30, used for fitting the target scene point cloud into a three-dimensional space surface;

[0131] A projection module 40, configured to project the target scene point cloud onto the three-dimensional space surface;

[0132] An interpolation module 50, used for performing an interpolation operation on the three-dimensional space surface;

[0133] A triangulation module 60, configured to triangulate the three-dimensional space surface and obtain triangulation triangles;

[0134] The depth map creation module 70 is used to create a rigid object geometric information depth map according to the triangulated triangles.

[0135] The present application provides a device for recovering geometric information of a rigid object, which can execute a method for recovering geometric information of a rigid object provided by the above steps.

[0136] Reference below Figure 3 , which shows a schematic diagram of the structure of an electronic device 100 suitable for implementing the embodiment of the present disclosure. The electronic device in the embodiment of the present disclosure may include, but is not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle terminals (such as vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 3 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0137] like Figure 3 As shown, the electronic device 100 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 101, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 102 or a program loaded from a storage device 108 to a random access memory (RAM) 103. In the RAM 103, various programs and data required for the operation of the electronic device 100 are also stored. The processing device 101, the ROM 102, and the RAM 103 are connected to each other via a bus 104. An input / output (I / O) interface 105 is also connected to the bus 104.

[0138] Typically, the following devices may be connected to the I / O interface 105: an input device 106 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 107 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 108 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 109. The communication device 109 may allow the electronic device 100 to communicate with other devices wirelessly or by wire to exchange data. Although the electronic device 100 having various devices is shown in the figure, it should be understood that it is not required to implement or have all the devices shown. More or fewer devices may be implemented or have alternatively.

[0139] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 109, or installed from the storage device 108, or installed from the ROM 102. When the computer program is executed by the processing device 101, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0140] Reference below Figure 4 , which shows a structural schematic diagram of a computer-readable storage medium suitable for implementing an embodiment of the present disclosure, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it can implement any of the rigid object geometric information recovery methods described above.

[0141] It should be noted that the computer-readable medium disclosed above may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0142] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0143] The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device: obtains at least two Internet Protocol addresses; sends a node evaluation request including the at least two Internet Protocol addresses to a node evaluation device, wherein the node evaluation device selects an Internet Protocol address from the at least two Internet Protocol addresses and returns it; receives the Internet Protocol address returned by the node evaluation device; wherein the obtained Internet Protocol address indicates an edge node in a content distribution network.

[0144] Alternatively, the computer-readable medium carries one or more programs, and when the one or more programs are executed by the electronic device, the electronic device: receives a node evaluation request including at least two Internet Protocol addresses; selects an Internet Protocol address from the at least two Internet Protocol addresses; and returns the selected Internet Protocol address; wherein the received Internet Protocol address indicates an edge node in a content distribution network.

[0145] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0146] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0147] The units involved in the embodiments described in the present disclosure may be implemented by software or hardware. The name of a unit does not limit the unit itself in some cases. For example, the first acquisition unit may also be described as a "unit for acquiring at least two Internet Protocol addresses".

[0148] The present application provides a method, device and storage medium for recovering geometric information of rigid objects. The method uses surface fitting, point cloud projection, interpolation and other methods to create a high-resolution point cloud map. The depth map generated based on the point cloud map is more accurate than the depth map generated directly, and is closer to the real scene in terms of layer distribution. The method enables low-priced LiDAR sensors to have performance close to that of high-priced LiDAR sensors, improves the functional performance of LiDAR sensors, and saves unnecessary expenses.

[0149] It should be understood that the above specific embodiments of the present invention are only used to illustrate or explain the principles of the present invention, and do not constitute a limitation of the present invention. Therefore, any modifications, equivalent substitutions, improvements, etc. made without departing from the spirit and scope of the present invention should be included in the protection scope of the present invention. In addition, the appended claims of the present invention are intended to cover all changes and modifications that fall within the scope and boundaries of the appended claims, or the equivalent forms of such scope and boundaries.

