A recognition method and lidar for traffic control

Abstract

The application relates to a recognition method and a laser radar for traffic control, the method comprising the following steps: dividing a scanning area into multiple rectangular scanning areas in response to an obtained starting signal or a cooperation signal; giving multiple scanning lines to each rectangular area, and the multiple scanning lines belonging to the same rectangular area are arranged in parallel; performing multiple laser scanning at positions where the scanning lines are located, and the angle and direction of the scanning laser remain unchanged during the laser scanning process; obtaining N diffraction echoes of each scanning laser in a time sequence; deleting the diffraction echoes before and after a constraint time length; and constructing independent feature objects according to the remaining diffraction echoes. The recognition method and the laser radar for traffic control disclosed by the application reduce the interference of the detection laser in rainy and low-light conditions through a combined recognition mode of fuzzy recognition and directional feature recognition, and improve the recognition accuracy in rainy and low-light conditions.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to an identification method and a lidar for traffic control. Background Technology

[0002] The purpose of traffic control is to guide vehicles to travel safely and quickly on motor vehicle lanes. However, with the increase in the number of motor vehicles and roads, the complexity and difficulty of traffic control are also increasing, because it is necessary to coordinate the traffic flow at various intersections and take into account the usage needs of motor vehicles, non-motor vehicles and pedestrians.

[0003] To address this issue, traffic control at intersections is shifting from timed passage to dynamic passage, meaning passage time is automatically adjusted based on traffic flow to achieve higher throughput. This traffic control method uses devices such as cameras and lidar to collect information, which is then analyzed in the cloud to provide control recommendations.

[0004] However, in rainy weather, the imaging interference factors of the camera (rain, vehicle lights, ground reflection) increase sharply. At this time, it is necessary to rely on LiDAR to collect more vehicle information. However, the detection light emitted by LiDAR will also be interfered with, resulting in a large amount of interference information in the generated 3D point cloud data, so that accurate information cannot be obtained. Summary of the Invention

[0005] This application provides an identification method and a lidar for traffic control. By combining fuzzy recognition and directional feature recognition, the method reduces interference with the detection laser under rainy and low-light conditions, thereby improving the accuracy of identification under these conditions.

[0006] The above-mentioned objective of this application is achieved through the following technical solution:

[0007] Firstly, this application provides an identification method, including:

[0008] In response to the acquired start signal or coordination signal, the scanning area is divided to obtain multiple rectangular scanning areas;

[0009] Assign multiple scan lines to each rectangular area, and set multiple scan lines belonging to the same rectangular area in parallel;

[0010] Multiple laser scans are performed at the location of the scan line, and the angle and direction of the scanning laser remain unchanged during the laser scan process;

[0011] For each scanning laser beam, obtain its Nth diffraction echo in time series, where N≥1 and is a natural number;

[0012] The diffraction echoes before and after the constraint time length are deleted; and

[0013] Construct independent feature objects based on the remaining diffraction echoes.

[0014] In one possible implementation of the first aspect, multiple scan lines belonging to the same rectangular region move intermittently during the scanning process;

[0015] The scan line forms an angle with any axis of the rectangular region.

[0016] In one possible implementation of the first aspect, the scan line moves on a horizontal plane and the direction of movement of the scan line is perpendicular to its own length direction.

[0017] One possible implementation of the first aspect also includes:

[0018] Convert the location points corresponding to the diffraction echoes within the constrained time length into coordinate points in three-dimensional space;

[0019] By filtering coordinate points in three-dimensional space using standard shapes, a cluster of coordinate points is obtained; and

[0020] Calculate the distance between adjacent coordinate point clusters and merge adjacent coordinate point clusters when the distance is less than a set distance.

[0021] In one possible implementation of the first aspect, the filtering method includes:

[0022] Construct a standard circle with the coordinate points as its centers;

[0023] Project all the standard circles within the standard shape onto a horizontal plane;

[0024] Calculate the ratio of the total projected area within the standard shape to the area of ​​the standard shape; and

[0025] When the ratio is greater than or equal to the set ratio, the portion of the 3D point cloud data is treated as a coordinate point cluster.

[0026] In one possible implementation of the first aspect, when the distance between two clusters of coordinate points is greater than the maximum required distance, it includes:

[0027] A reference constraint time length is given based on the constraint time length of the previous coordinate point cluster and the constraint time length of the next coordinate point cluster. The reference constraint time length is located between the corresponding constraint time lengths of the two coordinate point clusters.

