WO2023015407A1 - 一种伪像点识别方法、终端设备及计算机可读存储介质 - Google Patents

一种伪像点识别方法、终端设备及计算机可读存储介质 Download PDF

Info

Publication number
WO2023015407A1
WO2023015407A1 PCT/CN2021/111483 CN2021111483W WO2023015407A1 WO 2023015407 A1 WO2023015407 A1 WO 2023015407A1 CN 2021111483 W CN2021111483 W CN 2021111483W WO 2023015407 A1 WO2023015407 A1 WO 2023015407A1
Authority
WO
WIPO (PCT)
Prior art keywords
point
statistical
points
artifact
analysis area
Prior art date
Application number
PCT/CN2021/111483
Other languages
English (en)
French (fr)
Inventor
宋妍
Original Assignee
深圳市速腾聚创科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳市速腾聚创科技有限公司 filed Critical 深圳市速腾聚创科技有限公司
Priority to CN202180100803.2A priority Critical patent/CN117677862A/zh
Priority to PCT/CN2021/111483 priority patent/WO2023015407A1/zh
Publication of WO2023015407A1 publication Critical patent/WO2023015407A1/zh

Links

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01SRADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/02Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
    • G01S13/50Systems of measurement based on relative movement of target

Definitions

  • the present application belongs to the technical field of laser radar, and in particular relates to a false image point identification method, a terminal device and a computer-readable storage medium.
  • lidar Due to its high resolution, high sensitivity, strong anti-interference ability, and unaffected by dark conditions, lidar is often used in areas such as autonomous driving, logistics vehicles, robots, and public smart transportation.
  • the laser beam Since the laser beam is irrational, its light spot has a certain area.
  • two echoes When there are two target objects that are close to each other in the same direction, two echoes will be generated when the light spot hits the junction of the two objects.
  • the distance between the two target objects is relatively short, so the aliasing effect of the two echoes will be generated, resulting in artifact points in the received point cloud image between the two target objects, that is, the drag point phenomenon, the drag point phenomenon Existence can lead to errors in the distance determination of the target object.
  • the prior art has the problem of being unable to effectively identify whether the point cloud data is a real object or a drag point artifact.
  • the embodiment of the present application provides a method for identifying artifact points, a terminal device, and a computer-readable storage medium to solve the problem that the existing technology cannot effectively identify whether the point cloud data is a real object or a drag point artifact point. question.
  • the embodiment of the present application provides a false image point identification method, including:
  • the method before determining the statistical results of the points satisfying the preset statistical conditions in the data analysis area according to the distance difference value of each point, the method further includes:
  • the determining the statistical results of points satisfying preset statistical conditions in the data analysis area according to the distance difference value of each point includes:
  • the differential distance value of each point in the first statistical result the number of points whose distance difference between two consecutive points is not equal or does not satisfy the proportional relationship is counted, and recorded as the fifth statistical result;
  • variable flag is set.
  • the identifying the artifact points in the point cloud data based on the statistical results includes:
  • the statistical results are judged based on the identification conditions, and artifact points in the point cloud data are determined.
  • the center point of the data analysis area is determined to be an artifact point.
  • the first statistical result is greater than a first statistical threshold
  • the second statistical result is greater than a second statistical threshold
  • the fifth statistical result is greater than a third statistical threshold
  • the The variable flag is set, the third statistical result or the fourth statistical result is less than or equal to the fourth statistical threshold, and the previous point of the current point is an artifact point, then it is determined that the center point of the data analysis area is an artifact point.
  • the artifact point identification method further includes:
  • the artifact points are removed from the point cloud data.
  • the embodiment of the present application provides a terminal device, including:
  • the determination unit is used to slide and traverse each point in the point cloud data returned by the radar, and determine the data analysis area of each point;
  • a statistical unit configured to determine statistical results of points satisfying preset statistical conditions in the data analysis area according to the distance difference value of each point;
  • An identifying unit configured to identify artifact points in the point cloud data based on the statistical results.
  • an embodiment of the present application provides a terminal device, the terminal device includes a processor, a memory, and a computer program stored in the memory and operable on the processor, and the processor executes the The computer program implements the method described in the first aspect or any optional manner of the first aspect.
  • an embodiment of the present application provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, any operable solution in the first aspect or the first aspect can be implemented. method as described in the selection method.
  • an embodiment of the present application provides a computer program product, which, when the computer program product is run on a terminal device, causes the terminal device to execute the method described in the first aspect or any optional manner of the first aspect.
  • the statistical results of the points that meet the preset statistical conditions are calculated based on the distance difference results of the point cloud, and then the artifact points in the point cloud data are accurately identified based on the statistical results, which can effectively It can accurately identify the artifact points in the point cloud data, and solve the problem that the artifact points in the point cloud data cannot be effectively identified at present.
  • FIG. 1 is a schematic diagram of a scene where the light spot emitted by the radar hits the junction of two objects provided by the embodiment of the present application;
  • Fig. 2 is a schematic diagram of the echoes when the light spot emitted by the radar hits the junction of two objects provided by the embodiment of the present application;
  • FIG. 3 is a schematic flowchart of a method for identifying artifact points provided by an embodiment of the present application
  • FIG. 4 is a schematic diagram of the implementation flow of S12 provided by the embodiment of the present application.
  • Fig. 5 is a schematic diagram of artifact points formed by the drag point phenomenon provided by the embodiment of the present application.
  • Fig. 6 is a comparison diagram of the point cloud image containing artifact points provided by the embodiment of the present application and the point cloud image after removing the artifact points through the artifact point identification method provided by the embodiment of the present application;
  • FIG. 7 is a schematic structural diagram of a terminal device provided in an embodiment of the present application.
  • FIG. 8 is a schematic structural diagram of a terminal device provided by another embodiment of the present application.
  • FIG. 9 is a schematic structural diagram of a computer-readable storage medium provided by an embodiment of the present application.
  • references to “one embodiment” or “some embodiments” or the like described in the specification of the present application mean that a particular feature, structure or characteristic described in connection with the embodiment is included in one or more embodiments of the present application .
  • appearances of the phrases “in one embodiment,” “in some embodiments,” “in other embodiments,” “in other embodiments,” etc. in various places in this specification are not necessarily All refer to the same embodiment, but mean “one or more but not all embodiments” unless specifically stated otherwise.
  • the terms “including”, “comprising”, “having” and variations thereof mean “including but not limited to”, unless specifically stated otherwise.
  • Lidar is an automatic remote sensing device that uses a laser as a source of emission and uses photoelectric detection technology for detection.
  • Lidar can include a transmitting system, a receiving system, a scanning control system, and a data processing system.
  • the working principle of the laser radar is to transmit a detection signal to the target object. After the detection signal hits the target object, the target object will reflect the detection signal to form an echo signal.
  • the receiving system can receive the echo signal and analyze the received signal. The echo signal is processed to obtain the distance, size, speed, reflectivity and other information of the target object.
  • the artifact points in the point cloud data cannot be effectively identified. If the artifact points are not processed, the false point cloud (artifact points) will lead to the formation of false target objects, resulting in ranging errors. If the artifacts are excessively removed If the point cloud of the real slope object is misjudged, it will also affect the distance measurement result.
