A laser radar three-dimensional reconstruction method, device, electronic device and storage medium

Abstract

The present application relates to the field of three-dimensional reconstruction technology, and in particular to a laser radar three-dimensional reconstruction method, device, electronic device and storage medium. The method comprises: performing three-dimensional reconstruction scanning on a work scene by means of a laser radar, and obtaining rotation angle data and radar measurement data of the laser radar in real time; the rotation angle data comprises angle data and corresponding timestamp data, and the radar measurement data comprises point cloud data and corresponding timestamp data; comparing the timestamp data in the radar measurement data and the rotation angle data, and based on the time difference between the two, screening out matching point cloud data and angle data to obtain laser radar data with precise angular posture; parsing and processing the screened point cloud data through a callback function, performing coordinate conversion on the parsed point cloud data, and splicing and fusing the point cloud data in different time series based on the rotation angle. The present application effectively reduces the occupancy of spatial structures and improves the clarity and accuracy of three-dimensional reconstruction.

Description

Technical Field

[0001] The present application relates to the field of three-dimensional reconstruction technology, and in particular to a laser radar three-dimensional reconstruction method, device, electronic device and storage medium. Background Art

[0002] When an outdoor live power distribution robot is performing remote control or intelligent operations, it first needs to perform a three-dimensional reconstruction of the operating scene to determine the operating position of the robotic arm and provide basic data for the robotic arm's obstacle avoidance strategy.

[0003] Three-dimensional reconstruction can be achieved using visual perception devices such as binocular cameras and lidar. However, binocular cameras are prone to produce a large number of void areas when exposed to sufficient sunlight outdoors, resulting in unsatisfactory reconstruction results.

[0004] The principle of LiDAR ranging is that a laser with a certain wavelength is emitted from the laser center. The distance of the light spot emitted by its internal ranging module is directly proportional to the ranging range. That is, the longer the ranging range, the larger the light spot size. If the light spot illuminates a small target object and there is a larger object in the area behind the target, it is very easy to cause multiple reflections of the light spot, resulting in erroneous results in laser ranging. In addition, traditional LiDAR modeling will cause smearing problems, resulting in unsatisfactory modeling results.

[0005] When using lidar as the front-end sensor, a slide is often used in conjunction with a multi-line lidar. When using this method, the slide is generally designed to be longer in order to increase the field of view, which will take up more operating platform space and increase the complexity of the structural design. Summary of the Invention

[0006] The embodiments of the present application provide a laser radar three-dimensional reconstruction method, device, electronic device and storage medium to at least solve the problems in the related art of using laser radar for three-dimensional reconstruction, which occupy a large operating platform space, easily produce erroneous modeling results due to multiple reflections of the light spot, and cause unclear and inaccurate modeling due to the appearance of ghosting.

[0007] In a first aspect, an embodiment of the present application provides a method for three-dimensional reconstruction of a laser radar, the method comprising:

[0008] Step S1, performing a three-dimensional reconstruction scan of the work scene using a laser radar, and obtaining the current rotation angle data and radar measurement data of the laser radar in real time; the rotation angle data includes angle data and corresponding timestamp data, and the radar measurement data includes point cloud data and corresponding timestamp data;

[0009] Step S2, comparing the timestamp data in the radar measurement data with the timestamp data in the rotation angle data, and screening out matching point cloud data and angle data based on the time difference between the two to obtain lidar data with accurate angle pose;

[0010] In step S3, the point cloud data filtered out in step S2 is parsed and processed by the callback function, and the coordinates of the parsed point cloud data are converted, and the point cloud data in different time series are spliced ​​and fused according to their rotation angles.

[0011] In some embodiments, in step S2, the obtained rotation angle data and radar measurement data are stored in a queue, and the rotation angle data and radar measurement data are extracted in sequence starting from the head of the queue, and the difference between the timestamp data of the two sets of data is analyzed and compared, and the difference is compared with a preset first threshold.

