Apparatus for classifying objects and method thereof

Abstract

The invention relates to an object classification device. The object classification device comprises a LiDAR and a processor. The processor may identify that a point corresponding to a first previous object box and a point corresponding to a second previous object box are included in an integrated object box including a contour point at a predetermined time and included in an object box representing an integrated object, separating and clustering the contour points at the preset time into contour points representing a first object and contour points representing a second object, and obtaining the contour points representing the first object and the second object based on the number of the separated contour points representing the first object and the number of the separated contour points representing the second object. Separation contour points representing the first object are stored in association with the first object, and separation contour points representing the second object are stored in association with the second object.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to Korean Patent Application No. 10-2023-0161395, filed on November 20, 2023, which is hereby incorporated by reference in its entirety for all purposes. Technical Field

[0003] The present disclosure relates to an object classification apparatus and method, and more particularly to a technique for recognizing an object based on contour points obtained by light detection and ranging (LiDAR). Background Art

[0004] Technology for detecting the surrounding environment and avoiding obstacles is crucial for autonomous vehicles.

[0005] The vehicle can obtain data indicating the position of objects around the vehicle through LiDAR. The distance from the LiDAR to the object can be obtained by the interval between the time when the LiDAR emits laser light and the time when the laser light reflected by the object is received. Then, the vehicle can recognize the position of a point included in the object in the space where the vehicle is located based on the angle at which the laser light is emitted and the distance to the object.

[0006] The data obtained by LiDAR is characterized by high resolution and a large number of points included in the data. The importance of technology for identifying objects around the vehicle from the data is increasing.

[0007] The information disclosed in the Background of the Disclosure is only for enhancement of understanding of the general background of the disclosure and should not be taken as an acknowledgement or any form of suggestion that this information constitutes the prior art already known to a person skilled in the art. Summary of the invention

[0008] Various aspects of the present disclosure are directed to providing an object classification apparatus and method for separating and recognizing a single merged object into multiple objects when multiple objects are recognized as a merged object.

[0009] Various aspects of the present disclosure are directed to providing an object classification apparatus and method for improving tracking performance of multiple objects.

[0010] Various aspects of the present disclosure are directed to providing an object classification apparatus and method for identifying the location of an object through LiDAR points.

[0011] Various aspects of the present disclosure are directed to providing an object classification apparatus and method for allocating a memory space for storing information related to an object.

[0012] Various aspects of the present disclosure are directed to providing an object classification apparatus and method for improving the accuracy of object separation.

[0013] The technical problems to be solved by the present disclosure are not limited to the above-mentioned problems, and any other technical problems not mentioned herein will be clearly understood by those skilled in the art to which the present disclosure belongs from the following description.

[0014] According to one aspect of the present disclosure, an object classification device includes a LiDAR and a processor.

[0015] According to an exemplary embodiment of the present disclosure, the processor may: through the LiDAR, when part or all of the contour points of the multiple external objects identified as an integrated object as one object at the predetermined time satisfy a distribution condition, a dispersion condition or a distribution shape condition, identify points corresponding to a first previous object frame included in an object frame representing multiple external objects at a previous time before the predetermined time, and points corresponding to a second previous object frame included in the object frame representing the multiple external objects at the previous time are included in the integrated object frame and included in the object frame representing the integrated object, the integrated object frame including the contour points at the predetermined time; separate and cluster the contour points at the predetermined time into contour points representing a first object corresponding to the first previous object frame and contour points representing a second object corresponding to the second previous object frame; based on the number of separated contour points representing the first object and the number of separated contour points representing the second object, store the separated contour points representing the first object in association with the first object; and store the separated contour points representing the second object in association with the second object.

[0016] According to an exemplary embodiment of the present disclosure, the processor may: when the number of objects stored in the memory space is greater than a predetermined number, exchange information related to the integrated object with information related to one of the first object and the second object, and store the information related to one of the first object and the second object in the memory space storing information related to the integrated object; when the number of objects stored in the memory space is greater than the predetermined number, exchange information related to an object having a lowest priority according to a predetermined criterion among the objects stored in the memory space with information related to one of the first object and the second object, and store the information related to one of the first object and the second object in the memory space storing information related to the integrated object. The information related to one of the objects is stored in a memory space storing information related to the object with the lowest priority; when the number of objects stored in the memory space is less than or equal to the predetermined number, the information related to the integrated object and the information related to the one object are exchanged, and the information related to the one object is stored in the memory space storing information related to the integrated object; and when the number of objects stored in the memory space is less than or equal to the predetermined number, memory space in the memory space that does not store information related to an object is allocated to information related to a different object, and then the information related to the different object is stored in the allocated memory space.

[0017] According to an exemplary embodiment of the present disclosure, the processor may identify that the dispersion condition is satisfied based on that part or all of the contour points at the predetermined time are identified in two areas separated by a straight line connecting two end points of the contour points at the predetermined time.

[0018] According to an exemplary embodiment of the present disclosure, the processor can identify whether part or all of the contour points at the predetermined time satisfy the dispersion condition or the distribution shape condition based on the peak point, and the peak point is the contour point farthest from the straight line connecting the two endpoints of the contour points at the predetermined time, and the contour point is identified as representing the integrated object.

[0019] According to an exemplary embodiment of the present disclosure, the processor may: identify a first dispersion of distances between a first straight line connecting one of the two endpoints and the peak point and at least one contour point located between the one endpoint and the peak point; identify a second dispersion of distances between a second straight line connecting the other of the two endpoints and the peak point and at least one contour point located between the other endpoint and the peak point; and determine that the dispersion condition is satisfied based on identifying that a dispersion value of a reference straight line including a smaller dispersion value of the first dispersion and the second dispersion is included in a reference dispersion threshold range, and identifying that a dispersion value of a non-reference straight line including a larger dispersion value of the first dispersion and the second dispersion is included in a non-reference dispersion threshold range different from the reference dispersion threshold range. At least one contour point located between the one endpoint and the peak point may be located in a region between a straight line passing through the one endpoint and perpendicular to the first straight line and a straight line passing through the peak point and perpendicular to the first straight line. At least one contour point located between the other endpoint and the peak point may be located in a region between a straight line passing through the other endpoint and perpendicular to the second straight line and a straight line passing through the peak point and perpendicular to the second straight line.

[0020] According to an exemplary embodiment of the present disclosure, the processor may: when the contour point at the predetermined time is located on the left side of the main vehicle, identify that the distribution shape condition is satisfied based on that the area where the peak points in the left and right areas separated by the straight line connecting the two endpoints are located is the left area; when the contour point at the predetermined time is located on the right side of the main vehicle, identify that the distribution shape condition is satisfied based on that the area where the peak points in the left and right areas are located is the right area; and when the contour point at the predetermined time is located in front of or behind the main vehicle, identify that the distribution shape condition is satisfied based on that the area where the peak points in the two areas separated by the straight line connecting the two endpoints are located is different from the area where the main vehicle is located.

[0021] According to an exemplary embodiment of the present disclosure, the processor may: identify a first breakpoint and a second breakpoint assumed to represent different objects based on the fact that the length between the contour points located between one of the two endpoints included in the reference straight line and the peak point is greater than a reference length and the distribution density of the contour points is less than or equal to a reference distribution density; identify a first group of contour points at the predetermined time including the first breakpoint and a second group of contour points at the predetermined time including the second breakpoint; store the contour points included in the first group as one of the first object or the second object; and store the contour points included in the second group as an object different from the one of the first object or the second object. At least one contour point located between one of the two endpoints included in the reference straight line and the peak point may be located in a region between a straight line passing through one of the two endpoints included in the reference line and perpendicular to the reference line and a straight line passing through the peak point and perpendicular to the reference line.

[0022] According to an exemplary embodiment of the present disclosure, the processor may: identify two areas separated by a straight line passing through a midpoint of a line segment connecting the first breakpoint and the second breakpoint and identified as separated; determine contour points included in one of the two areas including the first breakpoint as the first group; and identify contour points included in another area different from the one area and including the second breakpoint as the second group.

[0023] According to an exemplary embodiment of the present disclosure, the point corresponding to the first previous object frame may represent a center point of the first previous object frame. The point corresponding to the second previous object frame may represent a center point of the second previous object frame.

[0024] According to an exemplary embodiment of the present disclosure, the processor may: when the effectiveness of the separation is identified as being greater than a reference value based on the number of separation contour points representing the first object and the number of separation contour points representing the second object, store the separation contour points representing the first object in association with the first object, and store the separation contour points representing the second object in association with the second object; and when the effectiveness of the separation is identified as being less than or equal to the reference value based on the number of separation contour points representing the first object and the number of separation contour points representing the second object, store the contour points at the predetermined time in association with the integrated object, rather than separating and clustering the contour points.

[0025] According to one aspect of the present disclosure, an object classification method includes the following steps: by light detection and ranging (LiDAR), when part or all of the contour points of the integrated object at a predetermined time that the multiple external objects are identified as one object meet a distribution condition, a dispersion condition or a distribution shape condition, identifying: points corresponding to a first previous object frame included in an object frame representing multiple external objects at a previous time before a predetermined time, and points corresponding to a second previous object frame included in the object frame representing the multiple external objects at the previous time are included in the integrated object frame and included in the object frame representing the integrated object, and the integrated object frame includes the contour points at the predetermined time; separating and clustering the contour points at the predetermined time into contour points representing a first object corresponding to the first previous object frame and contour points representing a second object corresponding to the second previous object frame; based on the number of separated contour points representing the first object and the number of separated contour points representing the second object, storing the separated contour points representing the first object in association with the first object; and storing the separated contour points representing the second object in association with the second object.