  • the embodiment of the present application traverses each point in the point cloud data by sliding, counts the statistical results of the points satisfying the preset statistical conditions based on the distance difference results of the point cloud, and then accurately identifies the point cloud based on the statistical results
  • the artifact points in the data can effectively identify the artifact points in the point cloud data, and solve the problem that the artifact points in the point cloud data cannot be effectively identified at present.
  • the method for identifying artifact points provided in the embodiments of the present application will be described in detail below:
  • FIG. 3 is a schematic flowchart of a method for identifying artifact points provided by an embodiment of the present application.
  • the execution subject of the false image point recognition method provided in the embodiment of the present application is a terminal device connected to the laser radar in communication.
  • the above-mentioned terminal device can be a mobile terminal such as a smart phone, a tablet computer, or a wearable device, or it can be a computer, cloud server, radar auxiliary computer, etc.
  • the false image point identification method provided in the embodiment of the present application is applicable to any form of lidar products, including but not limited to mechanical lidar, MEMS lidar, FLASH lidar, OPA lidar, etc.
  • the method for identifying artifact points may include S11 ⁇ S13, which are described in detail as follows:
  • S11 Slidingly traverse each point in the point cloud data returned by the radar, and determine the data analysis area of each point.
  • the terminal device realizes the identification of each point by acquiring the point cloud data uploaded by the radar, and then performing sliding traversal on each point in the point cloud data.
  • the radar system after the radar system receives the echo, it can upload the received data to a terminal device connected to it, and the terminal device analyzes the point cloud data.
  • the radar system may transmit the received data to the radar data processing module, and output the data after processing. It can be understood that the data processing module can be integrated in the radar control system.
  • the data analysis area may be determined based on the horizontal neighborhood and/or the vertical neighborhood.
  • the horizontal neighborhood refers to a horizontal data analysis area determined with the current point as the center point. For example, if the current point data is in the first row and the radius of the data analysis area is determined to be L/2, the horizontal neighborhood of data analysis can be It is an area of 1 row and L columns; the vertical neighborhood refers to a vertical data analysis area determined with the current point as the center point, for example, if the current data is in the first column, and the radius of the data analysis area is determined to be L/2, Then the vertical neighborhood of data analysis can be an area with L rows and 1 column, where L is a positive integer greater than 1, where the radius of the horizontal data analysis area and the radius of the longitudinal analysis area can be the same or different, without limitation here.
  • the point neighborhood is (m, n-floor(L/2):n+floor(L/ 2)).
  • the area of the point can be (m-floor(L/2) : m+floor(L/2),n-floor(L/2):n+floor(L/2)).
  • taking each point in the point cloud data as the center point can determine its corresponding data analysis area, and analyze whether the point is an artifact point based on the data analysis area.
  • the data analysis area of each point can be a single horizontal neighborhood (that is, the data analysis area of the point is only the horizontal neighborhood), a single vertical neighborhood (that is, the data analysis area of the point is only the vertical neighborhood). Domain), horizontal neighborhood and vertical neighborhood (that is, the data analysis area of this point includes both horizontal neighborhood and vertical neighborhood), without limitation here.
  • S12 Determine statistical results of points satisfying preset statistical conditions in the data analysis area according to the distance difference value of each point.
  • the preset statistical conditions can be set according to the actual situation, for example, if the absolute value of the difference value of each point is set to be greater than a specific threshold, if the preset statistical conditions are met, the point will be recorded in the first statistical result .
  • the distance difference value of each point in the data analysis area can be calculated.
  • the distance difference value described in the embodiment of the present application refers to the difference between the distance of the point at the next moment and the distance at the current moment, and the distance difference value of each point can be expressed as formula (1).
  • Diff(i) is the distance difference value of point i
  • Dist(i+1) is the distance of point i at the next moment
  • Dist(i) is the distance of point i at the current moment.
  • the above S12 may include S21-S25.
  • the data analysis area is taken as an example of a horizontal neighborhood for illustration.
  • the statistical result is recorded as the first statistical result Cnt1, and record the position IndexList (record coordinate value) of the point satisfying the condition and its distance difference value DiffList.
  • the judgment condition is shown in the following formula (2), where Diff represents the difference result of each point, Dist represents the distance value of the current point, ⁇ represents the angular resolution of the lidar, and ⁇ represents the angle judgment threshold, with a typical value of 0.2rad .
  • the first statistical result Cnt1 is obtained by weighting the judgment result of the formula (2) and the position of the current point in the horizontal neighborhood.
  • the weighted weight of each point in the data analysis area can be determined based on the distance from the central point, for example, the closer the distance to the central point, the greater the weighted weight; the farther the distance from the central point, the smaller the weighted weight.
  • the calculation of the weighting coefficient of each point can be adjusted according to the actual situation, for example, the weighting coefficient obeys linear decrease, Gaussian decrease, etc.
  • the coordinate number of the point can be determined based on the position IndexList of the point in S21, and then it is judged whether the coordinates of each point are continuous. If the coordinates are continuous, the point is counted as a point with continuous coordinates, and recorded in the second statistical result Cnt2.
  • the distance difference value DiffList in S21 the number of points whose distance increases in the judgment list, that is, the number of Diff greater than 0, is recorded as the third statistical result Cnt3; according to the distance difference value DiffList in S21 The number of points whose distance decreases in the judgment list, that is, the number of points whose Diff is less than 0, is recorded as the fourth statistical result Cnt4.
  • the distance difference value DiffList in S21 determine whether the distance difference between two consecutive points in the table is equal or satisfies the proportional relationship, and counts the number of points whose distance difference between two consecutive points is not equal or does not satisfy the proportional relationship , recorded as the fifth statistical result Cnt5.
  • the two consecutive points refer to two points whose coordinates are adjacent.
  • the first and last points of the data analysis area refer to the first point and the last point of the data analysis area
  • the first and last points of the data analysis area can be determined based on the coordinates of each point in the data analysis area. And can calculate the distance difference between the first and last two points.
  • first preset threshold and the above-mentioned second preset threshold may be set according to actual applications, which are not limited in this application.
  • variable flag Logic1 can be set to 1 or 0. For example, when the distance difference between the first and last points of the data analysis area is greater than the first preset threshold and smaller than the second preset threshold, the variable flag Logic1 set to 1.
  • the above S13 may include the following steps:
  • the statistical results are judged based on the identification conditions, and artifact points in the point cloud data are determined.
  • Recognition accuracy refers to the strength of the removal of artifact points. If the removal strength of artifact points needs to be strengthened, the recognition accuracy should be set higher; if the removal strength of artifact points needs to be reduced, the recognition accuracy should be set lower.
  • multiple joint recognition conditions can be set, and only when the multiple recognition conditions are satisfied, the center point of the data analysis area is determined as the artifact point.
  • the recognition conditions When setting the recognition conditions, it can be set based on the statistical results of S12. If the recognition accuracy is high, the recognition conditions can be set based on multiple statistical results. If the recognition accuracy is low, it can be set based on only one or two statistical results. The recognition condition can be set based on the actual situation in the application, which will not be repeated in this application.