[0012] If the difference is less than the first threshold, the angle data of the current timestamp is used as the rotation angle of the current point cloud data;

[0013] If the difference is greater than the first threshold, continue to judge the size of the timestamp data in the radar measurement data and the timestamp data in the rotation angle data, discard the group of data with smaller timestamp data values ​​in the radar measurement data and the rotation angle data, and select the next group of data from the discarded data queue in sequence to compare the difference in timestamp data between the two with the data that has not been discarded, and compare the difference with the preset first threshold.

[0014] In some embodiments, in step S1, before the laser radar operates, it is determined whether the laser radar is in a preset initial posture;

[0015] If it is in the preset initial posture, the laser radar performs rotation measurement to obtain the current rotation angle data and radar measurement data in real time;

[0016] If it is not in the preset initial posture, the laser radar is controlled to rotate to the preset initial posture and then perform rotation measurement to obtain the current rotation angle data and radar measurement data in real time.

[0017] In some embodiments, step S3 includes applying point cloud filtering to the point cloud data filtered in step S2 to remove abnormal measurement points, including:

[0018] The filtered point cloud data is further segmented to obtain a local point cloud set;

[0019] Calculate the spatial distance from the points in the local point cloud set to the lidar origin, and calculate the standard deviation of the local point cloud set in combination with the preset point cloud distance mean in the local point cloud set;

[0020] Compare the standard deviation with the preset second threshold,

[0021] If the standard deviation is greater than a preset second threshold, it is determined that there are abnormal measurement points in the local point cloud collection area, and a filtering operation is performed to remove the abnormal measurement points;

[0022] If the standard deviation is less than the preset second threshold, it is determined that there is no abnormal measurement point in the local point cloud collection area, and other local point cloud collections are determined at this time.

[0023] In some embodiments, when removing abnormal measurement points:

[0024] When the standard deviation is greater than the preset second threshold, the points in the local point cloud set are judged.

[0025] If the spatial distance from a point in the local point cloud set to the lidar origin is less than or equal to the preset point cloud distance average, the point is considered to be normal and saved.

[0026] If the spatial distance from a point in the local point cloud set to the lidar origin is greater than the preset point cloud distance mean, the point is judged as an abnormal point and is removed.

[0027] In some embodiments, in step S3, during the coordinate conversion processing of the point cloud data, the radar measurement data after the point cloud filtering processing is subjected to a rotational translation transformation according to the rotation angle data matched therewith to obtain a three-dimensional reconstruction result based on the initial posture state, and the obtained result is saved.

[0028] In the second aspect, an embodiment of the present application provides a laser radar three-dimensional reconstruction device, which includes a laser radar and a rotating structure. The rotating structure can rotate the laser radar in the horizontal direction for measurement. The origin position of the laser radar and the center of the rotating structure are in the same coordinate system. The device is also provided with a transmission structure for uploading the measurement data captured by the laser radar and the rotation angle data of the laser radar in real time.

[0029] In some embodiments, the horizontal rotation range of the rotating structure is 0° to 180°, and one end of the laser radar in the horizontal rotation range is set as the initial posture of the laser radar. A sensor is provided in the initial posture to detect whether the laser radar is in the initial posture.

[0030] In a third aspect, an embodiment of the present application provides an electronic device, comprising a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, implements any of the methods described above.

[0031] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method as described in any one of the above items is implemented.

[0032] According to the above content, the technical solution of the present invention has the following beneficial effects compared with the prior art:

[0033] 1. This invention uses single-point rotating 3D mapping of a laser radar, which can complete rapid 3D scene mapping of a large area and high precision while occupying less structural space.