[0028] Take any point between two clusters of coordinate points as the scanning reference point;

[0029] The exploration scan line is constructed based on the scanning reference point, which serves as the rotation center of the exploration scan line;

[0030] Obtain the diffraction echo generated based on the probe scan line and convert the position points corresponding to the diffraction echo into coordinate points in three-dimensional space;

[0031] The rotation path of the probe scan line is used to generate a filtering range, and this filtering range is used to filter the coordinate points generated from the diffraction echoes based on the probe scan line; and

[0032] Feature surfaces are created using selected coordinate points. When a feature surface exists, it is assumed that there are additional feature objects between two clusters of coordinate points.

[0033] In one possible implementation of the first aspect, the axis of rotation of the probe scan line is the line connecting the scan reference point and the signal source;

[0034] The process of generating the filter range includes:

[0035] The exploration surface is generated using the rotation path of the exploration scan line;

[0036] The probe surface is moved along the line connecting the scanning reference point and the signal source; and

[0037] The spatial region between the two exploration surfaces is used as the filtering range.

[0038] Secondly, this application provides an identification device, comprising:

[0039] The division unit is used to divide the scanning area in response to the acquired start signal or coordination signal, resulting in multiple rectangular scanning areas;

[0040] The assignment unit is used to assign multiple scan lines to each rectangular area, with multiple scan lines belonging to the same rectangular area set in parallel;

[0041] The scanning unit is used to perform multiple laser scans at the location of the scanning line. During the laser scanning process, the angle and orientation of the scanning laser remain unchanged.

[0042] The receiving unit is used to obtain N diffraction echoes in the time series for each scanning laser beam, where N≥1 and is a natural number;

[0043] The first filtering unit is used to delete diffraction echoes before and after the constraint time length; and

[0044] Object building units are used to construct independent feature objects based on the remaining diffraction echoes.

[0045] Thirdly, this application provides a lidar for traffic control, the lidar comprising:

[0046] Signal transmitting system and signal receiving system;

[0047] One or more memories for storing instructions; and

[0048] One or more processors are configured to retrieve and execute the instructions from the memory to drive the signal transmitting system and the signal receiving system to perform the methods described in the first aspect and any possible implementation thereof.

[0049] Fourthly, this application provides a computer-readable storage medium, the computer-readable storage medium comprising:

[0050] The program, when run by a processor, is executed as described in the first aspect and any possible implementation thereof.

[0051] Fifthly, this application provides a computer program product, including program instructions that, when run by a computing device, execute the method described in the first aspect and any possible implementation thereof.

[0052] Sixthly, this application provides a chip system including a processor for implementing the functions involved in the foregoing aspects, such as generating, receiving, transmitting, or processing the data and / or information involved in the foregoing methods.

[0053] This chip system can consist of chips or include chips and other discrete components.

[0054] In one possible design, the chip system also includes a memory for storing necessary program instructions and data. The processor and the memory can be decoupled and located on different devices, connected via wired or wireless means, or the processor and the memory can be coupled to the same device. Attached Figure Description

[0055] Figure 1 This is a flowchart illustrating the steps of an identification method provided in this application.

[0056] Figure 2 This is a schematic diagram of the division of a scanning area provided in this application.

[0057] Figure 3 This is a schematic diagram showing the relative position of a scan line and a rectangular region provided in this application.

[0058] Figure 4This is a schematic diagram illustrating the principle of generating multiple diffraction echoes, as provided in this application.

[0059] Figure 5 This is a schematic diagram illustrating the principle of screening diffraction echoes provided in this application.

[0060] Figure 6 This is a schematic diagram illustrating the principle of filtering coordinate points in three-dimensional space, as provided in this application.

[0061] Figure 7 This is a flowchart illustrating the specific steps for filtering 3D point cloud data provided in this application.

[0062] Figure 8 This is a schematic diagram illustrating the principle of filtering three-dimensional point cloud data based on area ratio, as provided in this application.

[0063] Figure 9 This is a schematic diagram of a method for merging clusters of adjacent coordinate points provided in this application.

[0064] Figure 10 This is a schematic diagram illustrating the principle of generating a probe scan line provided in this application.