[0026] According to an exemplary embodiment of the present disclosure, the object classification method may further include the following steps: when the number of objects stored in the memory space is greater than a predetermined number, exchanging information related to the integrated object with information related to one of the first object and the second object, and storing the information related to one of the first object and the second object in the memory space storing information related to the integrated object; when the number of objects stored in the memory space is greater than the predetermined number, exchanging information related to an object having the lowest priority according to a predetermined criterion among the objects stored in the memory space with information related to one of the first object and the second object, and storing the information related to one of the first object and the second object in the memory space storing information related to the integrated object. The information related to one of the two objects is stored in the memory space storing information related to the object with the lowest priority; when the number of objects stored in the memory space is less than or equal to the predetermined number, the information related to the integrated object and the information related to the one object are exchanged, and the information related to the one object is stored in the memory space storing information related to the integrated object; and when the number of objects stored in the memory space is less than or equal to the predetermined number, the memory space in the memory space that does not store information related to the object is allocated to information related to a different object, and then the information related to the different object is stored in the allocated memory space.

[0027] According to an exemplary embodiment of the present disclosure, through the LiDAR, when the multiple external objects are identified as part or all of the contour points of the integrated object at the predetermined time as an object satisfy the distribution condition, the dispersion condition or the distribution shape condition, identifying that the points corresponding to the first previous object frame included in the object frame representing the multiple external objects at the previous time before the predetermined time, and the points corresponding to the second previous object frame included in the object frame representing the multiple external objects at the previous time are included in the integrated object frame and included in the object frame representing the integrated object, and the integrated object frame includes the contour points at the predetermined time, may include: identifying that the dispersion condition is satisfied based on that part or all of the contour points at the predetermined time are identified in two areas separated by a straight line connecting two endpoints of the contour points at the predetermined time.

[0028] According to an exemplary embodiment of the present disclosure, through the LiDAR, when the multiple external objects are identified as part or all of the contour points of the integrated object at the predetermined time as an object satisfy the distribution condition, the dispersion condition or the distribution shape condition, identifying the points corresponding to the first previous object frame included in the object frame representing the multiple external objects at the previous time before the predetermined time, and the points corresponding to the second previous object frame included in the object frame representing the multiple external objects at the previous time are included in the integrated object frame and included in the object frame representing the integrated object, the integrated object frame including the contour points at the predetermined time, may include: identifying whether part or all of the contour points at the predetermined time satisfy the dispersion condition or satisfy the distribution shape condition based on a peak point, the peak point being the contour point farthest from a straight line connecting two endpoints of the contour points at the predetermined time, and the contour point being identified as representing the integrated object.

[0029] According to an exemplary embodiment of the present disclosure, through the LiDAR, when the plurality of external objects are identified as part or all of the contour points of the integrated object at the predetermined time as one object satisfy the distribution condition, the dispersion condition or the distribution shape condition, it is identified that the points corresponding to the first previous object frame included in the object frame representing the plurality of external objects at the previous time before the predetermined time, and the points corresponding to the second previous object frame included in the object frame representing the plurality of external objects at the previous time are included in the integrated object frame and are included in the object frame representing the integrated object, and the integrated object frame includes the contour points at the predetermined time. , may include: identifying a first dispersion of distances between a first straight line connecting one of the two endpoints and the peak point and at least one contour point located between the one endpoint and the peak point; identifying a second dispersion of distances between a second straight line connecting the other of the two endpoints and the peak point and at least one contour point located between the other endpoint and the peak point; and determining that the dispersion condition is satisfied based on identifying that a dispersion value of a reference straight line including a smaller dispersion value of the first dispersion and the second dispersion is included in a reference dispersion threshold range, and identifying that a dispersion value of a non-reference straight line including a larger dispersion value of the first dispersion and the second dispersion is included in a non-reference dispersion threshold range different from the reference dispersion threshold range. At least one contour point located between the one endpoint and the peak point may be located in a region between a straight line passing through the one endpoint and perpendicular to the first straight line and a straight line passing through the peak point and perpendicular to the first straight line. At least one contour point located between the other endpoint and the peak point may be located in a region between a straight line passing through the other endpoint and perpendicular to the second straight line and a straight line passing through the peak point and perpendicular to the second straight line.

[0030] According to an exemplary embodiment of the present disclosure, through the LiDAR, when the plurality of external objects are identified as part or all of the contour points of the integrated object at the predetermined time as one object satisfy the distribution condition, the dispersion condition or the distribution shape condition, it is identified that the points corresponding to the first previous object frame included in the object frame representing the plurality of external objects at the previous time before the predetermined time, and the points corresponding to the second previous object frame included in the object frame representing the plurality of external objects at the previous time are included in the integrated object frame and are included in the object frame representing the integrated object, and the integrated object frame includes the contour points at the predetermined time. , may include: when the contour point at the predetermined time is located on the left side of the main vehicle, based on the fact that the area where the peak points in the left and right areas separated by the straight line connecting the two endpoints are located is the left area, identifying that the distribution shape condition is satisfied; when the contour point at the predetermined time is located on the right side of the main vehicle, based on the fact that the area where the peak points in the left and right areas are located is the right area, identifying that the distribution shape condition is satisfied; and when the contour point at the predetermined time is located in front of or behind the main vehicle, based on the fact that the area where the peak points in the two areas separated by the straight line connecting the two endpoints are located is different from the area where the main vehicle is located, identifying that the distribution shape condition is satisfied.

[0031] According to an exemplary embodiment of the present disclosure, separating and clustering the contour points at the predetermined time into the contour points representing the first object corresponding to the first previous object frame and the contour points representing the second object corresponding to the second previous object frame may include: identifying a first breakpoint and a second breakpoint assumed to represent different objects based on the fact that the length between the contour points located between one of the two endpoints included in the reference straight line and the peak point is greater than a reference length and the distribution density of the contour points is less than or equal to a reference distribution density; identifying a first group of contour points at the predetermined time including the first breakpoint and a second group of contour points at the predetermined time including the second breakpoint; storing the contour points included in the first group as one of the first object or the second object; and storing the contour points included in the second group as an object different from the one of the first object or the second object. At least one contour point located between one of the two endpoints included in the reference straight line and the peak point may be located in a region between a straight line passing through one of the two endpoints included in the reference line and perpendicular to the reference line and a straight line passing through the peak point and perpendicular to the reference line.

[0032] According to an exemplary embodiment of the present disclosure, separating and clustering the contour points at the predetermined time into contour points representing the first object corresponding to the first previous object frame and contour points representing the second object corresponding to the second previous object frame may include: identifying two areas separated by a straight line passing through the midpoint of a line segment connecting the first breakpoint and the second breakpoint and identified as separated; determining the contour points included in one of the two areas including the first breakpoint as the first group; and identifying the contour points included in another area different from the one area and including the second breakpoint as the second group.

[0033] According to an exemplary embodiment of the present disclosure, a point corresponding to the first previous object frame represents a center point of the first previous object frame, and wherein a point corresponding to the second previous object frame represents a center point of the second previous object frame.

[0034] According to an exemplary embodiment of the present disclosure, based on the number of separation contour points representing the first object and the number of separation contour points representing the second object, storing the separation contour points representing the first object in association with the first object, and storing the separation contour points representing the second object in association with the second object may include: when the effectiveness of the separation is identified as greater than a reference value based on the number of separation contour points representing the first object and the number of separation contour points representing the second object, storing the separation contour points representing the first object in association with the first object; when the effectiveness of the separation is identified as greater than a reference value based on the number of separation contour points representing the first object and the number of separation contour points representing the second object, storing the separation contour points representing the second object in association with the second object; and when the effectiveness of the separation is identified as less than or equal to the reference value based on the number of separation contour points representing the first object and the number of separation contour points representing the second object, storing the contour points at the predetermined time in association with the integrated object instead of separating and clustering the contour points.

[0035] The methods and apparatus of the present disclosure have other features and advantages that will be apparent from or set forth in more detail in the accompanying drawings incorporated herein and in the following detailed description, which together serve to explain certain principles of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 is a block diagram illustrating a configuration of an object classification apparatus according to an exemplary embodiment of the present disclosure;

[0037] Figure 2 An example of object tracking performed by an existing object classification apparatus is shown;

[0038] Figure 3 A flowchart showing the operation of an object classification device for distinguishing and storing objects in an object classification device or an object classification method according to an exemplary embodiment of the present disclosure;

[0039] Figure 4 An example of contour points satisfying a distribution condition or a dispersion condition in an object classification device or an object classification method according to an exemplary embodiment of the present disclosure is shown;

[0040] Figure 5 An example of a distribution shape of contour points satisfying a distribution shape condition in an object classification device or an object classification method according to an exemplary embodiment of the present disclosure is shown;

[0041] Figure 6 An example of resource allocation for storing information related to an object in an object classification device or an object classification method according to an exemplary embodiment of the present disclosure is shown;

[0042] Figure 7 A flowchart showing the operation of an object recognition device for separating and storing contour points representing a first object or a second object in an object recognition device or an object classification method according to an exemplary embodiment of the present disclosure;

[0043] Figure 8 An example of object classification according to a conventional object classification device in an object classification device or an object classification method according to an exemplary embodiment of the present disclosure and an example of object classification according to an object classification device according to an exemplary embodiment of the present disclosure are shown;

[0044] Fig. 9 An example of object classification according to a conventional object classification device in an object classification device or method according to an exemplary embodiment of the present disclosure and an example of object classification according to an object classification device according to an exemplary embodiment of the present disclosure are shown; and

[0045] Fig.10 A computing system related to an object classification apparatus and an object classification method according to an exemplary embodiment of the present disclosure is shown.

[0046] It is to be understood that the accompanying drawings are not necessarily drawn to scale, but rather present somewhat simplified representations of various features illustrating the basic principles of the present disclosure. The specific design features of the present disclosure as disclosed herein (including, for example, specific dimensions, orientations, locations, and shapes) will be determined in part by the specific intended application and use environment.