[0034] 2. By matching point cloud data and rotation angles according to timestamps, using a filtering algorithm to process the point cloud data, detecting and eliminating abnormal point cloud data, and fusing and splicing point cloud data with different timestamps based on the rotation angle, the filtered point cloud data avoids erroneous modeling results caused by multiple reflections of the LiDAR spot and solves the ghosting problem in traditional laser modeling, effectively improving the clarity and accuracy of 3D mapping. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0036] Figure 1 is a flowchart of a laser radar 3D reconstruction method according to an embodiment of the present application;

[0037] Figure 2 is a schematic structural diagram of a laser radar 3D reconstruction device according to an embodiment of the present application;

[0038] Figure 3 It is a structural block diagram of an electronic device according to an embodiment of the present application.

[0039] Explanation of the accompanying drawings: 1. LiDAR; 2. Rotating structure; 3. Mounting bracket; 4. Sensor. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solutions and advantages of this application more clearly understood, the present application is described and illustrated below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely used to explain this application and are not intended to limit this application. Based on the embodiments provided in this application, all other embodiments obtained by those of ordinary skill in the art without making any creative efforts are within the scope of protection of this application.

[0041] Obviously, the drawings described below are merely examples or embodiments of the present application. Those skilled in the art can, without inventive effort, apply the present application to other similar scenarios based on these drawings. Furthermore, it is also understood that, although the effort involved in such a development process may be complex and lengthy, for those skilled in the art related to the content disclosed in this application, changes in design, manufacturing, or production based on the technical content disclosed in this application are merely conventional technical means and should not be construed as an insufficiency of the content disclosed in this application.

[0042] References to "embodiments" in this application mean that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it refer to independent or alternative embodiments that are mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described in this application may be combined with other embodiments unless there is a conflict.

[0043] Unless otherwise defined, the technical or scientific terms used in this application should have the ordinary meaning understood by a person of ordinary skill in the technical field to which this application belongs. The words "one", "a", "the" and the like used in this application do not indicate a limit on quantity and may indicate the singular or plural. The terms "include", "comprise", "have" and any variations thereof used in this application are intended to cover non-exclusive inclusions; for example, a process, method, system, product or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units that are not listed, or may also include other steps or units that are inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like used in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The word "multiple" used in this application means greater than or equal to two. "And / or" describes the association relationship of associated objects, indicating that three relationships can exist. For example, "A and / or B" can mean: A exists alone, A and B exist at the same time, and B exists alone. The terms "first", "second", "third" and the like involved in this application are merely used to distinguish similar objects and do not represent a specific ordering of the objects.

[0044] The purpose of the embodiments of the present application is to provide a single-point rotational three-dimensional reconstruction technology based on laser radar to solve the problems of existing laser radar occupying a large operating space during three-dimensional reconstruction, easily producing erroneous modeling results due to multiple reflections of the light spot, and the occurrence of ghosting, resulting in unclear and inaccurate modeling.

[0045] This technology compares the timestamps of the radar measurement data measured by the lidar scanning and the rotation angle data obtained by the rotation structure of the lidar, pairs the point cloud data and angle data corresponding to the compared timestamps within the allowable error range, filters the point cloud data to eliminate abnormal data, and fuses the uneliminated point cloud data with the matching angle data to obtain a clear and accurate three-dimensional reconstructed structure based on the initial posture of the lidar.

[0046] Figure 1 is a flow chart of a laser radar 3D reconstruction method according to an embodiment of the present application. Figure 1 As shown, the laser radar 3D reconstruction method includes the following steps:

[0047] Step S1, performing a three-dimensional reconstruction scan of the work scene using a laser radar, and obtaining the current rotation angle data and radar measurement data of the laser radar in real time; the rotation angle data includes angle data and corresponding timestamp data, and the radar measurement data includes point cloud data and corresponding timestamp data;

[0048] Step S2, comparing the timestamp data in the radar measurement data with the timestamp data in the rotation angle data, and screening out matching point cloud data and angle data based on the time difference between the two to obtain lidar data with precise angular position;

[0049] In step S3, the point cloud data filtered out in step S2 is parsed and processed by the callback function, and the coordinates of the parsed point cloud data are converted, and the point cloud data in different time series are spliced ​​and fused according to their rotation angles.