  • the above identification condition may be that if the first statistical result is greater than the first statistical threshold, the second statistical result is greater than the second statistical threshold, the fifth statistical result is greater than the third statistical threshold, and the variable identification is set, and the distance difference monotonically increases or decreases, then it is determined that the center point of the data analysis area is an artifact point, otherwise it is determined that the center point of the data analysis area is not an artifact point.
  • the above identification condition may be that if the first statistical result is greater than the first statistical threshold, the second statistical result is greater than the second statistical threshold, the fifth statistical result is greater than the third statistical threshold, and the variable identification Set, the third statistical result or the fourth statistical result is less than or equal to the fourth statistical threshold, and the previous point of the current point is an artifact point, then it is determined that the center point of the data analysis area is an artifact point, Otherwise, it is determined that the central point of the data analysis area is not an artifact point.
  • the artifact point recognition method provided by the embodiment of the present application can traverse each point in the point cloud data by sliding, and calculate the statistical results of the points satisfying the preset statistical conditions based on the distance difference result of the point cloud. Then based on the statistical results, the artifact points in the point cloud data are accurately identified.
  • the above-mentioned artifact point identification method may further include the following steps:
  • the artifact points are removed from the point cloud data.
  • the point identifier Invalid of the artifact point may be set to 1 to identify the point as an artifact point.
  • Fig. 6 shows a point cloud image containing the artifact point (the position pointed by the arrow is the artifact point) and through the embodiment of the application
  • the provided artifact point recognition method removes the point cloud image after artifact points are removed.
  • the artifact point recognition method provided by the embodiment of the present application can effectively remove the artifact point in the point cloud image, and well retain the point cloud of the real object, effectively improving the distance measurement. precision.
  • the embodiments of the present invention further provide embodiments of a terminal device that implements the foregoing method embodiments.
  • FIG. 7 is a schematic structural diagram of a terminal device provided in an embodiment of the present application.
  • each unit included in the terminal device is configured to execute each step in the embodiment corresponding to FIG. 3 .
  • the terminal device 70 includes: a determination unit 71 , a statistics unit 72 and an identification unit 73 . in:
  • the determination unit 71 is configured to perform sliding traverse on each point in the point cloud data returned by the radar, and determine the data analysis area of each point.
  • the statistical unit 72 is used for determining the statistical results of the points satisfying the preset statistical conditions in the data analysis area according to the distance difference value of each point.
  • the identifying unit 73 is configured to identify artifact points in the point cloud data based on the statistical results.
  • the above-mentioned terminal device may further include a calculation unit, and the above-mentioned calculation unit is configured to calculate a distance difference value of each point in the data analysis area.
  • the recognition unit includes a condition setting unit and a judgment unit, wherein: the condition setting unit is used to set the recognition condition according to the recognition accuracy; the judgment unit is used to judge the statistical result based on the recognition condition, The artifact points in the point cloud data are determined.
  • the above statistical unit 72 is specifically used for weighted statistics of points in the data analysis area whose distance difference value is greater than a specific threshold, and record it as the first statistical result; according to each point in the first statistical result The number of points with continuous coordinates of the statistical coordinates of the statistics is recorded as the second statistical result; according to the difference distance value of each point in the first statistical result, the number of points with increased distance and the number of points with reduced distance are counted, And record as the third statistical result and the fourth statistical result respectively; According to the differential distance value of each point in the first statistical result, count the number of points where the distance difference between two consecutive points is not equal or does not satisfy the proportional relationship, and record It is the fifth statistical result; if the distance difference between the first and last points of the data analysis area is greater than the first preset threshold and smaller than the second preset threshold, then the variable flag is set.
  • Fig. 8 is a schematic structural diagram of a terminal device provided by another embodiment of the present application.
  • the terminal device 8 provided in this embodiment includes: a processor 80 , a memory 81 , and a computer program 82 stored in the memory 81 and operable on the processor 80 , such as an image segmentation program.
  • the processor 80 executes the computer program 82, it realizes the steps in the above-mentioned embodiments of the method for identifying false image points, such as S11-S13 shown in FIG. 3 .
  • the processor 80 executes the computer program 82, it realizes the functions of the modules/units in the above terminal device embodiments, for example, the functions of the units 71-73 shown in FIG. 7 .
  • the computer program 82 can be divided into one or more modules/units, and the one or more modules/units are stored in the memory 81 and executed by the processor 80 to complete the application .
  • the one or more modules/units may be a series of computer program instruction segments capable of accomplishing specific functions, and the instruction segments are used to describe the execution process of the computer program 82 in the terminal device 8 .
  • the computer program 82 may be divided into a first acquisition unit and a first processing unit. For specific functions of each unit, please refer to the relevant description in the embodiment corresponding to FIG. 5 , which will not be repeated here.
  • the terminal device may include, but not limited to, a processor 80 and a memory 81 .
  • FIG. 8 is only an example of the terminal device 8, and does not constitute a limitation to the terminal device 8. It may include more or less components than those shown in the figure, or combine certain components, or different components. , for example, the terminal device may also include an input and output device, a network access device, a bus, and the like.
  • the so-called processor 80 can be a central processing unit (Central Processing Unit, CPU), and can also be other general-purpose processors, a digital signal processor (Digital Signal Processor, DSP), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), off-the-shelf programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • a general-purpose processor may be a microprocessor, or the processor may be any conventional processor, or the like.
  • the storage 81 may be an internal storage unit of the terminal device 8 , such as a hard disk or memory of the terminal device 8 .
  • the memory 81 may also be an external storage device of the terminal device 8, such as a plug-in hard disk equipped on the terminal device 8, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, flash memory card (Flash Card), etc. Further, the memory 81 may also include both an internal storage unit of the terminal device 8 and an external storage device.
  • the memory 81 is used to store the computer program and other programs and data required by the terminal device.
  • the memory 81 can also be used to temporarily store data that has been output or will be output.
  • FIG. 9 is a schematic structural diagram of a computer-readable storage medium provided by an embodiment of the present application. As shown in FIG. During execution, the above-mentioned artifact point identification method can be realized.
  • An embodiment of the present application provides a computer program product.
  • the terminal device can implement the above method for identifying artifact points when executed.
  • Module completion means that the internal structure of the terminal device is divided into different functional units or modules, so as to complete all or part of the functions described above.
  • Each functional unit and module in the embodiment may be integrated into one processing unit, or each unit may exist separately physically, or two or more units may be integrated into one unit, and the above-mentioned integrated units may adopt hardware It can also be implemented in the form of software functional units.