[0047] In the drawings, reference numbers refer to the same or equivalent parts of the present disclosure throughout the several figures of the drawing. DETAILED DESCRIPTION

[0048] Reference will now be made in detail to various embodiments of the present disclosure, examples of which are shown in the accompanying drawings and described below. Although the present disclosure will be described in conjunction with exemplary embodiments of the present disclosure, it should be understood that this description is not intended to limit the present disclosure to those exemplary embodiments of the present disclosure. On the other hand, the present disclosure is intended to cover not only the exemplary embodiments of the present disclosure, but also various alternatives, modifications, equivalents and other embodiments, which may be included within the spirit and scope of the present disclosure as defined by the appended claims.

[0049] Various exemplary embodiments of the present disclosure will be described in detail below with reference to the exemplary drawings. When adding reference numbers to the components of each drawing, it should be noted that the same or equivalent components are designated by the same numbers even when shown on other drawings. In addition, when describing the exemplary embodiments of the present disclosure, detailed descriptions of well-known features or functions will be excluded so as not to unnecessarily obscure the main points of the present disclosure.

[0050] When describing the components of the exemplary embodiments of the present disclosure, terms such as first, second, "A", "B", "(a), (b)", etc. may be used. These terms are intended only to distinguish one component from another and do not limit the nature, order, or sequence of the constituent components. Unless otherwise defined, all terms (including technical or scientific terms) used herein include the same meanings as those commonly understood by those skilled in the art to which the present disclosure belongs. Such terms (such as those defined in commonly used dictionaries) should be interpreted as having the same meanings as the contextual meanings in the relevant art and should not be interpreted as having ideal or overly formal meanings unless expressly defined as having such meanings in this application.

[0051] In addition, in the exemplary embodiments of the present disclosure, the expressions "greater than" or "less than" may be used to indicate whether a specific condition is satisfied or achieved, but are only used to indicate examples and do not exclude "greater than or equal to" or "less than or equal to". The condition indicating "greater than or equal to" may be replaced with "greater than", the condition indicating "less than or equal to" may be replaced with "less than", and the condition indicating "greater than or equal to and less than" may be replaced with "greater than and less than or equal to". In addition, "A" to "B" represent at least one element from A (including A) to B (including B).

[0052] The following will refer to Figures 1 to 10 Various exemplary embodiments of the present disclosure are described in detail.

[0053] Figure 1 is a block diagram illustrating a configuration of an object classification device according to an exemplary embodiment of the present disclosure.

[0054] Reference Figure 1 , the object classification device 101 may include a LiDAR 103 and a processor 105.

[0055] At least one of the LiDAR 103 or the processor 105 or any combination thereof may be electronically and / or operatively coupled to each other via electronic components such as a communication bus.

[0056] According to an exemplary embodiment of the present disclosure, hereinafter, operably combining hardware may mean establishing a direct connection or an indirect connection between hardware in a wired or wireless manner so that a first hardware in the hardware is controlled by a second hardware in the hardware. The type and / or number of hardware included in the object classification device 101 is not limited to Figure 1 For example, the object classification device 101 may only include Figure 1 Some of the hardware components shown.

[0057] According to an exemplary embodiment of the present disclosure, the processor 105 of the object classification device 101 may obtain position information of points including objects around the vehicle of the object classification device 101 through the LiDAR 103. The processor 105 of the object classification device 101 may obtain points representing objects through the LiDAR 103. Points obtained through the LiDAR and representing objects may be referred to as point clouds, but embodiments of the present disclosure may not be limited thereto.

[0058] According to an exemplary embodiment of the present disclosure, the processor 105 of the object classification device 101 may extract contour points including part or all of the external points among points that represent the object and can represent the external shape of the object.

[0059] According to an exemplary embodiment of the present disclosure, contour points may be identified in each of the layers formed based on the z-axis in the x-axis, the y-axis, and the z-axis. For example, contour points may be obtained based on representative points included in the point cloud in each layer formed on the z-axis in the x-axis, the y-axis, and the z-axis. For example, the representative point may include all or part of the points located outside of the plurality of points included in the point cloud. For example, the point cloud may be obtained by performing clustering based on each point obtained by LiDAR being identified within a predetermined distance.

[0060] According to an exemplary embodiment of the present disclosure, the processor 105 of the object classification device 101 may classify the object based on contour points.

[0061] According to an exemplary embodiment of the present disclosure, the processor 105 of the object classification device 101 may classify the contour points into multiple objects at a previous time before a predetermined time. For example, at a previous time, the contour points may be classified into points representing a first object and points representing a second object.

[0062] According to an exemplary embodiment of the present disclosure, when the first object or the second object is an object in a moving state, the first object and the second object may move to positions close to each other.

[0063] According to an exemplary embodiment of the present disclosure, when a first object and a second object are located close to each other at a predetermined time, at least one contour point representing the first object and at least one contour point representing the second object may be recognized as being located close to each other.

[0064] At a predetermined time, the processor of the conventional object classification apparatus may recognize that at least one contour point representing the first object and at least one point representing the second object represent an integrated object as one object.

[0065] To prevent this situation, when contour points representing multiple external objects represent an integrated object as a single object, the processor 105 of the object classification device 101 according to various exemplary embodiments of the present disclosure may classify contour points identified as representing the integrated object into contour points representing a first object and contour points representing a second object.

[0066] According to an exemplary embodiment of the present disclosure, in order to identify whether contour points representing multiple external objects represent an integrated object as a single object, the processor 105 of the object classification device 101 may identify whether some or all of the contour points at a predetermined time satisfy a distribution condition, a dispersion condition, or a distribution shape condition, and identify: points corresponding to a first previous object frame representing a first object at a previous time before the predetermined time, and points corresponding to a second previous object frame representing a second object at the previous time are included in an integrated object frame included in an object frame, which includes contour points at the predetermined time and representing the integrated object. According to an exemplary embodiment of the present disclosure, the object frame may include a virtual frame to which information related to the external object is assigned. For example, the object frame may be referred to as a contour frame.

[0067] According to an exemplary embodiment of the present disclosure, when some or all of the contour points at a predetermined time satisfy the distribution condition or the dispersion condition, the processor 105 of the object classification device 101 may identify the shape of the contour points at the predetermined time as a shape including two breakpoints. A breakpoint may refer to a point where a line connecting the contour points in sequence is bent at a predetermined angle or greater. Figure 4 The reasons for identifying a contour point as representing an integrated object based on the presence of two breakpoints are described.

[0068] When part or all of the contour points at a predetermined time satisfy the distribution shape condition, the processor 105 of the object classification device 101 according to various exemplary embodiments of the present disclosure may identify a shape including one breakpoint. Figure 5 The reasons for identifying a contour point as representing an integrated object based on the presence of a breakpoint are described.

[0069] According to an exemplary embodiment of the present disclosure, based on the fact that a point corresponding to a first previous object frame at a previous time before a predetermined time and a point corresponding to a second previous object frame at a previous time are included in an integrated object frame at a predetermined time, the processor 105 of the object classification device 101 may separate and cluster the contour points representing the integrated object into contour points representing the first object and contour points representing the second object. The reason is that as the first object and the second object move to positions close to each other, the contour points representing the first object and the contour points representing the second object may be identified as representing the integrated object. According to an exemplary embodiment of the present disclosure, a point corresponding to the first previous object frame may represent a center point of the first previous object frame, and a point corresponding to the second previous object frame may represent a center point of the second previous object frame.

[0070] According to an exemplary embodiment of the present disclosure, the processor 105 of the object classification device 101 may identify effectiveness of contour point distinction based on the number of separated contour points representing the first object and the number of separated contour points representing the second object.

[0071] When the contour points are not distinguished based on the appropriate distinguishing point, the number of contour points representing one of the first and second objects may be identified as deviating from the number of contour points representing another object different from one of the first and second objects.

[0072] According to an exemplary embodiment of the present disclosure, when the effectiveness of distinction is identified to be greater than a reference value based on the number of separation contour points representing a first object and the number of separation contour points representing a second object, the processor 105 of the object classification device 101 may store the separation contour points representing the first object in association with the first object and store the separation contour points representing the second object in association with the second object.

[0073] When the effectiveness of differentiation is identified as being less than or equal to a reference value based on the number of separation contour points representing the first object and the number of separation contour points representing the second object, the processor 105 of the object classification device 101 may store the contour points at a predetermined time in association with the integrated object without differentiation and clustering.

[0074] In addition, the processor 105 of the object classification device 101 can manage the memory space used to store information related to the object. Figure 6 Describes the management of memory space.

[0075] Figure 2 An example of object tracking performed by a conventional object classification apparatus is shown.

[0076] Reference Figure 2, in the first frame 201, the processor of the existing object classification device may recognize the contour points as the first object 203 and the second object 205. In the second frame 211 after the first frame 201, the processor of the existing object classification device may recognize the contour points as the third object 213. In the third frame 221 after the second frame 211, the processor of the existing object classification device may recognize the contour points as the fourth object 223 and the fifth object 225.

[0077] The processor of the existing object classification device may identify the third object 213 of the second frame 211 as the same object as the second object 205 of the first frame 201. In other words, the processor of the existing object classification device may assign the same identifier to the third object 213 of the second frame 211 as the second object 205 of the first frame 201. Therefore, the age (e.g., 17) of the third object 213 of the second frame 211 may be greater than or equal to the age (e.g., 16) of the second object 205 of the first frame 201. The age of an object may indicate the number of times an object has been recognized in at least one previous frame.

[0078] The processor of the existing object classification device may not be able to identify the contour points corresponding to the first object 203 in the second frame 211. Therefore, the tracking of the first object 203 may be ended. After the tracking of the first object 203 is ended, when the contour points representing the first object 203 are recognized again at a time after the second frame 211, it may be necessary to calculate again to determine the characteristics of the first object 203. In addition, the forward direction of the object frame of the third object 213 in the second frame 211 may be recognized as different from the forward direction of the actual object. Due to the incorrectly recognized forward direction of the object frame of the third object 213 in the second frame 211, the processor of the existing object classification device may mistakenly believe that the third object 213 is entering the lane where the main vehicle is located. Therefore, the performance of a system configured to control the main vehicle (e.g., an autonomous driving system or a driver assistance system) may be degraded.