[0065] Figure 11 This is a schematic diagram illustrating the principle of generating a filtering range, as provided in this application. Detailed Implementation

[0066] The technical solutions in this application will be further described in detail below with reference to the accompanying drawings.

[0067] This application discloses an identification method applied to LiDAR at intersections. The purpose of this method is to improve the recognition accuracy of LiDAR in rainy conditions. In terms of laser wavelength, the two wavelengths currently used are 905nm and 1550nm. The 1550nm wavelength has a higher safety threshold for the human eye, which allows for the emission of higher laser power to achieve higher ranging sensitivity. The 1550nm laser has begun to gradually replace the 905nm laser. However, in rainy environments, the 1550nm LiDAR still has certain difficulties in identifying targets because raindrops in the air will cause a large amount of interference data in the three-dimensional point cloud data. As a result, the amount of three-dimensional point cloud data generated based on target recognition is much smaller than the amount of three-dimensional point cloud data generated based on interference factors such as raindrops.

[0068] Please see Figure 1 To address this problem, this application discloses an identification method comprising the following steps:

[0069] S101, in response to the acquired start signal or coordination signal, divide the scanning area to obtain multiple rectangular scanning areas;

[0070] S102, assign multiple scan lines to each rectangular area, and set multiple scan lines belonging to the same rectangular area in parallel;

[0071] S103, Perform multiple laser scans at the location of the scan line, and keep the angle and direction of the scanning laser unchanged during the laser scan process;

[0072] S104, for each scanning laser beam, obtain its Nth diffraction echo in time series, where N≥1 and is a natural number;

[0073] S105, delete the diffraction echoes before and after the constraint time length; and

[0074] S106, construct independent feature objects based on the remaining diffraction echoes.

[0075] Specifically, in step S101, the LiDAR receives a start signal or a cooperation signal. This signal is emitted from the cloud or an edge coordination control center located at a traffic intersection. Upon receiving the start signal or cooperation signal, the LiDAR first divides the scanning area into multiple rectangular scanning areas, such as... Figure 2 As shown, the purpose of dividing the scanning area is to reduce interference when obtaining feature objects by using rectangular scanning areas.

[0076] It should be understood that in practical application scenarios, cars at intersections are arranged in lane order. The purpose of dividing the scanning area into multiple rectangular scanning areas is to pre-fix the position of the feature object through area division. The purpose of pre-fixing is to facilitate the improvement of the recognition accuracy of the feature object in the later fuzzy recognition process.

[0077] Because in the process of fuzzy recognition, feature objects are identified by the features of partial regions on the feature object. If the distance between two partial regions is close, there is a possibility that the two partial regions will be merged. This will lead to errors in the calculation of the number of feature objects and give incorrect guidance to the final decision.

[0078] In step S102, multiple scan lines are assigned to each rectangular region. Multiple scan lines belonging to the same rectangular region are set in parallel. The function of these scan lines is to generate three-dimensional point cloud data within their respective rectangular regions. The purpose of using scan lines is to reduce the amount of interference data generated.

[0079] It should be understood that during normal scanning, LiDAR uses a large-area scanning method to generate 3D point cloud data. However, in rainy environments, the generated 3D point cloud data will contain a large amount of interference data (due to the presence of raindrops). To reduce the impact of raindrops, this application uses scan lines to limit the scanning area of ​​the LiDAR.

[0080] Scan lines can significantly reduce the amount of generated 3D point cloud data, thus suppressing the generation of interference data. However, this also reduces the total amount of 3D point cloud data, so fuzzy recognition is needed to obtain feature objects.

[0081] In step S103, multiple laser scans are performed at the location of the scan line. During the laser scan, the angle and orientation of the scanning laser remain unchanged. The purpose of this method is to ensure that the laser emitted by the lidar can reach the feature object as much as possible.

[0082] It should be understood that the propagation of laser light in the air is affected by raindrops. Each time a laser encounters a raindrop, diffraction occurs, which changes the direction of laser propagation. This causes the actual contact position between the laser and the target object to deviate from the intended contact position, or even prevents the laser from making contact with the target object at all.

[0083] Performing multiple laser scans at the location of the scan line can increase the probability of the laser coming into contact with the feature object. This is because there are gaps between raindrops in the air, and the probability of a laser beam reaching the feature object can be increased during multiple laser scans at the same location.