[0050] In order to more clearly illustrate the present application, specific examples are given below and elaborated in detail.

[0051] Step S1, perform three-dimensional reconstruction scanning of the work scene through the laser radar, and obtain the current rotation angle data and radar measurement data of the laser radar in real time; the rotation angle data includes angle data and corresponding timestamp data, and the radar measurement data includes point cloud data and corresponding timestamp data.

[0052] As an example, the laser radar is installed in the middle position of the two mechanical arms of the working robot in a horizontally rotatable manner through a rotating structure. The rotation angle range of the rotating structure is set to 0°~180°. The laser radar is preset to face 0° as the initial position of the laser radar and recorded. The sensor is used in conjunction with the preset program to determine whether the laser radar is in the preset initial position.

[0053] Furthermore, before performing 3D reconstruction, it is determined whether the LiDAR is in the preset initial posture:

[0054] If the LiDAR is in the preset initial position, it will start to perform rotation measurement, and obtain and upload the current rotation angle data and LiDAR measurement data in real time;

[0055] If the laser radar is not in the preset initial position, the laser radar is controlled to rotate to the preset initial position through the rotation structure, and then the laser radar starts to perform rotation measurement, and obtains and uploads the current rotation angle data and radar measurement data in real time.

[0056] The radar measurement data includes point cloud data and timestamp data matching the point cloud data, and the rotation angle data includes angle data and timestamp data matching the angle data. The rotation angle data is uploaded at a frequency of 50 Hz.

[0057] Among them, the laser radar is connected to the rotating structure through a mounting bracket, and the origin position of the laser radar and the center of the rotating structure are in the same coordinate system.

[0058] Furthermore, the rotation axis of the laser radar is coaxially arranged with the rotation axis of the rotating structure, and the rotation axis of the rotating structure is perpendicular to the horizontal plane.

[0059] Preferably, the sensor can be a photoelectric switch or a proximity switch arranged at one end of the initial posture to determine whether the laser radar is in the initial posture.

[0060] Preferably, the laser radar is a 16-line 360° mechanical laser radar.

[0061] Preferably, the two robotic arms of the working robot are respectively arranged on the left and right sides, with the direction facing one of the robotic arms of the working robot being set to 0°, and the direction facing the other robotic arm of the working robot being set to 180°. When the field of view window of the laser radar is facing the 0° direction, the laser radar is in its initial position.

[0062] As an example, the data structure of a single measurement point in point cloud data is designed as follows:

[0063] Struct point{

[0064] double x;

[0065] double y;

[0066] double z;

[0067] double idensity;

[0068] int ring;

[0069] bool save;

[0070] }

[0071] Among them, x, y, z are the distances between the laser emitter and the target point in three-dimensional space along the x-axis, y-axis, and z-axis, respectively; indensity is the laser reflection intensity; ring is the laser emitter number; and save indicates whether to save the point.

[0072] Step S2, compare the timestamp data in the radar measurement data and the timestamp data in the rotation angle data, and based on the time difference between the two, filter out the matching point cloud data and angle data to obtain the lidar data with precise angular position.

[0073] As an example, the acquired rotation angle data and radar measurement data are stored in a queue in the order of data acquisition time, and the data acquired first is stored at the front of the queue; starting from the head of the queue, the rotation angle data is extracted from the rotation angle data queue and the radar measurement data is extracted from the radar measurement data queue in sequence, and the corresponding timestamp data is parsed from the extracted rotation angle data and radar measurement data, the difference between the two timestamp data is calculated, and the difference is compared with a preset first threshold.