[0079] The processor of the existing object classification device may identify the fifth object 225 of the third frame 221 as the same object as the second object 205 of the first frame 201 and the third object 213 of the second frame 211. In other words, the processor of the existing object classification device may assign the fifth object 225 of the third frame 221 the same identifier as the identifier of the second object 205 of the first frame 201 and the identifier of the third object 213 of the second frame 211. Therefore, the age of the fifth object 225 of the third frame 221 (e.g., 18) may be greater than or equal to the age of the third object 213 of the second frame 211 (e.g., 17) and the age of the second object 205 of the first frame 201 (e.g., 16).

[0080] The processor of the existing object classification device may recognize, in the third frame 221, a contour point of the fourth object 223 of the third frame 221 corresponding to the same first object 203 of the first frame 201. Therefore, tracking of the fourth object 223 of the third frame 221 may start.

[0081] The processor of the existing object classification device cannot reference the information of the first object 203 of the first frame 201 relative to the fourth object 223 of the third frame 221, even if the first object 203 of the first frame 201 and the fourth object 223 of the third frame 221 identify the same external object. Therefore, additional calculations may be required to determine the speed of the object, determine whether the object is in a moving state, and determine the direction of travel of the object.

[0082] According to various exemplary embodiments of the present disclosure, the processor of the object classification device can classify the contour points representing the third object 213 in the second frame 211 into multiple objects, thereby reducing the increase in the amount of calculation and the degree of occurrence of the system performance degradation problem. Figure 3 An object classification method of an object classification device according to an exemplary embodiment of the present disclosure is described.

[0083] Figure 3 A flowchart illustrating an operation of an object classification device for distinguishing and storing objects in an object classification device or an object classification method according to an exemplary embodiment of the present disclosure.

[0084] In the following, it is assumed Figure 1 The object classification device 101 is configured to perform Figure 3 In addition, Figure 3 In the description, the operations described as being performed by the object classification device may be understood as being controlled by the processor 105 of the object classification device 101.

[0085] Reference Figure 3 In the first stage 301, the processor of the object classification device according to various exemplary embodiments of the present disclosure may be configured to determine an object to be separated.

[0086] In the first operation 303 of the first stage 301 , the processor of the object classification apparatus according to various exemplary embodiments of the present disclosure may be configured to determine an object to be separated based on a distribution type of contour points and history information.

[0087] According to an exemplary embodiment of the present disclosure, the contour points to be separated may be limited to objects in a moving state (e.g., a moving vehicle) or objects that can be in a moving state (e.g., a vehicle), but the exemplary embodiments of the present disclosure may not be limited thereto. For example, the contour points may represent objects in a moving state (e.g., a moving vehicle) and objects that cannot be in a moving state (e.g., a road edge).

[0088] When some or all of the contour points at a predetermined time satisfy a distribution condition, a dispersion condition or a distribution shape condition, based on identifying that a point corresponding to a first previous object frame at a previous time before the predetermined time and a point corresponding to a second previous object frame at the previous time are included in an integrated object frame including the contour points at the predetermined time and are included in an object frame representing the integrated object, the processor of the object classification device can identify whether the contour points are targets to be separated.

[0089] In the second stage 311 , the processor of the object classification device according to various exemplary embodiments of the present disclosure may separate objects to be separated.

[0090] In the second operation 313 of the second stage 311 , the processor of the object classification device according to various exemplary embodiments of the present disclosure may perform a service function for object differentiation.

[0091] The processor of the object classification device according to various exemplary embodiments of the present disclosure may find a breakpoint based on the fact that the distribution of contour points is lower than a reference value at a point where two different objects are attached to each other.

[0092] A processor of an object classification device according to another exemplary embodiment of the present disclosure may find a plurality of breakpoints bent in a lightning shape and separate contour points such that the plurality of breakpoints are included in different objects.

[0093] In the third operation 315 of the second stage 311 , the processor of the object classification apparatus according to various embodiments of the present disclosure may verify the validity of the two separated objects.

[0094] According to an embodiment of the present disclosure, the processor of the object classification device may identify the effectiveness of separation of contour points based on the number of contour points representing the separated first object and the number of contour points representing the separated second object.

[0095] In the third stage 321 , the processor of the object classification device according to various embodiments of the present disclosure may manage resources for object storage.

[0096] In the fourth operation 323 of the third stage 321, the processor of the object classification device according to various embodiments of the present disclosure may identify whether the number of stored objects is about 70 or more. When the number of stored objects is about 70 or more, the processor of the object classification device may perform the fifth operation 325 of the third stage 321. When the number of stored objects is less than about 70, the processor of the object classification device may perform the sixth operation 327 of the third stage 321.

[0097] In a fifth operation 325 of the third stage 321 , the processor of the object classification device according to various exemplary embodiments of the present disclosure may rewrite the information related to the object A and the information related to the object B in the existing memory space.

[0098] In the sixth operation 327 of the third stage 321, the processor of the object classification device according to various exemplary embodiments of the present disclosure may rewrite the information related to object A in the existing memory space, and allocate new memory space for the information related to object B and then store the information related to object B.

[0099] Figure 4 An example of contour points satisfying a distribution condition or a dispersion condition in an object classification device or an object classification method according to an exemplary embodiment of the present disclosure is shown. The processor of the object classification device according to various exemplary embodiments of the present disclosure may identify the presence of two breakpoints based on identifying that the contour points satisfy the distribution condition or the dispersion condition. The breakpoint may refer to a contour point where a line connecting the contour points in sequence is bent at a predetermined angle or greater.

[0100] Reference Figure 4 , the first set 401 may represent contour points at a predetermined time that are identified as satisfying the distribution condition and representing the integrated object.

[0101] The first point 403 may represent one of the two endpoints of the contour point at the predetermined time (eg, point P1 and A). The second point 405 may represent the other of the two endpoints of the contour point at the predetermined time (eg, point P7 and C).

[0102] The second set 411 may represent contour points at a predetermined time that are identified as satisfying the dispersion condition and representing an integrated object. The third point 413 may represent one of the two endpoints of the contour point at the predetermined time (e.g., point P1 and A). The fourth point 415 may represent the other of the two endpoints of the contour point at the predetermined time (e.g., point P7 and C). The fifth point 417 may represent a peak point (e.g., point P4 and B), which is the contour point farthest from the straight line connecting the two endpoints of the contour point at the predetermined time.

[0103] According to an exemplary embodiment of the present disclosure, based on identifying some or all of the contour points (e.g., P2, P3, P4, P5, and P6) at a predetermined time in two areas separated by a straight line connecting two endpoints of the contour points at the predetermined time (e.g., the first point 403 of the first set 401 and the second point 405 of the first set 401), the processor of the object classification device can identify whether the distribution condition is met.

[0104] Typically, when contour points represent a single object, some or all of the contour points are identified only in one of two areas separated by a straight line connecting two endpoints of the contour points (e.g., first point 403 of first set 401, second point 405 of first set 401).

[0105] According to an exemplary embodiment of the present disclosure, the processor of the object classification device can identify whether some or all of the contour points at a predetermined time meet the dispersion condition based on a peak point (for example, the fifth point 417 of the second set 411), wherein the peak point is the contour point farthest from the straight line connecting the two endpoints of the contour points at the predetermined time, and the contour point is identified as representing an integrated object.

[0106] According to an exemplary embodiment of the present disclosure, a processor of an object classification device may identify a first dispersion of distances between a first straight line connecting one of the endpoints (e.g., the third point 413 of the second set 411) and a peak point (e.g., the fifth point 417 of the second set 411) and at least one contour point (e.g., P2, P3) located between a point (e.g., the third point 413 of the second set 411) and a peak point (e.g., the fifth point 417 of the second set 411).

[0107] According to an exemplary embodiment of the present disclosure, a processor of an object classification device may identify a second dispersion of distances between a second straight line connecting another point different from one of the endpoints (e.g., the fourth point 415 of the second set 411) and a peak point (e.g., the fifth point 417 of the second set 411) and at least one contour point (e.g., P5, P6) located between the other point (e.g., the fourth point 415 of the second set 411) and the peak point (e.g., the fifth point 417 of the second set 411).

[0108] According to an exemplary embodiment of the present disclosure, based on identifying that a dispersion value of a reference line including a smaller dispersion value of the first dispersion and the second dispersion (e.g., a straight line connecting the fourth point 415 of the second set 411 and the fifth point 417 of the second set 411) falls within a reference dispersion threshold range, and a dispersion value of a non-reference line including a larger dispersion value of the first dispersion and the second dispersion (e.g., a straight line connecting the third point 413 and the fifth point 417 of the second set 411) falls within a non-reference dispersion threshold range, a processor of the object classification device can identify whether a dispersion condition is met.

[0109] According to an exemplary embodiment of the present disclosure, at least one contour point (e.g., P2 or P3) located between a point (e.g., the third point 413 of the second set 411) and a peak point (e.g., the fifth point 417 of the second set 411) may be located in a region between a straight line passing through a point (e.g., the third point 413 of the second set 411) and perpendicular to the first straight line and a straight line passing through the peak point (e.g., the fifth point 417 of the second set 411) and perpendicular to the first straight line.

[0110] According to an exemplary embodiment of the present disclosure, at least one contour point (e.g., P5 or P6) located between another point (e.g., the fourth point 415 of the second set 411) and a peak point (e.g., the fifth point 417 of the second set 411) may be located in a region between a straight line passing through another point (e.g., the fourth point 415 of the second set 411) and perpendicular to the second straight line and a straight line passing through the peak point (e.g., the fifth point 417 of the second set 411) and perpendicular to the second straight line.