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Image Analysis (AREA)
  • Optical Radar Systems And Details Thereof (AREA)

Abstract

一种伪像点识别方法、终端设备及计算机可读存储介质,适用于激光雷达技术领域,包括:对雷达回传的点云数据中的每个点进行滑动遍历,确定每个点的数据分析区域;根据每个点的距离差分值确定所述数据分析区域中满足预设统计条件的点的统计结果;基于所述统计结果识别出所述点云数据中的伪像点,能够有效地识别出点云数据中的伪像点,解决目前无法有效地识别出点云数据中的伪像点的问题。

Description

一种伪像点识别方法、终端设备及计算机可读存储介质 技术领域
本申请属于激光雷达技术领域,尤其涉及一种伪像点识别方法、终端设备及计算机可读存储介质。
背景技术
激光雷达由于其分辨率高、灵敏度高、抗干扰能力强,不受黑暗条件影响等优势,常用于自动驾驶、物流车、机器人、公共智慧交通等领域。
由于激光雷达的光束是非理性化的,其光斑具有一定的面积,当同一方位上存在距离较近的两个目标物体时,光斑打到两个物体的交界处会产生两个回波,由于两个目标物体的距离较近,因此会产生两个回波的混叠效应,导致接收到的点云图像在两个目标物体之间的位置出现伪像点,即拖点现象,拖点现象的存在会导致目标物体的距离测定出现错误。
而现有技术存在无法有效识别出点云数据是真实物体还是拖点伪像点的问题。
技术问题
有鉴于此,本申请实施例提供了一种伪像点识别方法、终端设备及计算机可读存储介质,以解决现有技术存在无法有效识别出点云数据是真实物体还是拖点伪像点的问题。
技术解决方案
第一方面,本申请实施例提供一种伪像点识别方法,包括:
对雷达回传的点云数据中的每个点进行滑动遍历,确定每个点的数据分析区域;
根据每个点的距离差分值确定所述数据分析区域中满足预设统计条件的点的统计结果;
基于所述统计结果识别出所述点云数据中的伪像点。
在第一方面的一种实现方式中,在根据每个点的距离差分值确定所述数据分析区域中满足预设统计条件的点的统计结果之前,还包括:
计算所述数据分析区域中每个点的距离差分值。
在第一方面的一种实现方式中,所述根据每个点的距离差分值确定所述数据分析区域中满足预设统计条件的点的统计结果,包括:
加权统计所述数据分析区域中距离差分值大于特定阈值的点,并记录为第一统计结果;
根据所述第一统计结果中各个点的坐标位置统计坐标连续的点的个数,记录为第二统计结果;
根据所述第一统计结果中各个点的差分距离值统计距离增加的点的个数和距离减少的点的个数,并分别记录为第三统计结果和第四统计结果;
根据所述第一统计结果中各个点的差分距离值,统计连续两点距离差不相等或不满足等比关系的点的个数,记录为第五统计结果;
若所述数据分析区域首尾两点的距离差值大于第一预设阈值且小于第二预设阈值,则将变量标识置位。
在第一方面的一种实现方式中,所述基于所述统计结果识别出所述点云数据中的伪像点,包括:
根据识别精度设置识别条件;
基于所述识别条件对所述统计结果进行判断,确定出所述点云数据中的伪像点。
在第一方面的一种实现方式中,若所述第一统计结果大于第一统计阈值,所述第二统计结果大于第二统计阈值,所述第五统计结果大于第三统计阈值,所述变量标识置位,且距离差单调递增或递减,则确定所述数据分析区域的中心点为伪像点。
在第一方面的一种实现方式中,若所述第一统计结果大于第一统计阈值,所述第二统计结果大于第二统计阈值,所述第五统计结果大于第三统计阈值,所述变量标识置位,所述第三统计结果或所述第四统计结果小于或等于第四统计阈值,且当前点的前一点为伪像点,则确定所述数据分析区域的中心点为伪像点。
在第一方面的一种实现方式中,伪像点识别方法还包括:
将所述伪像点从所述点云数据中剔除。
第二方面,本申请实施例提供一种终端设备,包括:
确定单元,用于对雷达回传的点云数据中的每个点进行滑动遍历,确定每个点的数据分析区域;
统计单元,用于根据每个点的距离差分值确定所述数据分析区域中满足预设统计条件的点的统计结果;
识别单元,用于基于所述统计结果识别出所述点云数据中的伪像点。
第三方面,本申请实施例提供一种终端设备,所述终端设备包括处理器、存储器以及存储在所述存储器中并可在所述处理器上运行的计算机程序,所述处理器执行所述计算机程序时实现如第一方面或第一方面的任意可选方式所述的方法。
第四方面,本申请实施例提供一种计算机可读存储介质,所述计算机可读存储介质存储有计算机程序,所述计算机程序被处理器执行时实现如第一方面或第一方面的任意可选方式所述的方法。
第五方面,本申请实施例提供一种计算机程序产品,当计算机程序产品在终端设备上运行时,使得终端设备执行上述第一方面或第一方面的任意可选方式所述的方法。
有益效果
实施本申请实施例提供的一种伪像点识别方法、终端设备、计算机可读存储介质及计算机程序产品具有以下有益效果:
通过滑动遍历点云数据中的每个点,基于点云的距离差分结果来统计满足预设统计条件的点的统计结果,再基于统计结果准确识别出点云数据中的伪像点,能够有效地识别出点云数据中的伪像点,解决目前无法有效地识别出点云数据中的伪像点的问题。
附图说明
为了更清楚地说明本申请实施例中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本申请的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1是本申请实施例提供的雷达发射的光斑打到两个物体的交界处的场景示意图;
图2是本申请实施例提供的雷达发射的光斑打到两个物体的交界处时的回波示意图;
图3是本申请实施例提供的一种伪像点识别方法是示意性流程图;
图4是本申请实施例提供的S12的实现流程示意图;
图5是本申请实施例提供的拖点现象形成的伪像点的示意图;
图6是本申请实施例提供的包含伪像点的点云图像和通过本申请实施例提供的伪像点识别方法剔除掉伪像点之后的点云图像的对比图;
图7是本申请实施例提供的一种终端设备的结构示意图;
图8是本申请另一实施例提供的一种终端设备的结构示意图;
图9是本申请实施例提供的一种计算机可读存储介质的结构示意图。
本发明的实施方式
以下描述中,为了说明而不是为了限定,提出了诸如特定系统结构、技术之类的具体细节,以便透彻理解本申请实施例。然而,本领域的技术人员应当清楚,在没有这些具体细节的其它实施例中也可以实现本申请。在其它情况中,省略对众所周知的系统、装置、电路以及方法的详细说明,以免不必要的细节妨碍本申请的描述。
应当理解,在本申请说明书和所附权利要求书中使用的术语“和/或”是指相关联列出的项中的一个或多个的任何组合以及所有可能组合,并且包括这些组合。另外,在本申请说明书和所附权利要求书的描述中,术语“第一”、“第二”、“第三”等仅用于区分描述,而不能理解为指示或暗示相对重要性。
还应当理解,在本申请说明书中描述的参考“一个实施例”或“一些实施例”等意味着在本申请的一个或多个实施例中包括结合该实施例描述的特定特征、结构或特点。由此,在本说明书中的不同之处出现的语句“在一个实施例中”、“在一些实施例中”、“在其他一些实施例中”、“在另外一些实施例中”等不是必然都参考相同的实施例,而是意味着“一个或多个但不是所有的实施例”,除非是以其他方式另外特别强调。术语“包括”、“包含”、“具有”及它们的变形都意味着“包括但不限于”,除非是以其他方式另外特别强调。
激光雷达是一种使用激光器作为发射光源,采用光电探测技术进行探测的自动遥感设备。激光雷达可以包括发射系统、接收系统、扫描控制系统以及数据处理系统等部分。激光雷达的工作原理是通过向目标物体发射探测信号,探测信号在打到目标物体后,目标物体会反射该探测信号,形成回波信号,接收系统可以接收该回波信号,并且对接收到的回波信号进行处理,以得到目标物体的距离、大小、速度、反射率等信息。
请参阅图1,当同一方位上存在距离较近的两个目标物体时(目标物体1和目标物体2),激光雷达发射出的光斑打到两个物体的交界处会产生两个回波。如图2所示,光斑打到目标物体1和目标物体2后,分别在T1时刻和T2时刻发生反射,由于接收系统的分辨率限制,接收到的两个回波会重叠成一个宽波束回波,其回波时间会被估计为T3时刻,因此会导致测距错误。
目前无法有效地识别出点云数据中的伪像点,如果不将伪像点进行处理,则虚假点云(伪像点)会导致形成虚假目标物体,导致测距错误,如果过度得剔除伪像点,则会导致真实存在的斜坡物体点云被误判,同样会影响测距结果。