[0074] Furthermore, when the difference is compared with a preset first threshold, if the difference is less than the first threshold, the angle data of the current timestamp is used as the rotation angle of the current point cloud data;

[0075] If the difference is greater than the first threshold, continue to determine the size of the timestamp data in the radar measurement data and the timestamp data in the rotation angle data:

[0076] If the timestamp of the radar measurement data is less than the timestamp of the rotation angle data, the timestamp of the radar measurement data is considered to be the earlier time data, and this set of radar measurement data is discarded. The next set of radar measurement data is extracted from the radar measurement data queue, and the timestamp data difference is recalculated and the above judgment is performed;

[0077] If the timestamp of the rotation angle data is less than the timestamp of the radar measurement data, the timestamp of the rotation angle data is considered to be the earlier time data, and this set of rotation angle data is discarded. The next set of rotation angle data is extracted from the rotation angle data queue, the timestamp data difference is recalculated, and the above judgment is performed.

[0078] Save the filtered matched point cloud data and angle data to obtain lidar data with precise angle position.

[0079] When comparing timestamp data between two sets of data, the one with the smaller timestamp is discarded. This is because data queues are arranged in ascending chronological order, and data extracted for comparison is extracted in queue order. Therefore, the timestamp data after the queue must be larger than the timestamp data before the queue. If the larger timestamp is discarded, the timestamp of the next set of data extracted will result in a larger difference between the timestamps of the rotation angle data and the radar measurement data than the previous set, which would not meet the judgment purpose.

[0080] In step S3, the point cloud data filtered out in step S2 is parsed and processed by the callback function, and the coordinates of the parsed point cloud data are converted, and the point cloud data in different time series are spliced ​​and fused according to their rotation angles.

[0081] As an example, the callback function includes point cloud filtering and point cloud coordinate conversion.

[0082] In the application scenarios of working robots, the light spots emitted by the laser radar may be reflected multiple times, resulting in errors in laser ranging. This 3D reconstruction method analyzes the raw data of the laser radar, splits the ranging point cloud, and filters the point cloud data with a discrete point filtering algorithm to eliminate abnormal measurement points.

[0083] As an example, the point cloud data filtered in step S2 is parsed and divided into 16 groups of point cloud files according to the ring value (laser transmitter number) of the point cloud data, and the point cloud data in each group of point cloud files are arranged in order according to the angle value of 0 to 360 degrees;

[0084] Traverse all the point cloud data in the 16 groups of point cloud files, and when traversing the point cloud files, further segment the point cloud data in each group of point cloud files by the two constraints of the preset angle value and the preset number of adjacent points to obtain the local point cloud set N k ={p i ,......,p n}.

[0085] Specifically, different angle values ​​and numbers of adjacent points may be used to further differentiate the point cloud data according to different application scenarios.

[0086] Furthermore, we calculate the set N k The standard deviation s,

[0087] in, Represents the local point cloud set N kThe spatial distance from point i to the lidar origin (0, 0, 0); (x, y, z) is the three-dimensional coordinate of point i; n is the local point cloud set N k The number of point clouds within μ k is the preset local point cloud set N k Mean value of inner point cloud distance measurement;

[0088] Compare the standard deviation s with a preset second threshold,

[0089] If the standard deviation s is greater than the preset second threshold, the local point cloud set N is determined k If there are abnormal measurement points in the area, a filtering operation is performed to remove the abnormal measurement points;

[0090] If the standard deviation s is less than the preset second threshold, the local point cloud set N is determined k If there are no abnormal measurement points in the area, other local point cloud sets will be determined until all point cloud data in the 16 groups of point cloud files are traversed.

[0091] Furthermore, when the standard deviation s is greater than a preset second threshold, the abnormal measurement points are eliminated:

[0092] For the local point cloud set N k The points inside are judged, and the judged point is point i. The judgment formula is as follows:

[0093]

[0094] Among them, ThrAbnormal is the preset second threshold; p .i Represents the local point cloud set N k The spatial distance from point i to the origin of the laser radar; p i.save =false means p .i Save as false; p i.save =true means p .i Save is true;

[0095] The judgment process is:

[0096] If the local point cloud set N k The spatial distance p from point i to the origin of the laser radar .i Less than or equal to the preset point cloud distance mean μ k , then it is determined that there is no abnormality at this point p .save =true, save the point;

[0097] If the local point cloud set N k The spatial distance p from point i to the origin of the laser radar .i Greater than the preset point cloud distance mean μ k , then the point is determined to be an outlier point p.save =false, remove the point.