[0111] Figure 5 An example of a distribution shape of contour points satisfying a distribution shape condition in an object classification device or an object classification method according to an exemplary embodiment of the present disclosure is shown. The processor of the object classification device according to various exemplary embodiments of the present disclosure may identify the presence of a breakpoint based on identifying that the contour points satisfy the distribution shape condition. The breakpoint may refer to a point where a line connecting the contour points in sequence is bent at a predetermined angle or greater.

[0112] Reference Figure 5 Based on identifying that the contour points at a predetermined time are arranged into a distribution shape included in the first group 503, the second group 505, the third group 507 or the fourth group 509, the processor of the object classification device can identify whether the contour points meet the distribution shape condition.

[0113] According to an exemplary embodiment of the present disclosure, the processor of the object classification device may identify a global quadrant with the host vehicle 501 as the origin. According to an exemplary embodiment of the present disclosure, the processor of the object classification device may identify a local quadrant with each object represented by the contour point as the origin.

[0114] When the contour point at the predetermined time is located on the left side of the host vehicle 501, based on the fact that the area where the peak point is located is the left area of ​​the left area and the right area divided by the straight line connecting the two end points of the contour point (such as the distribution shape included in the first group 503), the processor of the object classification device according to various exemplary embodiments of the present disclosure can identify that the distribution shape condition is satisfied. The peak point can represent the contour point farthest from the straight line connecting the two end points of the contour point at the predetermined time, and the contour point is identified as representing the integrated object.

[0115] In other words, when a contour point belongs to the first quadrant (e.g., 1Q) or the second quadrant (e.g., 2Q) in the global quadrant, the processor of the object classification device can recognize that the distribution shape condition is satisfied based on that the contour point belongs to the first quadrant or the second quadrant in the local quadrant.

[0116] When the contour points represent a single object located on the left side of the host vehicle 501, in general, it is difficult for an object located in the lane (e.g., a vehicle or a sign pole) to include a shape that is concave to the left. The contour points of the distributed shape included in the first group 503 may be generated by an object relatively close to the host vehicle 501 and an object relatively far from the host vehicle 501 located near the near object.

[0117] When the contour point at a predetermined time is located on the right side of the main vehicle 501, based on the fact that the area where the peak point is located is the right area of ​​the left area and the right area divided by the straight line connecting the two endpoints of the contour point (such as the distribution shape included in the second group 505), the processor of the object classification device according to various exemplary embodiments of the present disclosure can identify that the distribution shape condition is satisfied.

[0118] In other words, when a contour point belongs to the third quadrant (e.g., 3Q) or the fourth quadrant (e.g., 4Q) in the global quadrant, the processor of the object classification device can recognize that the distribution shape condition is satisfied based on that the contour point belongs to the third quadrant or the fourth quadrant in the local quadrant.

[0119] When the contour points represent a single object located on the right side of the host vehicle 501, in general, it is difficult for an object located in the lane (e.g., a vehicle, a sign post) to include a shape that is concave to the right. The contour points of the distributed shape included in the second group 505 may be generated by an object relatively close to the host vehicle 501 and an object relatively far from the host vehicle 501 located near the near object.

[0120] When a contour point at a predetermined time is located in front of or behind the main vehicle 501, based on the fact that the area where the peak point in two areas separated by a straight line connecting the two endpoints is located is different from the area where the main vehicle 501 is located (such as the distribution shapes included in the third group 507 and the fourth group 509), the processor of the object classification device according to various exemplary embodiments of the present disclosure can identify that the distribution shape condition is satisfied.

[0121] In other words, when the contour point is located on the axis dividing the first quadrant (e.g., 1Q) or the fourth quadrant (e.g., 4Q) in the global quadrant, the processor of the object classification device can recognize that the distribution shape condition is satisfied based on that the contour point belongs to the first quadrant or the fourth quadrant in the local quadrant. When the contour point is located on the axis dividing the second quadrant (e.g., 2Q) or the third quadrant (e.g., 3Q) in the global quadrant, the processor of the object classification device can recognize that the distribution shape condition is satisfied based on that the contour point belongs to the second quadrant or the third quadrant in the local quadrant.

[0122] When the contour points represent a single object located in front of or behind the host vehicle 501, in general, it is difficult for an object located in a lane (e.g., a vehicle or a sign pole) to include a shape that is concave toward the host vehicle. The contour points of the distributed shape included in the third group 507 or the fourth group 509 may be generated by an object relatively close to the host vehicle 501 and an object relatively far from the host vehicle 501 located near the near object.

[0123] According to an exemplary embodiment of the present disclosure, when the contour points satisfy the following Figure 4 The distribution conditions or dispersion conditions shown or Figure 5 When the distribution shape condition is shown, based on the points corresponding to the first previous object frame representing the first object at a previous time before the predetermined time and the points corresponding to the second previous object frame representing the second object at the previous time being included in the integrated object frame representing the integrated object at the predetermined time, the processor of the object classification device can separate and cluster the contour points at the predetermined time into contour points representing the first object and contour points representing the second object.

[0124] According to an exemplary embodiment of the present disclosure, in order to separate and cluster contour points, based on the fact that the length between contour points located between one of two endpoints included in the reference straight line and the peak point is greater than the reference length and the distribution density of the contour points is less than or equal to the reference distribution density, the processor of the object classification device may identify the first breakpoint and the second breakpoint estimated to represent different objects because no contour point is identified between the two objects.

[0125] According to an exemplary embodiment of the present disclosure, the processor of the object classification device may identify a first group of contour points at a predetermined time including a first breakpoint, and a second group of contour points at a predetermined time including a second breakpoint.

[0126] According to an exemplary embodiment of the present disclosure, the processor of the object classification device may store the contour points included in the first group as one of the first object or the second object, and store the contour points included in the second group as the other of the first object or the second object different from the one. At least one contour point located between one of the two endpoints included in the reference straight line and the peak point may be located in a region between a straight line passing through one of the two endpoints included in the reference line and perpendicular to the reference line and a straight line passing through the peak point and perpendicular to the reference line.

[0127] According to another exemplary embodiment of the present disclosure, the processor of the object classification device may identify two areas separated by a straight line passing through a midpoint of a line segment connecting a first breakpoint and a second breakpoint and identified as separated, identify contour points included in the one area including the first breakpoint as a first group, and identify contour points included in another area different from the one area and including the second breakpoint as a second group.

[0128] The first breakpoint and the second breakpoint may refer to points where a line connecting the contour points in sequence bends at a predetermined angle or more. When there are two breakpoints, the contour points may be arranged in a lightning shape.

[0129] Figure 6 An example of resource allocation for storing information related to an object in an object classification device or an object classification method according to an exemplary embodiment of the present disclosure is shown.

[0130] Reference Figure 6 , the processor of the object classification device may store information related to one of the first object and the second object in the first memory space 601. The processor of the object classification device may store information related to an object different from one of the first object and the second object in the second memory space 603.

[0131] According to an exemplary embodiment of the present disclosure, a processor of an object classification device may store information related to an object into which a contour point is classified in a memory space. st_MergedObject may be a memory space for storing the state of a merged object. st_GlobMem may be a memory space for storing the state of a separate object. st_MergedObjectTrack may represent a memory space for storing the management state of a tracked object.

[0132] When the number of objects stored in a memory space (e.g., st_MergedObjectTrack) is greater than a predetermined number (e.g., approximately 70), the processor of the object classification device may exchange the stored information related to the integrated object and the information related to one of the first object and the second object (e.g., the first object), and store the information related to one object in the memory space (e.g., the first memory space 601) in which the information related to the integrated object is stored.

[0133] When the number of objects stored in a memory space (e.g., st_MergedObjectTrack) is greater than a predetermined number (e.g., approximately 70), the processor of the object classification device may exchange information related to an object having the lowest priority and information related to an object other than one of the first object and the second object (e.g., the second object 603), and store the information related to the object other than the one object in a memory space where the object having the lowest priority according to a predetermined criterion among the objects stored in the memory space is stored.

[0134] When the number of objects stored in a memory space (e.g., st_MergedObjectTrack) is less than or equal to a predetermined number (e.g., approximately 70), the processor of the object classification device may exchange information related to the integrated object and information related to one of the first object and the second object (e.g., the first object), and store the information related to one object in a memory space (e.g., the first memory space 601) in which information related to the integrated object is stored.

[0135] When the number of objects stored in a memory space (e.g., st_MergedObjectTrack) is less than or equal to a predetermined number (e.g., approximately 70), the processor of the object classification device may assign a memory space in the memory space (e.g., the second memory space 603) in which information related to the object is not stored to information related to a different object, and then store the information related to the different object in the assigned memory space.

[0136] Figure 7 A flowchart showing the operation of an object recognition device for separating and storing contour points representing a first object or a second object in an object recognition device or an object classification method according to an exemplary embodiment of the present disclosure is shown.

[0137] In the following, it is assumed Figure 1 The object classification device 101 is configured to perform Figure 7 In addition, Figure 7In the description, the operations referred to as being performed by the object classification device may be understood as being controlled by the processor 105 of the object classification device 101.

[0138] Reference Figure 7 In the first operation 701, when part or all of the contour points satisfy a distribution condition, a dispersion condition, or a distribution shape condition, the processor of the object classification device according to various exemplary embodiments of the present disclosure may recognize that points corresponding to the first previous object box and points corresponding to the second previous object box are included in the object box representing the integrated object.

[0139] According to an exemplary embodiment of the present disclosure, contour points may be acquired through LiDAR at a predetermined time, and a plurality of external objects may be recognized as an integrated object (ie, a single object).

[0140] In the second operation 703 , the processor of the object classification apparatus according to various exemplary embodiments of the present disclosure may separate and cluster the contour points at a predetermined time into contour points representing the first object and contour points representing the second object.