为了解决上述缺陷,本申请实施例通过滑动遍历点云数据中的每个点,基于点云的距离差分结果来统计满足预设统计条件的点的统计结果,再基于统计结果准确识别出点云数据中的伪像点,能够有效地识别出点云数据中的伪像点,解决目前无法有效地识别出点云数据中的伪像点的问题。以下将对本申请实施例提供的伪像点识别方法进行详细的说明:
请参阅图3,图3是本申请实施例提供的一种伪像点识别方法的示意性流程图。本申请实施例提供的伪像点识别方法的执行主体是与激光雷达通信连接的终端设备,上述终端设备可以是智能手机、平板电脑或可穿戴设备等移动终端,也可以是各种应用场景下的电脑、云服务器、雷达辅助计算机等。需要说明的是,本申请实施例提供的伪像点识别方法适用于任何形式的激光雷达产品,包括但不限于机械式激光雷达、MEMS激光雷达、FLASH激光雷达、OPA激光雷达等。
如图3所示,本申请实施例提供伪像点识别方法可以包括S11~S13,详述如下:
S11:对雷达回传的点云数据中的每个点进行滑动遍历,确定每个点的数据分析区域。
本申请实施例中,终端设备通过获取雷达上传的点云数据,然后对点云数据中的每个点进行滑动遍历,实现对每个点的识别。
在一种实现方式中,雷达系统在接收到回波后,可以将其接收到的数据上传至与其连接的终端设备,由终端设备对点云数据进行分析。在另一种可选的实施例中,所述雷达系统接收到回波后,可以将接收到的数据传到雷达数据处理模块,对数据处理完成后进行输出。可以理解的是,所述数据处理模块可以集成在雷达控制系统中。
在本申请实施例中,可以基于横向邻域和/或纵向邻域来确定数据分析区域。横向邻域是指以当前点为中心点确定的一个横向的数据分析区域,例如,若当前点数据在第1行中且确定数据分析区域半径为L/2,则数据分析的横向邻域可以是1行L列的一个区域;纵向邻域是指以当前点为中心点确定的一个纵向的数据分析区域,例如若当前数据在第1列中,且确定数据分析区域半径为L/2,则数据分析的纵向邻域可以是L行1列的一个区域,其中L为大于1的正整数,其中横向数据分析区域的半径和纵向分析区域半径可以相同也可以不同,在此不加限制。
示例性的,假设当前点的坐标为(m,n),以纵向邻域来确定数据分析区域,那么该点邻域就是(m,n-floor(L/2):n+floor(L/2))。再举例来说,当同时以横向邻域和纵向邻域来确定数据分析区域,若当前点的坐标为(m,n),则那么该点的领域可以为(m-floor(L/2):m+floor(L/2),n-floor(L/2):n+floor(L/2))。
需要说明的是,以点云数据中每个点为中心点都可以确定出其对应的数据分析区域,基于该数据分析区域对该点是否为伪像点进行分析。
还需要说明的是,每个点的数据分析区域可以为单横向邻域(即该点的数据分析区域仅为横向邻域)、单纵向邻域(即该点的数据分析区域仅为纵向邻域)、横向邻域和纵向邻域(即该点的数据分析区域既包括横向邻域也包括纵向邻域),在此不加限制。
S12:根据每个点的距离差分值确定所述数据分析区域中满足预设统计条件的点的统计结果。
在本申请实施例中,在确定了每个点的数据分析区域后,可以基于该数据分析区域中包含的每个点的距离差分值判断该数据分析区域中的点是否满足预设统计条件。
需要说明的是,预设统计条件可以根据实际情况进行设置,例如设置每个点的差分值的绝对值大于特定阈值,若满足该预设统计条件,则将该点记录到第一统计结果中。
在本申请一实施例中,在S12之前,还可以包括以下步骤:
计算所述数据分析区域中每个点的距离差分值。
对于点云数据中任意一个点,在确定出其数据分析区域后,就可以计算出该数据分析区域中的每个的距离差分值。
本申请实施例所述的距离差分值是指该点下一时刻的距离和当前时刻的距离的差值,每个点的距离差分值可以表示为公式(1)。
其中Diff(i)是指点i的距离差分值,Dist(i+1)是指点i在下一时刻的距离,Dist(i)是指点i在当前时刻的距离。
Diff(i)=Dist(i+1)-Dist(i)(1)
请参阅图4,在本申请一实施例中,上述S12可以包括S21~S25。
S21:加权统计所述数据分析区域中距离差分值大于特定阈值的点,并记录为第一统计结果。
在本申请实施例中,以数据分析区域为横向邻域为例进行说明,通过加权统计1行*L列邻域内距离差分结果的绝对值大于特定阈值的点,统计结果记为第一统计结果Cnt1,并记录满足条件的点的位置IndexList(记录坐标值)以及其距离差分值DiffList。
具体地,判断条件如下公式(2)所示,其中,Diff代表每一点的差分结果,Dist代表当前点的距离值,θ代表激光雷达角分辨率,ψ代表角度判断阈值,典型值为0.2rad。
|Diff|>Dis×sin(θ)/tan(ψ)(2)
需要说明的是,由于拖点点云距离从一个距离迅速连续变化到另一个距离,并且角分辨率足够小,因此其点云组成的方向向量看起来近似指向点云坐标系原点。因此,将判断条件设置如公式(2)。
示例性的,请参阅图5,其中,A和B为拖点现象形成的伪像点,若向量BA近似指向圆心O,那么∠ABC要足够小,假设为ψ,以此反推点云距离数据所需要满足的条件。当Dist足够大且角分辨率θ足够小时,CO的距离近似等于AO,那么BC可近似认为是A、B两点的距离差Diff,根据三角几何关系即可得到公式(2)。
需要说明的是,第一统计结果Cnt1是通过公式(2)的判断结果和当前点在横向邻域内的位置经过计算加权得到的。数据分析区域中每个点的加权权重可以基于与中心点的距离来确定,例如与中心点的距离越近,则加权权重越大;与中心点的距离越远,则加权权重越小。
还需要说明的是,每个点的加权系数计算可根据实际情况调整,例如加权系数服从线性递减、高斯递减等。
S22:根据所述第一统计结果中各个点的坐标位置统计坐标连续的点的个数,记录为第二统计结果。
需要说明的是,上述坐标连续是指第一统计结果中的点的坐标序号1、2、3、4、5,即是连续的,而不是跳跃的1、4、8。
具体地,可以基于S21中的点的位置IndexList来确定点的坐标序号,进而判断出各个点的坐标是否连续,若坐标连续,则将该点统计为坐标连续的点,记录到第二统计结果Cnt2。
S23:根据所述第一统计结果中各个点的差分距离值统计距离增加的点的个数和距离减少的点的个数,并分别记录为第三统计结果和第四统计结果。
在本申请实施例中,根据S21中的距离差分值DiffList判断列表中距离增加的点的个数,即Diff大于0的个数,记录为第三统计结果Cnt3;根据S21中的距离差分值DiffList判断列表中距离减少的点的个数,即Diff小于0的个数,记录为第四统计结果Cnt4。
S24:根据所述第一统计结果中各个点的差分距离值,统计连续两点距离差不相等或不满足等比关系的点的个数,记录为第五统计结果。
在本申请实施例中,根据S21中的距离差分值DiffList断列表中连续两点距离差是否相等或满足等比关系,统计连续两点距离差不相等或不满足等比关系的点的个数,记录为第五统计结果Cnt5。
需要说明的是,连续两点是指坐标相邻的两个点。
S25:若所述数据分析区域首尾两点的距离差值大于第一预设阈值且小于第二预设阈值,则将变量标识置位。
在本申请实施例中,数据分析区域首尾两点是指数据分析区域的第一个点和最后一个点,基于数据分析区域中每个点的坐标就能确定出数据分析区域的首尾两点,并且能够计算出首尾两点的距离差值。
需要说明的是,上述第一预设阈值和上述第二预设阈值可以根据实际应用进行设置,本申请在此不加以限制。
将变量标识Logic1置位可以是置1或置0,示例性的,当所述数据分析区域首尾两点的距离差值大于第一预设阈值且小于第二预设阈值时,将变量标识Logic1置1。
S13:基于所述统计结果识别出所述点云数据中的伪像点。
在本申请实施例中,在得到统计结果后,就能够针对统计结果对该点是否为伪像点进行判断,通过对点云数据中的每个点对应的统计结果都基于识别条件进行判断,就能够识别出点云数据中所有的伪像点。
在本申请一实施例中,上述S13可以包括以下步骤:
根据识别精度设置识别条件;
基于所述识别条件对所述统计结果进行判断,确定出所述点云数据中的伪像点。
识别精度是指去除伪像点的强弱程度,若需要加强伪像点的去除强度,则将识别精度设置偏高;若需要降低伪像点的去除强度,则将识别精度设置偏低。
在确定了识别精度后,可以设置多个联合的识别条件,只有多个识别条件均满足时,才将该数据分析区域的中心点确定为伪像点。