[0098] Furthermore, the filtered point cloud data is rotated and translated using a rotation and translation matrix according to the matching angle data and timestamp data to obtain a three-dimensional reconstruction result based on the initial posture state, and the result after multi-frame fusion is saved.

[0099] Specifically, since the rotating structure rotates only around one axis, the rotation and translation only perform rotation and translation on the corresponding axis.

[0100] Preferably, if there is a certain error during the installation phase, manual calibration is used to calibrate the initial posture. The calibration method uses laser point cloud data in the initial posture state to scan areas with obvious features such as planes and corners, and manually level the collected point cloud data. The calibration parameters are used as fixed parameters, stored and used in the program.

[0101] Step S4: determine whether there is still unprocessed data in the radar measurement data queue. If so, repeat the above steps.

[0102] In step S5, a three-dimensional reconstruction result based on the rotating structure is obtained according to the above steps, and a statistical filtering algorithm is applied to remove the burr area of ​​the point cloud in the reconstruction result to obtain relatively smooth reconstructed point cloud data.

[0103] Furthermore, according to the spatial mapping relationship between the laser center point and the center position of the robotic arm, the point cloud result obtained in step S5 is subjected to translation transformation, and finally the entire 3D reconstruction task is completed.

[0104] Therefore, based on the above, by comparing the timestamps of the real-time uploaded radar measurement data and rotation angle data, we generate matching point cloud data and angle data based on the timestamps. Furthermore, we use a callback function to filter the point cloud data and remove abnormal data. Compared with traditional methods, this method not only improves the overall accuracy of the 3D reconstruction results and prevents the impact of abnormal data on the modeling results, but also solves the problem of smearing during LiDAR 3D reconstruction.

[0105] It should be noted that the steps shown in the above process or the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0106] An embodiment of the present application also provides a laser radar three-dimensional reconstruction device, which is used to implement the above-mentioned embodiments and preferred implementation methods, and will not be repeated here.

[0107] Figure 2 is a schematic structural diagram of a laser radar 3D reconstruction device according to an embodiment of the present application. Figure 2 As shown, the device includes a laser radar 1 and a rotating structure 2. The laser radar 1 can be horizontally rotated and measured through the rotating structure 2 and is installed in the middle position of the two mechanical arms of the working robot. The rotation angle range of the rotating structure 2 is 0° to 180°. The laser radar 1 is preset to face 0° as the initial posture of the laser radar 1. A sensor 4 is provided at the initial posture to detect whether the laser radar 1 is in the initial posture.

[0108] Among them, the laser radar 1 is connected to the rotating structure 2 through the mounting bracket 3. The origin position of the laser radar 1 and the center of the rotating structure 2 are in the same coordinate system. The device is also provided with a transmission structure for uploading the measurement data captured by the laser radar 1 and the rotation angle data of the laser radar 1 in real time.

[0109] Furthermore, the rotation axis of the laser radar 1 is coaxially arranged with the rotation axis of the rotating structure 2, and the rotation axis of the rotating structure 2 is perpendicular to the horizontal plane.

[0110] Preferably, the sensor 4 cooperates with a preset program to determine whether the laser radar 1 is in a preset initial posture.

[0111] Preferably, the sensor 4 can be a photoelectric switch or a proximity switch arranged at one end of the initial posture to determine whether the laser radar 1 is in the initial posture.

[0112] Preferably, the laser radar 1 uses a 16-line 360° mechanical laser radar.