[0084] In step S104, the lidar receives the diffraction echoes generated by the scanning laser. For each scanning laser beam, the lidar obtains N diffraction echoes in its time series, where N ≥ 1 and is a natural number. The diffraction echoes are as follows: Figure 4 As shown, when a lidar comes into contact with a raindrop or a feature object, it will generate diffraction echoes. If these diffraction echoes are considered as data, then the diffraction echoes generated by contact with raindrops are interference data, while the diffraction echoes generated by contact with feature objects are valid data.

[0085] In subsequent processing, valid data needs to be filtered out from the data. This filtering process is performed in step S105. Please refer to [link / reference]. Figure 5 In step S105, the diffraction echoes before and after the constraint time length are deleted, and the data generated corresponding to the deleted diffraction echoes are interference data.

[0086] It should be understood that since the location of the lidar is fixed and the height of the vehicle is also within a certain range, the data can be filtered by obtaining the diffraction echo time at a fixed time. The remaining data includes valid data and some interference data, while the data that is filtered out includes a large amount of interference data and a certain amount of valid data.

[0087] Finally, step S106 is executed. In this step, independent feature objects are constructed based on the remaining diffraction echoes. These feature objects are generated based on valid data and a small amount of interference data, and are used for traffic control. The specific control process involves changing the original fixed time interval to a dynamic time interval. That is, the amount of vehicles gathering at the intersection is dynamically correlated with the interval of the traffic lights. When the amount of vehicles gathering reaches the required level, the traffic lights switch to a green light state. The time interval between two green light states (red light) is dynamically adjusted according to the amount of vehicles gathering.

[0088] In rainy conditions, drivers' visibility is impaired, reducing both acceleration and speed. Fixed time intervals can cause traffic congestion at intersections, decreasing traffic flow efficiency. To improve traffic efficiency, it's necessary to determine the number of vehicles at each intersection and then dynamically adjust traffic lights based on that number.

[0089] Rainy weather limits the image clarity provided by cameras, making it more difficult to count vehicles, especially when combined with low light intensity (nighttime). Furthermore, vehicles often turn on their lights, and the reflections from these lights can blur parts of the image. Thermal imaging is also unusable due to obstructions.

[0090] This application uses LiDAR scanning to count the number of vehicles. After obtaining the feature objects in step S106, the number of feature objects is counted to obtain the number of vehicles in the LiDAR coverage area.

[0091] In some cases, multiple scan lines belonging to the same rectangular region move intermittently during the scan, and the scan lines form an angle with any axis of the rectangular region. Figure 3 (As shown). Intermittent movement allows sufficient time for multiple scans along a single scan line. Compared to continuous movement, this method reduces the amount of 3D point cloud data while also increasing the concentration of the 3D point cloud data. This increased concentration leads to lower data processing volume and faster data processing speed.

[0092] At this point, both interference data and valid data will be concentrated, and the concentration of valid data means that a larger amount of valid data can be obtained. The more valid data at a given location, the more lasers can come into contact with the feature object, resulting in a more accurate description of the feature object.

[0093] The presence of an angle between the scan line and any axis of the rectangular area can increase the contact length between the laser and the feature object, which also helps to obtain more effective data.

[0094] In some possible implementations, the scan line moves on a horizontal plane and the direction of movement of the scan line is perpendicular to its own length direction.

[0095] In some examples, the following steps have been added:

[0096] S201 converts the position points corresponding to the diffraction echoes within the constrained time length into coordinate points in three-dimensional space.

[0097] S202, using standard shapes to filter coordinate points in three-dimensional space, resulting in a cluster of coordinate points; and

[0098] S203, calculate the distance between adjacent coordinate point clusters and perform fusion processing on adjacent coordinate point clusters when the distance is less than the set distance.

[0099] Steps S201 to S203 involve processing the diffraction echoes located within the constrained time length. Specifically, the processing involves displaying the location of the diffraction echoes in three-dimensional space using coordinates.

[0100] Please see Figure 6 After transferring the diffraction echoes within the constrained time length into three-dimensional space, each diffraction echo will generate a coordinate. When some coordinates are generated based on a feature object, these coordinates will cluster at one location, while the coordinates generated by the interference data will be scattered in space.