设置识别条件时可以基于S12的统计结果来进行设置,如果识别精度较高,则可以基于多个统计结果来设置识别条件,如果识别精度较低,则可以只基于一个或两个统计结果来设置识别条件,在应用中可以基于实际情况进行设置,本申请对此不再加以赘述。
示例性的,上述识别条件可以是若所述第一统计结果大于第一统计阈值,所述第二统计结果大于第二统计阈值,所述第五统计结果大于第三统计阈值,所述变量标识置位,且距离差单调递增或递减,则确定所述数据分析区域的中心点为伪像点,否则确定所述数据分析区域的中心点不是伪像点。
示例性的,上述识别条件可以是若所述第一统计结果大于第一统计阈值,所述第二统计结果大于第二统计阈值,所述第五统计结果大于第三统计阈值,所述变量标识置位,所述第三统计结果或所述第四统计结果小于或等于第四统计阈值,且当前点的前一点为伪像点,则确定所述数据分析区域的中心点为伪像点,否则确定所述数据分析区域的中心点不是伪像点。
以上可以看出,本申请实施例提供的伪像点识别方法,能够通过滑动遍历点云数据中的每个点,基于点云的距离差分结果来统计满足预设统计条件的点的统计结果,再基于统计结果准确识别出点云数据中的伪像点。
在本申请另一实施例中,上述伪像点识别方法还可以包括以下步骤:
将所述伪像点从所述点云数据中剔除。
在本申请实施例中,在识别出伪像点后,可以将伪像点的点标识Invalid置1,以标识该点为伪像点。
将点云数据中点标识Invalid为1的点全部剔除,就能够将伪像点从点云数据中剔除掉。
为了更直观地表现出本申请实施例提供的伪像点识别方法的有益效果,图6示出了包含伪像点的点云图像(箭头指向的位置就是伪像点)和通过本申请实施例提供的伪像点识别方法剔除掉伪像点之后的点云图像。
由图6明显可以看出,本申请实施例提供的伪像点识别方法能够有效地剔除掉点云图像中的伪像点,并且很好地保留真实物体的点云,有效地提高了测距精度。
应理解,上述实施例中各步骤的序号的大小并不意味着执行顺序的先后,各过程的执行顺序应以其功能和内在逻辑确定,而不应对本申请实施例的实施过程构成任何限定。
基于上述实施例所提供的伪像点识别方法,本发明实施例进一步给出实现上述方法实施例的终端设备的实施例。
请参阅图7,图7是本申请实施例提供的一种终端设备的结构示意图。本申请实施例中,终端设备包括的各单元用于执行图3对应的实施例中的各步骤。具体请参阅图3以及图3对应的实施例中的相关描述。为了便于说明,仅示出了与本实施例相关的部分。如图7所示,终端设备70包括:确定单元71、统计单元72和识别单元73。其中:
确定单元71用于对雷达回传的点云数据中的每个点进行滑动遍历,确定每个点的数据分析区域。
统计单元72用于根据每个点的距离差分值确定所述数据分析区域中满足预设统计条件的点的统计结果。
识别单元73用于基于所述统计结果识别出所述点云数据中的伪像点。
在本申请一实施例中,上述终端设备还可以包括计算单元,上述计算单元用于计算所述数据分析区域中每个点的距离差分值。
在本申请一实施例中,上述识别单元包括条件设置单元和判断单元,其中:条件设置单元用于根据识别精度设置识别条件;判断单元用于基于所述识别条件对所述统计结果进行判断,确定出所述点云数据中的伪像点。
在本申请一实施例中,上述统计单元72具体用于加权统计所述数据分析区域中距离差分值大于特定阈值的点,并记录为第一统计结果;根据所述第一统计结果中各个点的坐标位置统计坐标连续的点的个数,记录为第二统计结果;根据所述第一统计结果中各个点的差分距离值统计距离增加的点的个数和距离减少的点的个数,并分别记录为第三统计结果和第四统计结果;根据所述第一统计结果中各个点的差分距离值,统计连续两点距离差不相等或不满足等比关系的点的个数,记录为第五统计结果;若所述数据分析区域首尾两点的距离差值大于第一预设阈值且小于第二预设阈值,则将变量标识置位。
需要说明的是,上述各个单元之间的信息交互、执行过程等内容,由于与本申请方法实施例基于同一构思,其具体功能及带来的技术效果,具体可参照方法实施例部分,此处不再赘述。
图8是本申请另一实施例提供的一种终端设备的结构示意图。如图8所示,该实施例提供的终端设备8包括:处理器80、存储器81以及存储在所述存储器81中并可在所述处理器80上运行的计算机程序82,例如图像分割程序。处理器80执行所述计算机程序82时实现上述各个伪像点识别方法实施例中的步骤,例如图3所示的S11~S13。或者,所述处理器80执行所述计算机程序82时实现上述各终端设备实施例中各模块/单元的功能,例如图7所示单元71~73的功能。
示例性的,所述计算机程序82可以被分割成一个或多个模块/单元,所述一个或者多个模块/单元被存储在所述存储器81中,并由处理器80执行,以完成本申请。所述一个或多个模块/单元可以是能够完成特定功能的一系列计算机程序指令段,该指令段用于描述所述计算机程序82在所述终端设备8中的执行过程。例如,所述计算机程序82可以被分割成第一获取单元和第一处理单元,各单元具体功能请参阅图5对应地实施例中的相关描述,此处不赘述。
所述终端设备可包括但不仅限于,处理器80、存储器81。本领域技术人员可以理解,图8仅仅是终端设备8的示例,并不构成对终端设备8的限定,可以包括比图示更多或更少的部件,或者组合某些部件,或者不同的部件,例如所述终端设备还可以包括输入输出设备、网络接入设备、总线等。
所称处理器80可以是中央处理单元(Central Processing Unit,CPU),还可以是其他通用处理器、数字信号处理器 (Digital Signal Processor,DSP)、专用集成电路 (Application Specific Integrated Circuit,ASIC)、现成可编程门阵列 (Field-Programmable Gate Array,FPGA) 或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
所述存储器81可以是所述终端设备8的内部存储单元,例如终端设备8的硬盘或内存。所述存储器81也可以是所述终端设备8的外部存储设备,例如所述终端设备8上配备的插接式硬盘,智能存储卡(Smart Media Card, SMC),安全数字(Secure Digital, SD)卡,闪存卡(Flash Card)等。进一步地,所述存储器81还可以既包括所述终端设备8的内部存储单元也包括外部存储设备。所述存储器81用于存储所述计算机程序以及所述终端设备所需的其他程序和数据。所述存储器81还可以用于暂时地存储已经输出或者将要输出的数据。
本申请实施例还提供了一种计算机可读存储介质。请参阅图9,图9是本申请实施例提供的一种计算机可读存储介质的结构示意图,如图9所示,计算机可读存储介质90中存储有计算机程序91,计算机程序91被处理器执行时可实现上述伪像点识别方法。
本申请实施例提供了一种计算机程序产品,当计算机程序产品在终端设备上运行时,使得终端设备执行时实现可实现上述伪像点识别方法。
所属领域的技术人员可以清楚地了解到,为了描述的方便和简洁,仅以上述各功能单元、模块的划分进行举例说明,实际应用中,可以根据需要而将上述功能分配由不同的功能单元、模块完成,即将所述终端设备的内部结构划分成不同的功能单元或模块,以完成以上描述的全部或者部分功能。实施例中的各功能单元、模块可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个单元中,上述集成的单元既可以采用硬件的形式实现,也可以采用软件功能单元的形式实现。另外,各功能单元、模块的具体名称也只是为了便于相互区分,并不用于限制本申请的保护范围。上述系统中单元、模块的具体工作过程,可以参考前述方法实施例中的对应过程,在此不再赘述。
在上述实施例中,对各个实施例的描述都各有侧重,某个实施例中没有详述或记载的部分,可以参照其它实施例的相关描述。
本领域普通技术人员可以意识到,结合本文中所公开的实施例描述的各示例的单元及算法步骤,能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行,取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能,但是这种实现不应认为超出本申请的范围。
以上所述实施例仅用以说明本申请的技术方案,而非对其限制;尽管参照前述实施例对本申请进行了详细的说明,本领域的普通技术人员应当理解:其依然可以对前述各实施例所记载的技术方案进行修改,或者对其中部分技术特征进行等同替换;而这些修改或者替换,并不使相应技术方案的本质脱离本申请各实施例技术方案的精神和范围,均应包含在本申请的保护范围之内。