[0113] Preferably, the two robotic arms of the working robot are respectively arranged on the left and right sides, with the direction facing one of the robotic arms of the working robot being set to 0°, and the direction facing the other robotic arm of the working robot being set to 180°. When the field of view window of the laser radar 1 is facing the 0° direction, the laser radar 1 is in its initial position.

[0114] Preferably, the rotating structure 2 can also be installed on the working robot via a slide, so that the device can realize scanning measurement while rotating along the slide.

[0115] In summary, the embodiment of the present application is based on mechanical multi-line laser radar three-dimensional reconstruction technology. It drives the multi-line laser radar to perform scanning measurement through a rotating structure, obtains and uploads radar measurement data and rotation angle data, and filters out matching point cloud data and angle data through timestamp comparison. The filtered point cloud data is filtered and abnormal points are removed. The point cloud data that is not removed is rotated, translated, and fused to obtain the three-dimensional reconstruction result.

[0116] In addition, an embodiment of the present application further provides an electronic device, Figure 3 is a structural block diagram of an electronic device according to an embodiment of the present application, such as Figure 3 As shown, the electronic device includes a processor 81 and a memory 82 storing computer program instructions.

[0117] Specifically, the processor 81 may include a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0118] Among them, the memory 82 may include a large-capacity memory for data or instructions. By way of example and not limitation, the memory 82 may include a hard disk drive (HDD), a floppy disk drive, a solid-state drive (SSD), a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory 82 may include a removable or non-removable (or fixed) medium. Where appropriate, the memory 82 may be inside or outside the data processing device. In a specific embodiment, the memory 82 is a non-volatile memory. In a specific embodiment, the memory 82 includes a read-only memory (ROM) and a random access memory (RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (Programmable Read-Only Memory, PROM for short), an erasable PROM (Erasable Programmable Read-Only Memory, EPROM for short), an electrically erasable PROM (Electrically Erasable Programmable Read-Only Memory, EEPROM for short), an electrically alterable ROM (Electrically Alterable Read-Only Memory, EAROM for short) or a flash memory (FLASH) or a combination of two or more of these. Under appropriate circumstances, the RAM can be a static random access memory (SRAM) or a dynamic random access memory (DRAM), where the DRAM can be a fast page mode dynamic random access memory (FPMDRAM), an extended data output dynamic random access memory (EDODRAM), a synchronous dynamic random access memory (SDRAM), etc.

[0119] The memory 82 may be used to store or cache various data files required for processing and / or communication, as well as possible computer program instructions executed by the processor 81 .

[0120] The processor 81 implements any one of the lidar three-dimensional reconstruction methods in the above embodiments by reading and executing computer program instructions stored in the memory 82.

[0121] In some embodiments, the electronic device may further include a communication interface 83 and a bus 80. Figure 3 As shown, the processor 81, the memory 82, and the communication interface 83 are connected via a bus 80 and communicate with each other.

[0122] The communication interface 83 is used to implement communication between the various modules, devices, units, and / or devices in the embodiments of the present application. The communication interface 83 can also implement data communication with other components such as: external devices, image / data acquisition equipment, databases, external storage, and image / data processing workstations.

[0123] The bus 80 includes hardware, software, or both, and couples the components of the electronic device to each other. The bus 80 includes, but is not limited to, at least one of the following: a data bus, an address bus, a control bus, an expansion bus, and a local bus. By way of example, and not limitation, bus 80 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses, or a combination of two or more of these. Bus 80 may include one or more buses, where appropriate. Although embodiments of the present application describe and illustrate a particular bus, this application contemplates any suitable bus or interconnect.

[0124] In addition, in conjunction with the lidar 3D reconstruction method in the above embodiments, embodiments of the present application may provide a computer-readable storage medium for implementation. The computer-readable storage medium stores computer program instructions; when the computer program instructions are executed by a processor, any of the lidar 3D reconstruction methods in the above embodiments is implemented.