[0101] For coordinate points in three-dimensional space, a standard shape is used for filtering, resulting in a cluster of coordinate points. Here, the standard shape refers to a cube. Figure 6 As shown, the cube has length, width, and height, and the screening process involves the cube moving within three-dimensional space.

[0102] The cube represents the screening precision. When some coordinates fall within the cube and meet the requirements, these coordinate points are considered to form a coordinate point cluster. Considering a practical scenario, the 3D point cloud data generated by LiDAR during scanning can potentially result in a coordinate point cluster on any surface of a vehicle. However, the shape, position, and spatial orientation of these coordinate point clusters in 3D space are difficult to determine and involve many possibilities.

[0103] To address this issue, this application uses a time constraint to filter 3D point cloud data. This filtering concentrates the location of 3D point cloud data near the vehicle roof, which can be considered a plane or a combination of a plane and a curved surface. The problem of different vehicle heights is solved using a cube. A cube has a certain height, which allows it to retain 3D point cloud data within a specific height region before further processing.

[0104] Please see Figure 7 The specific filtering method for 3D point cloud data is as follows:

[0105] S301, construct a standard circle with the coordinate point as the center;

[0106] S302, Project all standard circles within the standard shape onto a horizontal plane;

[0107] S303, calculate the ratio of the total projected area within the standard shape to the area of ​​the standard shape; and

[0108] S304: When the ratio is greater than or equal to the set ratio, the part of the 3D point cloud data is treated as a coordinate point cluster.

[0109] Steps S301 to S304 involve filtering the 3D point cloud data using area ratios. Since the dimensions of the standard shapes mentioned earlier are fixed, standard circles are constructed centered on the coordinate points. These standard circles are then projected onto a plane. When the ratio of the total projected area within the standard shapes to the area of ​​the standard shapes meets the requirement, this portion of the 3D point cloud data is treated as a cluster of coordinate points. Figure 8 As shown.

[0110] Here, the dispersion of coordinate points and data precision also need to be considered. For a single feature object, it may have multiple clusters of coordinate points. Therefore, it is necessary to calculate the distance between adjacent clusters of coordinate points and merge them when the distance is less than a set distance. Figure 9 As shown, this corresponds to step S203. After the fusion process, the blank areas between the two coordinate point clusters are filled with standard data.

[0111] When the distance between two clusters of coordinate points is greater than the maximum required distance, the following steps are performed:

[0112] S401, a reference constraint time length is given based on the constraint time length of the previous coordinate point cluster and the constraint time length of the next coordinate point cluster. The reference constraint time length is located between the corresponding constraint time lengths of the two coordinate point clusters.

[0113] S402, take any point between two coordinate point clusters as the scanning reference point;

[0114] S403, construct the exploration scan line based on the scanning reference point, with the scanning reference point serving as the rotation center of the exploration scan line;

[0115] S404, acquire the diffraction echo generated based on the probe scan line and convert the position points corresponding to the diffraction echo into coordinate points in three-dimensional space;

[0116] S405, using the rotation path of the probe scan line to generate a filtering range, and using the filtering range to filter the coordinate points generated based on the diffraction echo generated by the probe scan line; and

[0117] S406 uses selected coordinate points to create feature surfaces. When a feature surface exists, it is assumed that there are additional feature objects between two clusters of coordinate points.

[0118] Specifically, at this point, it is assumed that there is a missing region between these two clusters of coordinate points with a spacing greater than the maximum required distance. A missing region implies the possibility of a feature object existing in that region, but further confirmation is needed. Please refer to [link / reference]. Figure 10 Specifically, a scanning reference point is first determined between two coordinate point clusters (step S402), and then a rotational scan is performed using the line connecting the scanning reference point and the lidar (signal source) as the axis. This rotational scan can be considered as scanning the side of the vehicle.

[0119] The scan results are processed between steps S404 and S406, and the process is the same as described above, so it will not be repeated here. During the processing, feature surfaces (the sides of the vehicle) are obtained. When a feature surface exists, it is assumed that an additional feature object exists between the two coordinate point clusters. This additional feature object is considered to be located between the two feature objects.

[0120] Please see Figure 11 In step S405, the rotation path of the probe scan line needs to be used to generate the filtering range. The process of generating the filtering range is as follows:

[0121] The exploration surface is generated using the rotation path of the exploration scan line;

[0122] The probe surface is moved along the line connecting the scanning reference point and the signal source; and

[0123] The spatial region between the two exploration surfaces is used as the filtering range.