Claims

1. A method for recovering geometric information of a rigid object, characterized in that: The method comprises the steps of: Get the original environment point cloud map; Generate a target scene point cloud according to the original environment point cloud map; Fitting the target scene point cloud into a three-dimensional space surface; Projecting the target scene point cloud onto the three-dimensional space surface; Performing an interpolation operation on the three-dimensional space surface; Triangulating the three-dimensional space surface and obtaining triangulated triangles; A depth map of rigid object geometry information is created based on the triangulated triangles.

2. The method for recovering geometric information of a rigid object according to claim 1, characterized in that: The step of obtaining the original environment point cloud image comprises: Obtain the original environment to be detected; Arranging a laser radar in the original environment; Using the laser radar to detect the original environment; Obtain the original environment point cloud map generated by the laser radar.

3. The method for recovering geometric information of a rigid object according to claim 1, characterized in that: Generating the target scene point cloud according to the original environment point cloud map comprises the following steps: Acquire all point clouds in the original environment point cloud map; Obtaining location information of all the point clouds; Determine the target scene and the irrelevant scene according to all the position information; Eliminate all point cloud data of the irrelevant scenes; Acquire the remaining point cloud data of all the target scenes; The target scene point cloud is generated according to all the remaining point cloud data.

4. The method for recovering geometric information of a rigid object according to claim 1, characterized in that: The step of fitting the target scene point cloud into a three-dimensional space surface comprises the following steps: Obtaining a background point cloud corresponding to the background in the target scene point cloud; Obtaining a foreground point cloud corresponding to the foreground in the target scene point cloud; Obtaining a rigid object point cloud corresponding to the rigid object in the foreground point cloud; Fitting the background point cloud into a plane; The rigid object point cloud is fitted into a quadratic surface.

5. The method for recovering geometric information of a rigid object according to claim 1, characterized in that: The projecting of the target scene point cloud onto the three-dimensional space surface comprises the steps of: Acquire a background point cloud in the target scene point cloud and a rigid object point cloud in the foreground; Acquire a plane and a quadratic surface in the three-dimensional space surface; Projecting the background point cloud onto the plane; The rigid object point cloud is projected onto the quadratic surface.

6. The method for recovering geometric information of a rigid object according to claim 5, characterized in that: The projecting of the rigid object point cloud onto the quadratic surface comprises the steps of: Acquire a first rigid object point cloud corresponding to a first rigid object in the rigid object point cloud; Acquire a second rigid object point cloud corresponding to a second rigid object in the rigid object point cloud; Acquire a first quadratic surface corresponding to the first rigid object in the quadratic surface; Obtaining a second quadratic surface corresponding to the second rigid object in the quadratic surface; Projecting the first rigid object point cloud onto the first quadratic surface; The second rigid object point cloud is projected onto the second quadratic surface.

7. The method for recovering geometric information of a rigid object according to claim 1, characterized in that: The interpolation operation on the three-dimensional space surface comprises the steps of: Obtaining real scene data information in the original environment to be detected; Generate a real scene point cloud according to the real scene data information; Interpolation is performed on the three-dimensional space surface according to the real scene point cloud.

8. A device for recovering geometric information of a rigid object, characterized in that: The device comprises the steps of: An acquisition module is used to obtain the original environment point cloud map; A target scene point cloud generation module, used to generate a target scene point cloud according to the original environment point cloud map; A three-dimensional space surface fitting module, used for fitting the target scene point cloud into a three-dimensional space surface; A projection module, used for projecting the target scene point cloud onto the three-dimensional space surface; An interpolation module, used for performing an interpolation operation on the three-dimensional space surface; A triangulation module, used for triangulating the three-dimensional space surface and obtaining triangulation triangles; A depth map creation module is used to create a depth map of rigid object geometric information based on the triangulated triangles.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the rigid object geometric information recovery method according to any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, characterized in that: The non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the rigid object geometric information recovery method described in any one of claims 1-7.

Images