[0141] According to an exemplary embodiment of the present disclosure, the first object may correspond to a first previous object frame, and the second object may correspond to a second previous object frame.

[0142] In a third operation 705 , based on the number of contour points representing the first object and the number of contour points representing the second object, the processor of the object classification apparatus according to various exemplary embodiments of the present disclosure may store the contour points representing the first object in association with the first object.

[0143] In a fourth operation 707 , the processor of the object classification apparatus according to various exemplary embodiments of the present disclosure may store the contour point representing the second object in association with the second object.

[0144] Figure 8 An example of object classification according to a conventional object classification device in an object classification device or an object classification method according to an exemplary embodiment of the present disclosure and an example of object classification according to an object classification device according to an exemplary embodiment of the present disclosure are shown.

[0145] Reference Figure 8 , the first result 801 may include an example of object classification according to an existing object classification device. The first result 801 may include object classification results according to contour points in the first frame 803, the second frame 805, and the third frame 807. The second result 811 may include object classification results according to contour points in the fourth frame 813, the fifth frame 815, and the sixth frame 817.

[0146] Referring to the first result 801, if Figure 2As shown, the processor of the conventional object classification device may not recognize the contour point corresponding to the object located in the upper portion of the first frame 803 in the second frame 805. Therefore, the tracking of the object located in the upper portion of the first frame 803 may be terminated.

[0147] In the first result 801 , the upper object of the first frame 803 and the upper object of the third frame 807 may be the same external object, but may be recognized as different external objects by the existing object classification apparatus.

[0148] Therefore, the identifier (e.g., 41) of the upper object of the first frame 803 and the identifier (e.g., 24) of the upper object of the third frame 807 may be different from each other. In addition, the age (e.g., 4) of the upper object of the first frame 803 may be greater than or equal to the age (e.g., 1) of the upper object of the third frame 807.

[0149] In the first result 801, the lower object of the first frame 803 and the lower object of the third frame 807 may represent the same external object. Therefore, the lower object of the first frame 803 and the lower object of the third frame 807 may include the same identifier (e.g., 44). In addition, the age of the lower object of the first frame 803 (e.g., 16) may be smaller than the age of the lower object of the second frame 805 (e.g., 17). The age of the lower object of the second frame 805 (e.g., 17) may be smaller than the age of the lower object of the third frame 807 (e.g., 18).

[0150] When the contour points representing the object located in the upper portion of the first frame 803 are identified again at a time after the second frame 805 after the tracking of the object located in the upper portion of the first frame 803 has ended, it may be necessary to repeat the determination to identify the characteristics of the object. In addition, the traveling direction of the object frame of the object in the second frame 805 is identified as being different from the traveling direction of the actual object, which may deteriorate the performance of a system configured to control the host vehicle (e.g., an autonomous driving system or a driver assistance system).

[0151] Referring to the second result 811, the processor of the object classification device according to various exemplary embodiments of the present disclosure may identify contour points corresponding to the object located in the upper portion of the fourth frame 813 in the fifth frame 815. Therefore, tracking of the object located in the upper portion of the fourth frame 813 may be continuously performed.

[0152] In the second result 811, the upper object of the fourth frame 813, the upper object of the fifth frame 815, and the upper object of the sixth frame 817 may represent the same external object. Therefore, the upper object of the fourth frame 813, the upper object of the fifth frame 815, and the upper object of the sixth frame 817 may include the same identifier (e.g., 41). In addition, the age (e.g., 4) of the upper object of the fourth frame 813 may be smaller than the age (e.g., 5) of the upper object of the fifth frame 815. The age (e.g., 5) of the upper object of the fifth frame 815 may be smaller than the age (e.g., 6) of the upper object of the sixth frame 817.

[0153] In the second result 811, the lower object of the fourth frame 813, the lower object of the fifth frame 815, and the lower object of the sixth frame 817 may represent the same external object. Therefore, the lower object of the fourth frame 813, the lower object of the fifth frame 815, and the lower object of the sixth frame 817 may include the same identifier (e.g., 44). In addition, the age of the lower object of the fourth frame 813 (e.g., 16) may be smaller than the age of the lower object of the fifth frame 815 (e.g., 17). The age of the lower object of the fifth frame 815 (e.g., 17) may be smaller than the age of the lower object of the sixth frame 817 (e.g., 18).

[0154] Therefore, compared to existing object classification devices, the processor of the object classification device according to various exemplary embodiments of the present disclosure may reduce repeated calculations when identifying characteristics of an object and improve the performance of a system configured to control a host vehicle.

[0155] Fig. 9 In the object classification device or method according to the exemplary embodiment of the present disclosure, another example of object classification according to the existing object classification device and another example of object classification according to the object classification device of the exemplary embodiment of the present disclosure are shown.

[0156] Reference Fig. 9 , the first result 901 may include an example of object classification according to an existing object classification device. The first result 901 may include object classification results according to contour points in the first frame 903, the second frame 905, the third frame 907, and the fourth frame 909. The second result 911 may include object classification results according to contour points in the fifth frame 913, the sixth frame 915, the seventh frame 917, and the eighth frame 919.

[0157] Referring to the first result 901, the processor of the existing object classification device may not recognize the contour points corresponding to the object located in the lower portion of the first frame 903 in the second frame 905 and the third frame 907. Therefore, the tracking of the object located in the lower portion of the first frame 903 may be ended. In addition, the object located in the lower portion of the first frame 903 may be recognized again in the fourth frame 909. However, the processor of the existing object classification device may not recognize whether the object recognized in the first frame 903 and the object recognized in the fourth frame 909 are the same external object. Therefore, in the first result 901, the lower object of the first frame 903 and the lower object of the fourth frame 909 may be the same external object, but may be recognized as different external objects by the existing object classification device.

[0158] Therefore, the identifier (e.g., 43) of the lower object of the first frame 903 and the identifier (e.g., 19) of the lower object of the fourth frame 909 may be different from each other. Therefore, the age (e.g., 38) of the lower object of the first frame 903 may be greater than or equal to the age (e.g., 1) of the lower object of the fourth frame 909.

[0159] In the first result 901, the upper object of the first frame 903 and the upper object of the fourth frame 909 may represent the same external object. Therefore, the lower object of the first frame 903 and the upper object of the fourth frame 909 may include the same identifier (e.g., 44). In addition, the age of the upper object of the first frame 903 (e.g., 29) may be less than the age of the upper object of the second frame 905 (e.g., 30). The age of the object of the second frame 905 (e.g., 30) may be less than the age of the object of the third frame 907 (e.g., 31). The age of the object of the third frame 907 (e.g., 31) may be less than the age of the object of the fourth frame 909 (e.g., 32).

[0160] When the contour points representing the lower object of the first frame 903 are recognized again at a time after the second frame 905 after the tracking of the lower object of the first frame 903 has ended, it may be necessary to repeat calculations to recognize the characteristics of the object. In addition, the traveling direction of the object frame of the object in the second frame 905 is recognized to be different from the traveling direction of the actual object, which may deteriorate the performance of a system configured to control the host vehicle (e.g., an autonomous driving system or a driver assistance system).

[0161] Referring to the second result 911, the processor of the object classification device according to various exemplary embodiments of the present disclosure may identify contour points corresponding to the objects located in the lower portions of the fifth frame 913, the sixth frame 915, and the seventh frame 917. Therefore, tracking of the object located in the lower portion of the fifth frame 913 may be continuously performed.

[0162] In the second result 911, the lower object of the fifth frame 913, the lower object of the sixth frame 915, the lower object of the seventh frame 917, and the lower object of the eighth frame 919 may represent the same external object. Therefore, the lower object of the fifth frame 913, the lower object of the sixth frame 915, the lower object of the seventh frame 917, and the lower object of the eighth frame 919 include the same identifier (e.g., 43).

[0163] Therefore, the age of the lower object of the fifth frame 913 (e.g., 38) may be smaller than the age of the lower object of the sixth frame 915 (e.g., 39). The age of the lower object of the sixth frame 915 (e.g., 39) may be smaller than the age of the lower object of the seventh frame 917 (e.g., 40). The age of the lower object of the seventh frame 917 (e.g., 40) may be smaller than the age of the lower object of the eighth frame 919 (e.g., 41).

[0164] In the second result 911, the upper object of the fifth frame 913, the upper object of the sixth frame 915, the upper object of the seventh frame 917, and the upper object of the eighth frame 919 may represent the same external object. Therefore, the upper object of the fifth frame 913, the upper object of the sixth frame 915, the upper object of the seventh frame 917, and the upper object of the eighth frame 919 include the same identifier (e.g., 44).

[0165] In addition, the age of the upper object of the fifth frame 913 (e.g., 29) may be smaller than the age of the upper object of the sixth frame 915 (e.g., 30). The age of the upper object of the sixth frame 915 (e.g., 30) may be smaller than the age of the upper object of the seventh frame 917 (e.g., 31). The age of the upper object of the seventh frame 917 (e.g., 31) may be smaller than the age of the upper object of the eighth frame 919 (e.g., 32).

[0166] Therefore, compared to existing object classification devices, the processor of the object classification device according to various exemplary embodiments of the present disclosure may reduce repeated determinations when identifying characteristics of an object and improve the performance of a system configured to control a host vehicle.

[0167] Fig.10 A determination system related to an object classification apparatus and an object classification method according to an exemplary embodiment of the present disclosure is shown.

[0168] Reference Fig.10 The computing system 1000 may include at least one processor 1010 , a memory 1030 , a user interface input device 1040 , a user interface output device 1050 , a storage device 1060 , and a network interface 1070 , which are interconnected via a bus 1020 .

[0169] The processor 1010 may be a central processing unit (CPU) or a semiconductor device that processes instructions stored in the memory 1030 and / or the storage device 1060. The memory 1030 and the storage device 1060 may include various types of volatile or non-volatile storage media. For example, the memory 1030 may include a read-only memory (ROM) 1031 and a random access memory (RAM) 1032.