Claims (20)

  1. 一种伪像点识别方法,其特征在于,包括:
    对雷达回传的点云数据中的每个点进行滑动遍历,确定每个点的数据分析区域;
    根据每个点的距离差分值确定所述数据分析区域中满足预设统计条件的点的统计结果;
    基于所述统计结果识别出所述点云数据中的伪像点。
  2. 根据权利要求1所述的伪像点识别方法,其特征在于,在根据每个点的距离差分值确定所述数据分析区域中满足预设统计条件的点的统计结果之前,还包括:
    计算所述数据分析区域中每个点的距离差分值。
  3. 根据权利要求2所述的伪像点识别方法,其特征在于,所述根据每个点的距离差分值确定所述数据分析区域中满足预设统计条件的点的统计结果,包括:
    加权统计所述数据分析区域中距离差分值大于特定阈值的点,并记录为第一统计结果;
    根据所述第一统计结果中各个点的坐标位置统计坐标连续的点的个数,记录为第二统计结果;
    根据所述第一统计结果中各个点的差分距离值统计距离增加的点的个数和距离减少的点的个数,并分别记录为第三统计结果和第四统计结果;
    根据所述第一统计结果中各个点的差分距离值,统计连续两点距离差不相等或不满足等比关系的点的个数,记录为第五统计结果;
    若所述数据分析区域首尾两点的距离差值大于第一预设阈值且小于第二预设阈值,则将变量标识置位。
  4. 根据权利要求1所述的伪像点识别方法,其特征在于,所述基于所述统计结果识别出所述点云数据中的伪像点,包括:
    根据识别精度设置识别条件;
    基于所述识别条件对所述统计结果进行判断,确定出所述点云数据中的伪像点。
  5. 根据权利要求3所述的伪像点识别方法,其特征在于,若所述第一统计结果大于第一统计阈值,所述第二统计结果大于第二统计阈值,所述第五统计结果大于第三统计阈值,所述变量标识置位,且距离差单调递增或递减,则确定所述数据分析区域的中心点为伪像点。
  6. 根据权利要求3所述的伪像点识别方法,其特征在于,若所述第一统计结果大于第一统计阈值,所述第二统计结果大于第二统计阈值,所述第五统计结果大于第三统计阈值,所述变量标识置位,所述第三统计结果或所述第四统计结果小于或等于第四统计阈值,且当前点的前一点为伪像点,则确定所述数据分析区域的中心点为伪像点。
  7. 根据权利要求1至6任意一项所述的伪像点识别方法,其特征在于,还包括:
    将所述伪像点从所述点云数据中剔除。
  8. 一种终端设备,其特征在于,包括:
    确定单元,用于对雷达回传的点云数据中的每个点进行滑动遍历,确定每个点的数据分析区域;
    统计单元,用于根据每个点的距离差分值确定所述数据分析区域中满足预设统计条件的点的统计结果;
    识别单元,用于基于所述统计结果识别出所述点云数据中的伪像点。
  9. 如权利要求8所述的终端设备,其特征在于,所述终端设备还包括:
    计算单元,用于计算所述数据分析区域中每个点的距离差分值。
  10. 如权利要求8所述的终端设备,其特征在于,所述识别单元包括:
    条件设置单元,用于根据识别精度设置识别条件;
    判断单元,用于基于所述识别条件对所述统计结果进行判断,确定出所述点云数据中的伪像点。
  11. 一种终端设备,包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机程序,其特征在于,所述处理器执行所述计算机可读指令时实现如下步骤:
    对雷达回传的点云数据中的每个点进行滑动遍历,确定每个点的数据分析区域;
    根据每个点的距离差分值确定所述数据分析区域中满足预设统计条件的点的统计结果;
    基于所述统计结果识别出所述点云数据中的伪像点。
  12. 如权利要求11所述的终端设备,其特征在于,在根据每个点的距离差分值确定所述数据分析区域中满足预设统计条件的点的统计结果之前,还包括:
    计算所述数据分析区域中每个点的距离差分值。
  13. 如权利要求12所述的终端设备,其特征在于,所述根据每个点的距离差分值确定所述数据分析区域中满足预设统计条件的点的统计结果,包括:
    加权统计所述数据分析区域中距离差分值大于特定阈值的点,并记录为第一统计结果;
    根据所述第一统计结果中各个点的坐标位置统计坐标连续的点的个数,记录为第二统计结果;
    根据所述第一统计结果中各个点的差分距离值统计距离增加的点的个数和距离减少的点的个数,并分别记录为第三统计结果和第四统计结果;
    根据所述第一统计结果中各个点的差分距离值,统计连续两点距离差不相等或不满足等比关系的点的个数,记录为第五统计结果;
    若所述数据分析区域首尾两点的距离差值大于第一预设阈值且小于第二预设阈值,则将变量标识置位。
  14. 如权利要求11所述的终端设备,其特征在于,所述基于所述统计结果识别出所述点云数据中的伪像点,包括:
    根据识别精度设置识别条件;
    基于所述识别条件对所述统计结果进行判断,确定出所述点云数据中的伪像点。
  15. 如权利要求11至14任一项所述的终端设备,其特征在于,所述处理器执行所述计算机可读指令时还实现如下步骤:
    将所述伪像点从所述点云数据中剔除。
  16. 一种计算机可读存储介质,所述计算机可读存储介质存储有计算机程序,其特征在于,所述计算机可读指令被处理器执行时实现如下步骤:
    对雷达回传的点云数据中的每个点进行滑动遍历,确定每个点的数据分析区域;
    根据每个点的距离差分值确定所述数据分析区域中满足预设统计条件的点的统计结果;
    基于所述统计结果识别出所述点云数据中的伪像点。
  17. 如权利要求16所述的计算机可读存储介质,其特征在于,在根据每个点的距离差分值确定所述数据分析区域中满足预设统计条件的点的统计结果之前,还包括:
    计算所述数据分析区域中每个点的距离差分值。
  18. 如权利要求17所述的计算机可读存储介质,其特征在于,所述根据每个点的距离差分值确定所述数据分析区域中满足预设统计条件的点的统计结果,包括:
    加权统计所述数据分析区域中距离差分值大于特定阈值的点,并记录为第一统计结果;
    根据所述第一统计结果中各个点的坐标位置统计坐标连续的点的个数,记录为第二统计结果;
    根据所述第一统计结果中各个点的差分距离值统计距离增加的点的个数和距离减少的点的个数,并分别记录为第三统计结果和第四统计结果;
    根据所述第一统计结果中各个点的差分距离值,统计连续两点距离差不相等或不满足等比关系的点的个数,记录为第五统计结果;
    若所述数据分析区域首尾两点的距离差值大于第一预设阈值且小于第二预设阈值,则将变量标识置位。
  19. 如权利要求17所述的计算机可读存储介质,其特征在于,所述基于所述统计结果识别出所述点云数据中的伪像点,包括:
    根据识别精度设置识别条件;
    基于所述识别条件对所述统计结果进行判断,确定出所述点云数据中的伪像点。
  20. 如权利要求16至19任一项所述的计算机可读存储介质,其特征在于,所述计算机可读指令被处理器执行时还实现如下步骤:
    将所述伪像点从所述点云数据中剔除。
PCT/CN2021/111483 2021-08-09 2021-08-09 一种伪像点识别方法、终端设备及计算机可读存储介质 WO2023015407A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN202180100803.2A CN117677862A (zh) 2021-08-09 2021-08-09 一种伪像点识别方法、终端设备及计算机可读存储介质
PCT/CN2021/111483 WO2023015407A1 (zh) 2021-08-09 2021-08-09 一种伪像点识别方法、终端设备及计算机可读存储介质