[0124] Specifically, a plane is obtained based on the probe scan line. This plane is then moved back and forth along the line connecting the scan reference point and the signal source. During the movement, there are two extreme positions. When the probe surface is located at these two extreme positions, the spatial area between the two probe surfaces is used as the filtering range. The purpose of the filtering range is also to filter the generated 3D point cloud data, with the aim of reducing the proportion of interference data in the data obtained in steps S401 to S406.

[0125] The length of the probe scan line also needs to be limited here. Its length is 60%-80% of the minimum distance between two coordinate point clusters. The purpose of this limitation is to reduce the coverage area of ​​the filtering range, thereby reducing the amount of data that needs to be processed.

[0126] This application also provides an identification device, including:

[0127] The division unit is used to divide the scanning area in response to the acquired start signal or coordination signal, resulting in multiple rectangular scanning areas;

[0128] The assignment unit is used to assign multiple scan lines to each rectangular area, with multiple scan lines belonging to the same rectangular area set in parallel;

[0129] The scanning unit is used to perform multiple laser scans at the location of the scanning line. During the laser scanning process, the angle and orientation of the scanning laser remain unchanged.

[0130] The receiving unit is used to obtain N diffraction echoes in the time series for each scanning laser beam, where N≥1 and is a natural number;

[0131] The first filtering unit is used to delete diffraction echoes before and after the constraint time length; and

[0132] Object building units are used to construct independent feature objects based on the remaining diffraction echoes.

[0133] Furthermore, multiple scan lines belonging to the same rectangular area move intermittently during the scanning process;

[0134] The scan line forms an angle with any axis of the rectangular region.

[0135] Furthermore, the scan line moves on a horizontal plane and the direction of movement of the scan line is perpendicular to its own length direction.

[0136] Furthermore, it also includes:

[0137] The coordinate point transformation unit is used to convert the position points corresponding to the diffraction echoes within the constrained time length into coordinate points in three-dimensional space.

[0138] The second filtering unit is used to filter coordinate points in three-dimensional space using standard shapes to obtain a cluster of coordinate points; and

[0139] The fusion unit is used to calculate the distance between adjacent coordinate point clusters and to perform fusion processing on adjacent coordinate point clusters when the distance is less than a set distance.

[0140] Furthermore, it also includes:

[0141] The first processing unit is used to construct a standard circle with the coordinate point as the center.

[0142] The second processing unit is used to project all the standard circles within the standard shape onto a horizontal plane.

[0143] A calculation unit is used to calculate the ratio of the total projected area within the standard shape to the area of ​​the standard shape; and

[0144] The cutoff unit is used to treat a portion of the 3D point cloud data as a coordinate point cluster when the ratio is greater than or equal to a set ratio.

[0145] Furthermore, it also includes:

[0146] The third processing unit is used to provide a reference constraint time length based on the constraint time length of the previous coordinate point cluster and the constraint time length of the next coordinate point cluster. The reference constraint time length is located between the corresponding constraint time lengths of the two coordinate point clusters.

[0147] The first selection unit is used to select any point between two coordinate point clusters as the scanning reference point;

[0148] The building unit is used to construct the exploration scan line based on the scanning reference point, which is the rotation center of the exploration scan line;

[0149] The fourth processing unit is used to acquire the diffraction echo generated based on the probe scan line and convert the position points corresponding to the diffraction echo into coordinate points in three-dimensional space;

[0150] The third filtering unit is used to generate a filtering range using the rotation path of the probe scan line and to filter the coordinate points generated by the diffraction echo based on the probe scan line using the filtering range; and

[0151] The identification unit is used to create feature surfaces using filtered coordinate points. When a feature surface exists, it is assumed that there is an additional feature object between two clusters of coordinate points.

[0152] Furthermore, the axis of rotation of the probe scan line is the line connecting the scan reference point and the signal source.

[0153] Furthermore, it also includes:

[0154] A generation unit is used to generate an exploration surface using the rotation path of the exploration scan line;

[0155] The moving unit is used to move the probe surface along the line connecting the scanning reference point and the signal source; and

[0156] The second selection unit is used to select the spatial region between two exploration surfaces as the filtering range.