[0170] Thus, the operations of a method or algorithm described in connection with the exemplary embodiments included herein may be embodied directly in hardware or in a software module executed by the processor 1010, or in a combination thereof. The software module may reside on a storage medium (i.e., memory 1030 and / or storage device 1060) such as RAM, flash memory, ROM, EPROM, EEPROM, registers, hard disk, removable disk, and CD-ROM.

[0171] An exemplary storage medium may be coupled to the processor 1010, and the processor 1010 may read information from the storage medium and may record information in the storage medium. Alternatively, the storage medium may be integrated with the processor 1010. The processor and the storage medium may reside in an application specific integrated circuit (ASIC). The ASIC may reside in a user terminal. In another case, the processor and the storage medium may reside in a user terminal as separate components.

[0172] The above description is merely to illustrate the technical idea of ​​the present disclosure, and those skilled in the art to which the present disclosure belongs may make various modifications and variations without departing from the essential features of the present disclosure.

[0173] Therefore, the exemplary embodiments included in the exemplary embodiments of the present disclosure are not intended to limit the technical ideas of the present disclosure, but are used to describe the present disclosure, and the scope of the technical ideas of the present disclosure is not limited by the embodiments. The protection scope of the present disclosure should be interpreted by the following claims, and all technical ideas within the equivalent scope thereof should be interpreted as included within the scope of the present disclosure.

[0174] When multiple objects are recognized as a merged object, the present technology may separate and recognize a single merged object as multiple objects.

[0175] This technique can improve the tracking performance of multiple objects.

[0176] This technology can identify the location of an object through LiDAR points.

[0177] The present technology may allocate memory space for storing information associated with an object.

[0178] This technique can improve the accuracy of object separation.

[0179] Furthermore, various effects directly or indirectly understood through the present disclosure can be provided.

[0180] In various embodiments of the present disclosure, the memory and the processor may be provided as one chip, or may be provided as separate chips.

[0181] In various embodiments of the present disclosure, the scope of the present disclosure includes software or machine executable commands (e.g., operating systems, applications, firmware, programs, etc.) for enabling operations according to the methods of the various embodiments to be executed on a device or computer, including non-transitory computer-readable media on which such software or commands are stored and can be executed on a device or computer.

[0182] In various embodiments of the present disclosure, the control device may be implemented in the form of hardware or software, or may be implemented in a combination of hardware and software.

[0183] In addition, terms such as “unit”, “module”, etc. included in the specification refer to a unit for processing at least one function or operation, which can be implemented by hardware, software, or a combination thereof.

[0184] In an exemplary embodiment of the present disclosure, a vehicle may be referred to as being based on a concept including various vehicles. In some cases, a vehicle may be interpreted as being based on a concept including not only various land vehicles such as cars, motorcycles, trucks, and buses that travel on roads, but also various vehicles such as airplanes, drones, ships, and the like.

[0185] For ease of interpretation and accurate definition of the appended claims, the terms "upper", "lower", "inner", "outer", "above", "below", "upward", "downward", "front", "rear", "back", "inner", "outer", "inwardly", "outwardly", "interior", "exterior", "interior", "exterior", "forward", and "rearward" are used to describe features of the exemplary embodiments with reference to the locations of such features, as shown in the figures. It should also be understood that the term "connected" or its derivatives directly connect and refer to both indirect connections.

[0186] The term "and / or" may include a combination of multiple related listed items or any one of the multiple related listed items. For example, "A and / or B" includes three cases such as "A", "B" and "A and B".

[0187] In this specification, unless otherwise stated, a singular expression includes a plural expression unless the context clearly states otherwise.

[0188] In exemplary embodiments of the present disclosure, “at least one of A and B” may refer to “at least one of A or B” or “at least one combination of combinations of at least one of A and B”. In addition, “one or more of A and B” may refer to “one or more of A or B” or “one or more combinations of combinations of one or more of A and B”.

[0189] In the exemplary embodiments of the present disclosure, it should be understood that terms such as “including” or “having” are intended to indicate the presence of features, numbers, steps, operations, elements, parts, or combinations thereof described in the specification, and do not exclude the possibility of adding or existing one or more other features, numbers, steps, operations, elements, parts, or combinations thereof.

[0190] According to exemplary embodiments of the present disclosure, components may be combined with each other to be implemented as one, or some components may be omitted.

[0191] The above descriptions of specific exemplary embodiments of the present disclosure are presented for the purpose of illustration and description. They are not intended to be exhaustive or to limit the present disclosure to the precise form disclosed, and it is apparent that many modifications and variations are possible in light of the above teachings. The exemplary embodiments are selected and described in order to explain certain principles of the present disclosure and their practical applications, so that other persons skilled in the art can make and utilize various exemplary embodiments of the present disclosure, as well as various alternatives and modifications thereof. The scope of the present disclosure is intended to be defined by the appended claims and their equivalents.

Claims

1. An object classification device, comprising: Light detection and ranging LiDAR; as well as a processor operably connected to the LiDAR, The processor is configured to: By means of the LiDAR, in response to a plurality of external objects being identified as part or all of the contour points of an integrated object as one object at a predetermined time satisfying a distribution condition, a dispersion condition or a distribution shape condition, identifying: points corresponding to a first previous object frame included in an object frame representing a plurality of external objects at a previous time before a predetermined time, and points corresponding to a second previous object frame included in the object frame representing the plurality of external objects at the previous time are included in an integrated object frame and are included in the object frame representing the integrated object, the integrated object frame including the contour points at the predetermined time; separating and clustering the contour points at the predetermined time into: contour points representing a first object corresponding to the first previous object frame and contour points representing a second object corresponding to the second previous object frame; Based on the number of separated contour points representing the first object and the number of separated contour points representing the second object, the separated contour points representing the first object are stored in association with the first object; and the separated contour points representing the second object are stored in association with the second object.

2. The object classification device according to claim 1, wherein the processor is further configured to: In response to the number of objects stored in the memory space being greater than a predetermined number, exchanging information related to the integration object and information related to one of the first object and the second object, and storing the information related to one of the first object and the second object in the memory space storing the information related to the integration object; In response to the number of objects stored in the memory space being greater than the predetermined number, exchanging information related to an object having a lowest priority according to a predetermined criterion among the objects stored in the memory space and information related to one of the first object and the second object, and storing the information related to one of the first object and the second object in the memory space storing the information related to the object having the lowest priority; In response to the number of objects stored in the memory space being less than or equal to the predetermined number, exchanging information related to the integration object and information related to the one object, and storing the information related to the one object in the memory space storing the information related to the integration object; as well as In response to the number of objects stored in the memory space being less than or equal to the predetermined number, memory space in the memory space that does not store information related to the object is allocated to information related to a different object, and then the information related to the different object is stored in the allocated memory space.

3. An object classification device according to claim 1, wherein the processor is further configured to: identify that the dispersion condition is satisfied based on that part or all of the contour points at the predetermined time are identified in two areas separated by a straight line connecting two endpoints of the contour points at the predetermined time.

4. An object classification device according to claim 1, wherein the processor is further configured to: identify whether part or all of the contour points at the predetermined time meet the dispersion condition or the distribution shape condition based on a peak point, the peak point being the contour point farthest from the straight line connecting the two endpoints of the contour points at the predetermined time, and the contour point is identified as representing the integrated object.

5. The object classification device according to claim 4, wherein the processor is further configured to: identifying a first dispersion of distances between a first straight line connecting one of the two endpoints and the peak point and at least one contour point located between the one endpoint and the peak point; identifying a second dispersion of distances between a second straight line connecting the other of the two endpoints and the peak point and at least one contour point located between the other endpoint and the peak point; as well as determining that the dispersion condition is satisfied based on identifying that a dispersion value of a reference straight line including a smaller dispersion value of the first dispersion and the second dispersion is included in a reference dispersion threshold range, and identifying that a dispersion value of a non-reference straight line including a larger dispersion value of the first dispersion and the second dispersion is included in a non-reference dispersion threshold range different from the reference dispersion threshold range, wherein at least one contour point located between the one end point and the peak point is located in a region between a straight line passing through the one end point and perpendicular to the first straight line and a straight line passing through the peak point and perpendicular to the first straight line, At least one contour point located between the other end point and the peak point is located in a region between a straight line passing through the other end point and perpendicular to the second straight line and a straight line passing through the peak point and perpendicular to the second straight line.

6. The object classification device according to claim 4, wherein the processor is further configured to: in response to the contour point at the predetermined time being located on the left side of the host vehicle, identifying that the distribution shape condition is satisfied based on that a region where a peak point in a left region and a right region separated by a straight line connecting the two end points is located is the left region; in response to the contour point at the predetermined time being located on the right side of the host vehicle, identifying that the distribution shape condition is satisfied based on that the region where the peak point in the left region and the right region is located is the right region; and In response to the contour point at the predetermined time being located in front of or behind the main vehicle, the distribution shape condition is identified as being satisfied based on the fact that the area where the peak point in the two areas separated by the straight line connecting the two endpoints is located is different from the area where the main vehicle is located.

7. The object classification device according to claim 5, wherein the processor is further configured to: identifying a first breakpoint and a second breakpoint assumed to represent different objects based on that a length between contour points located between one of the two endpoints included in the reference straight line and the peak point is greater than a reference length and a distribution density of the contour points is less than or equal to a reference distribution density; identifying a first set of contour points at the predetermined time including the first breakpoint, and a second set of contour points at the predetermined time including the second breakpoint; storing the contour points included in the first group as one of the first object or the second object; and storing the contour points included in the second group as an object different from the one of the first object or the second object, Wherein at least one contour point located between one of the two endpoints included in the reference straight line and the peak point is located in a region between a straight line passing through one of the two endpoints included in the reference line and perpendicular to the reference line and a straight line passing through the peak point and perpendicular to the reference line.