[0125] Those skilled in the art should understand that the various technical features of the above-described embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the various technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0126] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A laser radar three-dimensional reconstruction method, characterized in that: The method comprises: Step S1, performing a three-dimensional reconstruction scan of the work scene using a laser radar, and obtaining the current rotation angle data and radar measurement data of the laser radar in real time; the rotation angle data includes angle data and corresponding timestamp data, and the radar measurement data includes point cloud data and corresponding timestamp data; Step S2, comparing the timestamp data in the radar measurement data with the timestamp data in the rotation angle data, and screening out matching point cloud data and angle data based on the time difference between the two to obtain lidar data with accurate angle pose; Step S3, parsing and processing the point cloud data filtered out in step S2 through the callback function: further segmenting the filtered point cloud data to obtain a local point cloud set; Calculate the spatial distance from the points in the local point cloud set to the lidar origin, and calculate the standard deviation of the local point cloud set in combination with the preset point cloud distance mean in the local point cloud set; Compare the standard deviation with a preset second threshold, If the standard deviation is greater than the preset second threshold, the points in the local point cloud set are judged. If the spatial distance from a point in the local point cloud set to the lidar origin is less than or equal to the preset point cloud distance mean, the point is judged to be normal and saved. If the spatial distance from a point in the local point cloud set to the lidar origin is greater than the preset point cloud distance mean, the point is judged to be an abnormal point and removed. If the standard deviation is less than the preset second threshold, it is determined that there are no abnormal measurement points in the local point cloud collection area, and other local point cloud collections are determined at this time; The coordinates of the analyzed point cloud data are transformed, and the point cloud data in different time series are spliced ​​and fused according to their rotation angles.

2. The method according to claim 1, characterized in that In step S2, the rotation angle data and radar measurement data are stored in a queue, and the rotation angle data and radar measurement data are extracted in order from the head of the queue, and the difference between the timestamp data of the two sets of data is analyzed and compared, and the difference is compared with a preset first threshold. If the difference is less than the first threshold, the angle data of the current timestamp is used as the rotation angle of the current point cloud data; If the difference is greater than the first threshold, continue to judge the size of the timestamp data in the radar measurement data and the timestamp data in the rotation angle data, discard the group of data with smaller timestamp data values ​​in the radar measurement data and the rotation angle data, and select the next group of data from the discarded data queue in sequence to compare the difference in timestamp data between the two with the data that has not been discarded, and compare the difference with the preset first threshold.

3. The method according to claim 1, characterized in that In step S1, before the laser radar operates, it is determined whether the laser radar is in a preset initial posture; If it is in the preset initial posture, the laser radar performs rotation measurement to obtain the current rotation angle data and radar measurement data in real time; If it is not in the preset initial posture, the laser radar is controlled to rotate to the preset initial posture and then perform rotation measurement to obtain the current rotation angle data and radar measurement data in real time.

4. The method according to claim 1, wherein In step S3, during the coordinate conversion processing of the point cloud data, the radar measurement data after the point cloud filtering processing is subjected to rotation and translation transformation according to the rotation angle data matched therewith to obtain a three-dimensional reconstruction result based on the initial posture state, and the obtained result is saved.

5. The method according to claim 1, wherein: The invention relates to a laser radar three-dimensional reconstruction device, which includes a laser radar and a rotating structure. The rotating structure can rotate the laser radar in the horizontal direction for measurement. The origin position of the laser radar and the center of the rotating structure are in the same coordinate system. The device is also provided with a transmission structure for uploading the measurement data captured by the laser radar and the rotation angle data of the laser radar in real time. The horizontal rotation range of the rotating structure is 0°~180°. The end of the laser radar in the horizontal rotation range is set as the initial posture of the laser radar. A sensor for detecting whether the laser radar is in the initial posture is provided in the initial posture.

6. An electronic device comprising a processor and a memory storing a computer program, characterized in that: When the computer program is executed by the processor, the method according to any one of claims 1 to 4 is implemented.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images