[0157] In one example, the unit in any of the above devices may be one or more integrated circuits configured to implement the above methods, such as one or more application-specific integrated circuits (ASICs), or one or more digital signal processors (DSPs), or one or more field-programmable gate arrays (FPGAs), or a combination of at least two of these integrated circuit forms.

[0158] For example, when the units in the device can be implemented through a processing element scheduler, the processing element can be a general-purpose processor, such as a central processing unit (CPU) or other processor capable of calling programs. Alternatively, these units can be integrated together to form a system-on-a-chip (SOC).

[0159] In this application, various objects such as messages / information / devices / network elements / systems / apparatus / actions / operations / processes / concepts may be named. It is understood that these specific names do not constitute a limitation on the relevant objects. The names may be changed depending on the scenario, context, or usage habits. The understanding of the technical meaning of the technical terms in this application should be mainly determined from their functions and technical effects embodied / performed in the technical solution.

[0160] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0161] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0162] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0163] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0164] It should also be understood that in the various embodiments of this application, the terms "first," "second," etc., are merely to indicate that multiple objects are different. For example, a first time window and a second time window are only to indicate different time windows. They should not have any effect on the time windows themselves, and the aforementioned terms "first," "second," etc., should not impose any limitations on the embodiments of this application.

[0165] It should also be understood that, in the various embodiments of this application, unless otherwise specified or in case of logical conflict, the terms and / or descriptions between different embodiments are consistent and can be referenced by each other, and the technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationships.

[0166] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a computer-readable storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned computer-readable storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0167] This application also provides a computer program product including instructions that, when executed, cause the terminal device and the network device to perform operations corresponding to the methods described above.

[0168] This application also provides a lidar for traffic control, the lidar comprising:

[0169] Signal transmitting system and signal receiving system;

[0170] One or more memories for storing instructions; and

[0171] One or more processors are configured to retrieve and execute the instructions from the memory to drive the signal transmitting system and the signal receiving system to perform the methods described above.

[0172] This application also provides a chip system including a processor for implementing the functions involved in the above description, such as generating, receiving, transmitting, or processing the data and / or information involved in the above methods.

[0173] This chip system can consist of chips or include chips and other discrete components.

[0174] The processor mentioned above can be a CPU, a microprocessor, an ASIC, or one or more integrated circuits that execute a program to control the method of transmitting the feedback information described above.

[0175] In one possible design, the chip system also includes a memory for storing necessary program instructions and data. The processor and the memory can be decoupled and located on different devices, connected via wired or wireless means to support the chip system in implementing the various functions described in the above embodiments. Alternatively, the processor and the memory can also be coupled to the same device.

[0176] Optionally, the computer instructions are stored in memory.

[0177] Optionally, the memory can be a storage unit within the chip, such as a register or cache. Alternatively, the memory can be a storage unit located outside the chip within the terminal, such as a ROM or other types of static storage devices that can store static information and instructions, such as RAM.

[0178] It is understood that the memory in this application may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory.

[0179] Non-volatile memory can be ROM, programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory.

[0180] Volatile memory can be RAM, which is used as an external cache. There are many different types of RAM, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus random access memory.

[0181] The embodiments described in this specific implementation are preferred embodiments of this application and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A recognition method, characterized in that, include: In response to the acquired start signal or coordination signal, the scanning area is divided to obtain multiple rectangular scanning areas; Assign multiple scan lines to each rectangular area, and set multiple scan lines belonging to the same rectangular area in parallel; Multiple laser scans are performed at the location of the scan line, and the angle and direction of the scanning laser remain unchanged during the laser scan process; For each scanning laser beam, obtain its Nth diffraction echo in time series, where N≥1 and is a natural number; The diffraction echoes before and after the constraint time length are deleted. Construct independent feature objects based on the remaining diffraction echoes; Also includes: Convert the location points corresponding to the diffraction echoes within the constrained time length into coordinate points in three-dimensional space; By using standard shapes to filter coordinate points in three-dimensional space, a cluster of coordinate points is obtained; Calculate the distance between adjacent coordinate point clusters and merge adjacent coordinate point clusters when the distance is less than a set distance; Filtering methods include: Construct a standard circle with the coordinate points as its centers; Project all the standard circles within the standard shape onto a horizontal plane; Calculate the ratio of the total projected area within the standard shape to the area of ​​the standard shape; When the ratio is greater than or equal to the set ratio, the 3D point cloud data located within the standard shape will be treated as a cluster of coordinate points. When the distance between two clusters of coordinate points is greater than the maximum required distance, including: A reference constraint time length is given based on the constraint time length of the previous coordinate point cluster and the constraint time length of the next coordinate point cluster. The reference constraint time length is located between the corresponding constraint time lengths of the two coordinate point clusters. Take any point between two clusters of coordinate points as the scanning reference point; The exploration scan line is constructed based on the scanning reference point, which serves as the rotation center of the exploration scan line; Obtain the diffraction echo generated based on the probe scan line and convert the position points corresponding to the diffraction echo into coordinate points in three-dimensional space; The rotation path of the probe scan line is used to generate a filtering range, and the filtering range is used to filter the coordinate points generated by the diffraction echo based on the probe scan line. Feature surfaces are created using selected coordinate points. When a feature surface exists, it is assumed that there are additional feature objects between two clusters of coordinate points.