8. The object classification device according to claim 7, wherein the processor is further configured to: identifying two regions separated by a straight line passing through a midpoint of a line segment connecting the first breakpoint and the second breakpoint and identified as being separated; determining contour points included in one of the two regions including the first breakpoint as the first group; and Contour points included in another region different from the one region and including a second breakpoint are identified as a second group.

9. The object classification device according to claim 1, wherein the point corresponding to the first previous object frame represents a center point of the first previous object frame, and The point corresponding to the second previous object frame represents the center point of the second previous object frame.

10. The object classification device according to claim 1, wherein the processor is further configured to: In response to identifying the effectiveness of separation as being greater than a reference value based on the number of separated contour points representing the first object and the number of separated contour points representing the second object, storing the separated contour points representing the first object in association with the first object; In response to identifying the effectiveness of separation as being greater than a reference value based on the number of separated contour points representing the first object and the number of separated contour points representing the second object, storing the separated contour points representing the second object in association with the second object; as well as In response to identifying the effectiveness of the separation as less than or equal to the reference value based on the number of separated contour points representing the first object and the number of separated contour points representing the second object, the contour points at the predetermined time are stored in association with the integrated object instead of separating and clustering the contour points.

11. A method for object classification, the method comprising the following steps: By light detection and ranging LiDAR, in response to a distribution condition, a dispersion condition or a distribution shape condition being satisfied by part or all of the contour points of an integrated object of a plurality of external objects recognized as one object at a predetermined time, identifying: points corresponding to a first previous object frame included in an object frame representing a plurality of external objects at a previous time before the predetermined time, and points corresponding to a second previous object frame included in an object frame representing a plurality of external objects at the previous time are included in an integrated object frame and are included in an object frame representing the integrated object, the integrated object frame including the contour points at the predetermined time; separating and clustering the contour points at the predetermined time into contour points representing a first object corresponding to the first previous object frame and contour points representing a second object corresponding to the second previous object frame; Based on the number of separated contour points representing the first object and the number of separated contour points representing the second object, the separated contour points representing the first object are stored in association with the first object; and the separated contour points representing the second object are stored in association with the second object.

12. The object classification method according to claim 11 further comprises the following steps: In response to the number of objects stored in the memory space being greater than a predetermined number, exchanging information related to the integration object and information related to one of the first object and the second object, and storing the information related to one of the first object and the second object in the memory space storing the information related to the integration object; In response to the number of objects stored in the memory space being greater than the predetermined number, exchanging information related to an object having the lowest priority according to a predetermined criterion among the objects stored in the memory space and the information related to one of the first object and the second object, and storing the information related to one of the first object and the second object in the memory space storing information related to the object having the lowest priority; In response to the number of objects stored in the memory space being less than or equal to the predetermined number, exchanging information related to the integration object and information related to the one object, and storing the information related to the one object in the memory space storing the information related to the integration object; as well as In response to the number of objects stored in the memory space being less than or equal to the predetermined number, memory space in the memory space that does not store information related to the object is allocated to information related to a different object, and then the information related to the different object is stored in the allocated memory space.

13. The object classification method according to claim 11, wherein by light detection and ranging LiDAR, in response to the fact that part or all of the contour points of the integrated object at a predetermined time in which the plurality of external objects are identified as one object satisfy a distribution condition, a dispersion condition or a distribution shape condition, it is identified that: points corresponding to a first previous object frame included in an object frame representing a plurality of external objects at a previous time before a predetermined time, and points corresponding to a second previous object frame included in the object frame representing the plurality of external objects at the previous time are included in an integrated object frame and are included in the object frame representing the integrated object, the integrated object frame including the contour points at the predetermined time, including: Satisfaction of the dispersion condition is identified based on that part or all of the contour points at the predetermined time are identified in two areas separated by a straight line connecting two end points of the contour points at the predetermined time.

14. The object classification method according to claim 11, wherein by light detection and ranging LiDAR, in response to the fact that part or all of the contour points of the integrated object at a predetermined time that the multiple external objects are identified as one object satisfy a distribution condition, a dispersion condition, or a distribution shape condition, it is identified that: points corresponding to a first previous object frame included in an object frame representing multiple external objects at a previous time before a predetermined time, and points corresponding to a second previous object frame included in the object frame representing the multiple external objects at the previous time are included in an integrated object frame and are included in the object frame representing the integrated object, the integrated object frame including the contour points at the predetermined time, including: Based on the peak point, it is identified whether part or all of the contour points at the predetermined time meet the dispersion condition or the distribution shape condition. The peak point is the contour point farthest from the straight line connecting the two endpoints of the contour points at the predetermined time. The contour point is identified as representing the integrated object.

15. The object classification method according to claim 14, wherein by light detection and ranging LiDAR, in response to the fact that part or all of the contour points of the integrated object at a predetermined time where the plurality of external objects are identified as one object satisfy a distribution condition, a dispersion condition or a distribution shape condition, it is identified that: points corresponding to a first previous object frame included in an object frame representing a plurality of external objects at a previous time before a predetermined time, and points corresponding to a second previous object frame included in the object frame representing the plurality of external objects at the previous time are included in an integrated object frame and are included in the object frame representing the integrated object, the integrated object frame including the contour points at the predetermined time, including: identifying a first dispersion of distances between a first straight line connecting one of the two endpoints and the peak point and at least one contour point located between the one endpoint and the peak point; identifying a second dispersion of distances between a second straight line connecting the other of the two endpoints and the peak point and at least one contour point located between the other endpoint and the peak point; as well as determining that the dispersion condition is satisfied based on identifying that a dispersion value of a reference straight line including a smaller dispersion value of the first dispersion and the second dispersion is included in a reference dispersion threshold range, and identifying that a dispersion value of a non-reference straight line including a larger dispersion value of the first dispersion and the second dispersion is included in a non-reference dispersion threshold range different from the reference dispersion threshold range, wherein at least one contour point located between the one end point and the peak point is located in a region between a straight line passing through the one end point and perpendicular to the first straight line and a straight line passing through the peak point and perpendicular to the first straight line, and At least one contour point located between the other end point and the peak point is located in a region between a straight line passing through the other end point and perpendicular to the second straight line and a straight line passing through the peak point and perpendicular to the second straight line.

16. The object classification method according to claim 14, wherein by light detection and ranging LiDAR, in response to the fact that part or all of the contour points of the integrated object at a predetermined time where the plurality of external objects are identified as one object satisfy a distribution condition, a dispersion condition or a distribution shape condition, it is identified that: points corresponding to a first previous object frame included in an object frame representing a plurality of external objects at a previous time before a predetermined time, and points corresponding to a second previous object frame included in the object frame representing the plurality of external objects at the previous time are included in an integrated object frame and are included in the object frame representing the integrated object, the integrated object frame including the contour points at the predetermined time, including: in response to the contour point at the predetermined time being located on the left side of the host vehicle, identifying that the distribution shape condition is satisfied based on that a region where a peak point in a left region and a right region separated by a straight line connecting the two end points is located is the left region; In response to the contour point at the predetermined time being located on the right side of the host vehicle, identifying that the distribution shape condition is satisfied based on that the region where the peak point in the left region and the right region is located is the right region; as well as In response to the contour point at the predetermined time being located in front of or behind the main vehicle, the distribution shape condition is identified as being satisfied based on the fact that the area where the peak point in the two areas separated by the straight line connecting the two endpoints is located is different from the area where the main vehicle is located.

17. The object classification method according to claim 15, wherein: Separating and clustering the contour points at the predetermined time into the contour points representing the first object corresponding to the first previous object frame and the contour points representing the second object corresponding to the second previous object frame, comprising: identifying a first breakpoint and a second breakpoint assumed to represent different objects based on that a length between contour points located between one of the two endpoints included in the reference straight line and the peak point is greater than a reference length and a distribution density of the contour points is less than or equal to a reference distribution density; identifying a first set of contour points at the predetermined time including the first breakpoint, and a second set of contour points at the predetermined time including the second breakpoint; storing the contour points included in the first group as one of the first object or the second object; and storing the contour points included in the second group as an object different from the one of the first object or the second object, Wherein at least one contour point located between one of the two endpoints included in the reference straight line and the peak point is located in a region between a straight line passing through one of the two endpoints included in the reference line and perpendicular to the reference line and a straight line passing through the peak point and perpendicular to the reference line.

18. The object classification method according to claim 17, wherein: Separating and clustering the contour points at the predetermined time into the contour points representing the first object corresponding to the first previous object frame and the contour points representing the second object corresponding to the second previous object frame, comprising: identifying two regions separated by a straight line passing through a midpoint of a line segment connecting the first breakpoint and the second breakpoint and identified as being separated; determining contour points included in one of the two regions including the first breakpoint as the first group; and Contour points included in another region different from the one region and including a second breakpoint are identified as a second group.

19. The object classification method according to claim 11, wherein the point corresponding to the first previous object frame represents a center point of the first previous object frame, and The point corresponding to the second previous object frame represents the center point of the second previous object frame.

20. The object classification method according to claim 11, wherein based on the number of separation contour points representing the first object and the number of separation contour points representing the second object, the separation contour points representing the first object are stored in association with the first object, and the separation contour points representing the second object are stored in association with the second object, comprising: In response to identifying the effectiveness of separation as being greater than a reference value based on the number of separated contour points representing the first object and the number of separated contour points representing the second object, storing the separated contour points representing the first object in association with the first object; In response to identifying the effectiveness of separation as being greater than a reference value based on the number of separated contour points representing the first object and the number of separated contour points representing the second object, storing the separated contour points representing the second object in association with the second object; as well as In response to identifying the effectiveness of the separation as less than or equal to the reference value based on the number of separated contour points representing the first object and the number of separated contour points representing the second object, the contour points at the predetermined time are stored in association with the integrated object instead of separating and clustering the contour points.

Images