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2021/111483 WO2023015407A1 (zh) 2021-08-09 2021-08-09 一种伪像点识别方法、终端设备及计算机可读存储介质

Publications (1)

Publication Number Publication Date
WO2023015407A1 true WO2023015407A1 (zh) 2023-02-16

Family

ID=85199674

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/111483 WO2023015407A1 (zh) 2021-08-09 2021-08-09 一种伪像点识别方法、终端设备及计算机可读存储介质

Country Status (2)

Country Link
CN (1) CN117677862A (zh)
WO (1) WO2023015407A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115984803A (zh) * 2023-03-10 2023-04-18 安徽蔚来智驾科技有限公司 数据处理方法、设备、驾驶设备和介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101776760A (zh) * 2010-02-09 2010-07-14 中国科学院上海技术物理研究所 一种基于单光子探测器的激光三维成像装置
CN110580705A (zh) * 2019-11-08 2019-12-17 江苏省测绘工程院 一种基于双域图信号滤波检测建筑物边缘点的方法
CN110865365A (zh) * 2019-11-27 2020-03-06 江苏集萃智能传感技术研究所有限公司 一种基于毫米波雷达的停车场噪声消除方法
CN111199555A (zh) * 2019-12-13 2020-05-26 意诺科技有限公司 毫米波雷达目标识别方法
US20210033711A1 (en) * 2019-07-29 2021-02-04 Infineon Technologies Ag Optical crosstalk mitigation in lidar using digital signal processing
CN113155027A (zh) * 2021-04-27 2021-07-23 中铁工程装备集团有限公司 一种隧道岩壁特征物识别方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101776760A (zh) * 2010-02-09 2010-07-14 中国科学院上海技术物理研究所 一种基于单光子探测器的激光三维成像装置
US20210033711A1 (en) * 2019-07-29 2021-02-04 Infineon Technologies Ag Optical crosstalk mitigation in lidar using digital signal processing
CN110580705A (zh) * 2019-11-08 2019-12-17 江苏省测绘工程院 一种基于双域图信号滤波检测建筑物边缘点的方法
CN110865365A (zh) * 2019-11-27 2020-03-06 江苏集萃智能传感技术研究所有限公司 一种基于毫米波雷达的停车场噪声消除方法
CN111199555A (zh) * 2019-12-13 2020-05-26 意诺科技有限公司 毫米波雷达目标识别方法
CN113155027A (zh) * 2021-04-27 2021-07-23 中铁工程装备集团有限公司 一种隧道岩壁特征物识别方法

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115984803A (zh) * 2023-03-10 2023-04-18 安徽蔚来智驾科技有限公司 数据处理方法、设备、驾驶设备和介质
CN115984803B (zh) * 2023-03-10 2023-12-12 安徽蔚来智驾科技有限公司 数据处理方法、设备、驾驶设备和介质

Also Published As

Publication number Publication date
CN117677862A (zh) 2024-03-08

Similar Documents

Publication Publication Date Title
WO2021104497A1 (zh) 基于激光雷达的定位方法及系统、存储介质和处理器
WO2021008202A1 (zh) 一种基于毫米波雷达点云特征的核支持向量机目标分类方法
CN113671480A (zh) 雷达与视频融合交通目标跟踪方法、系统、设备、终端
CN110442120B (zh) 控制机器人在不同场景下移动的方法、机器人及终端设备
CN112213725B (zh) 车载雷达的多径虚警抑制方法、装置及终端设备
JP7251409B2 (ja) 車線変更車両の検出装置、方法及びビデオ監視装置
CN110443275B (zh) 去除噪声的方法、设备及存储介质
CN112513679A (zh) 一种目标识别的方法和装置
CN115018903B (zh) 一种堆场料堆体积计算方法及系统
WO2023015407A1 (zh) 一种伪像点识别方法、终端设备及计算机可读存储介质
WO2023024087A1 (zh) 处理激光雷达点云的方法、装置、设备及存储介质
CN112686951A (zh) 用于确定机器人位置的方法、装置、终端及存储介质
US11698459B2 (en) Method and apparatus for determining drivable region information
WO2023024347A1 (zh) 一种机器人自主探索的方法、终端设备及存储介质
CN112689842B (zh) 一种目标检测方法以及装置
JP6657934B2 (ja) 物体検出装置
CN115728772A (zh) 激光扫描点的类别检测方法、装置及终端设备
CN113112551B (zh) 相机参数的确定方法、装置、路侧设备和云控平台
CN110309741B (zh) 障碍物检测方法及装置
CN116359886A (zh) 一种雷达控制方法、终端设备及计算机可读存储介质
CN113514825A (zh) 道路边缘的获取方法、装置和终端设备
CN111723866A (zh) 点云聚类的方法及装置、无人车、可读存储介质
WO2024060209A1 (zh) 一种处理点云的方法和雷达
CN115754972A (zh) 激光雷达的滤波方法、终端设备及计算机可读存储介质
JP7254243B2 (ja) 物体検知システムおよび物体検知方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21953046

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 202180100803.2

Country of ref document: CN

NENP Non-entry into the national phase

Ref country code: DE