2. The identification method according to claim 1, characterized in that, Multiple scan lines belonging to the same rectangular area move intermittently during the scanning process; The scan line forms an angle with any axis of the rectangular region.

3. The identification method according to claim 2, characterized in that, The scan line moves on a horizontal plane and the direction of movement of the scan line is perpendicular to its own length direction.

4. The identification method according to claim 1, characterized in that, The axis of rotation of the probe scan line is the line connecting the scan reference point and the signal source; The process of generating the filter range includes: The exploration surface is generated using the rotation path of the exploration scan line; The probe surface is moved along the line connecting the scanning reference point and the signal source; The spatial region between the two exploration surfaces is used as the filtering range.

5. An identification device, characterized in that, include: The division unit is used to divide the scanning area in response to the acquired start signal or coordination signal, resulting in multiple rectangular scanning areas; The assignment unit is used to assign multiple scan lines to each rectangular area, with multiple scan lines belonging to the same rectangular area set in parallel; The scanning unit is used to perform multiple laser scans at the location of the scanning line. During the laser scanning process, the angle and orientation of the scanning laser remain unchanged. The receiving unit is used to obtain N diffraction echoes in the time series for each scanning laser beam, where N≥1 and is a natural number; The first filtering unit is used to delete diffraction echoes before and after the constraint time length; Object building units are used to construct independent feature objects based on the residual diffraction echoes; Also includes: Convert the location points corresponding to the diffraction echoes within the constrained time length into coordinate points in three-dimensional space; By using standard shapes to filter coordinate points in three-dimensional space, a cluster of coordinate points is obtained; Calculate the distance between adjacent coordinate point clusters and merge adjacent coordinate point clusters when the distance is less than a set distance; Filtering methods include: Construct a standard circle with the coordinate points as its centers; Project all the standard circles within the standard shape onto a horizontal plane; Calculate the ratio of the total projected area within the standard shape to the area of ​​the standard shape; When the ratio is greater than or equal to the set ratio, the 3D point cloud data located within the standard shape will be treated as a cluster of coordinate points. When the distance between two clusters of coordinate points is greater than the maximum required distance, including: A reference constraint time length is given based on the constraint time length of the previous coordinate point cluster and the constraint time length of the next coordinate point cluster. The reference constraint time length is located between the corresponding constraint time lengths of the two coordinate point clusters. Take any point between two clusters of coordinate points as the scanning reference point; The exploration scan line is constructed based on the scanning reference point, which serves as the rotation center of the exploration scan line; Obtain the diffraction echo generated based on the probe scan line and convert the position points corresponding to the diffraction echo into coordinate points in three-dimensional space; The rotation path of the probe scan line is used to generate a filtering range, and the filtering range is used to filter the coordinate points generated by the diffraction echo based on the probe scan line. Feature surfaces are created using selected coordinate points. When a feature surface exists, it is assumed that there are additional feature objects between two clusters of coordinate points.

6. A lidar for traffic control, characterized in that, The lidar includes: Signal transmitting system and signal receiving system; One or more memories for storing instructions; and One or more processors are configured to retrieve and execute the instructions from the memory to drive the signal transmitting system and the signal receiving system to perform the method as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes: The program, when run by the processor, executes the method as described in any one of claims 1 to 4.

Images