Object recognition device and method

Abstract

The invention relates to an object recognition device and method. Such an object recognition device includes a sensor (e.g., LIDAR) and a processor. The processor may determine an object in a first frame via the sensor, determine a first direction of movement of the object in the first frame, determine a confidence value for the first direction of movement, determine a point representing the object in the first frame based on a representative direction of movement of the object, and output a signal indicating that the object is a moving object or a movable stationary object.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the priority of Korean Patent Application No. 10-2023-0128422 filed in the Korean Intellectual Property Office on September 25, 2023, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present invention relates to an object recognition device and method, and in particular to a technology for recognizing object features based on contour points obtained by light detection and ranging (LIDAR). Background Art

[0004] For autonomous vehicles or vehicles with driver assistance devices enabled, technology for detecting the surrounding environment and distinguishing obstacles is required in order to adjust the driving path and avoid obstacles while minimizing driver input.

[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. The vehicle can identify the position of the external point of the object in the space where the vehicle is located based on the angle of the laser light emission and the distance to the object.

[0006] An autonomous driving vehicle or a vehicle with a driver assistance device enabled can recognize information of an object represented by a point based on the acquired movement information of the point. In particular, technology for recognizing whether an object is an object in a moving state, an object capable of being in a moving state, or an object that cannot be in a moving state is essential to ensure the stability of autonomous driving or driver assistance driving and reduce the risk of accidents. Summary of the invention

[0007] The present invention aims to solve the above-mentioned problems occurring in the prior art while retaining the advantages of the prior art.

[0008] One aspect of the present invention provides an object recognition apparatus and method for recognizing whether an object is a moving object or an object capable of being in a moving state by selecting a representative point in a point cloud.

[0009] One aspect of the present invention provides an object recognition apparatus and method for improving the accuracy of determining whether an object is a moving object or an object capable of being in a moving state by selecting a representative point in a point cloud based on the moving direction of an object having a reliability value greater than a specified reliability value.

[0010] One aspect of the present invention provides an object recognition apparatus and method for improving the accuracy of determining whether an identified object is a moving object or an object capable of being in a moving state by selecting a representative point in a point cloud based on whether the identified object satisfies a specified condition.

[0011] The technical problems to be solved by the present invention 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 invention belongs from the following description.

[0012] According to one or more example embodiments of the present invention, an object recognition device may include: a sensor associated with a vehicle; and a processor. The processor may be configured to: determine an object in a first frame via the sensor; determine a first moving direction of the object in the first frame; determine a credibility value of the first moving direction; and determine a point representing the object in the first frame based on a representative moving direction of the object. The representative moving direction may be one of the following: a first moving direction based on a credibility value greater than a threshold, or a second moving direction of the object in a second frame before the first frame based on a credibility value less than or equal to a threshold. The processor may be further configured to output a signal indicating that the object is a moving object or a movable stationary object based on a first position of the point in the first frame and a second position of the point in at least one frame before the first frame.

[0013] The processor may be configured to determine the point by: determining an object frame representing the object in the first frame; determining a line segment associated with the object frame based on the representative movement direction; and further determining the point based on the line segment.

[0014] The processor may be configured to determine a point by the following steps: determining an object frame representing an object in a first frame; determining a first heading of the object frame in the first frame based on a third position of the object in the first frame and a fourth position of the object in at least one frame before the first frame; determining a trajectory heading of the object frame in the first frame based on the first heading and a second heading of the object frame in at least one frame before the first frame; determining a line segment associated with the object frame based on the first heading being within a first range, the trajectory heading being within a second range, and the representative moving direction being a specified direction; and further determining a point based on the line segment.

[0015] The processor may be configured to determine the point by: determining an object frame representing the object in a first frame; assigning a different index from a plurality of indexes to each of a plurality of vertices forming the object frame; and further determining the point based on the plurality of indexes.

[0016] The processor may be configured to determine a point by the following steps: determining an object frame representing an object in a first frame; determining a first heading of the object frame based on a first position of the object in the first frame and a second position of the object in at least one frame before the first frame; determining a trajectory heading of the object frame in the first frame based on the first heading and the second heading of the object frame in at least one frame before the first frame; assigning a different index from a plurality of indexes to each of a plurality of vertices forming the object frame based on the first heading being within a first range, the trajectory heading being within a second range, and the representative moving direction being a specified direction; and further determining a point based on the plurality of indexes.

[0017] The processor may be configured to determine the point by: determining an object frame representing the object in a first frame; and determining the point based on one of: a first line segment having a minimum longitudinal distance to the vehicle among line segments associated with the object frame based on a representative direction of movement being parallel to a longitudinal axis of the vehicle, or a second line segment having a minimum lateral distance to the vehicle among line segments associated with the object frame based on a representative direction of movement being perpendicular to the longitudinal axis of the vehicle.

[0018] The processor may be configured to determine the point by: determining an object frame representing the object in the first frame; and determining the point based on a center point of one of the line segments constituting the object frame.

[0019] The processor may be configured to determine a first moving direction of the object in the first frame based on a value obtained by adding a first distance moved by the vehicle to a second distance moved by the object relative to the vehicle. The second distance may be determined based on a difference between the first position and the second position.

[0020] The processor may be configured to determine whether the object is a moving object or a movable stationary object further based on the speed of the object.The speed of the object may be determined based on the first position and the second position.

[0021] The processor may be further configured to assign to the object one of: a first identifier indicating that the object is a mobile object, or a second identifier indicating that the object is a movable stationary object.

[0022] According to one or more example embodiments of the present invention, a method may include: determining an object in a first frame via a sensor associated with a vehicle; determining a first moving direction of the object in the first frame; determining a credibility value of the first moving direction; and determining a point in the first frame representing the object based on a representative moving direction of the object. The representative moving direction may be one of the following: a first moving direction based on a credibility value greater than a threshold, or a second moving direction of the object in a second frame preceding the first frame based on a credibility value less than or equal to a threshold. The method may also include outputting a signal indicating whether the object is a moving object or a movable stationary object based on a first position of the point in the first frame and a second position of the point in at least one frame preceding the first frame.

[0023] Determining the point may include: determining an object frame representing the object in the first frame; determining a line segment associated with the object frame based on the representative movement direction; and further determining the point based on the line segment.

[0024] Determining a point may include: determining an object frame representing an object in a first frame; determining a first heading of the object frame in the first frame based on a third position of the object in the first frame and a fourth position of the object in at least one frame before the first frame; determining a trajectory heading of the object frame in the first frame based on the first heading and a second heading of the object frame in at least one frame before the first frame; determining a line segment associated with the object frame based on the first heading being within a first range, the trajectory heading being within a second range, and the representative moving direction being a specified direction; and further determining a point based on the line segment.

[0025] Determining the point may include: determining an object frame representing the object in the first frame; assigning a different index from among a plurality of indexes to each of a plurality of vertices forming the object frame; and determining the point further based on the plurality of indexes.

[0026] Determining a point may include: determining an object frame representing an object in a first frame; determining a first heading of the object frame based on a first position of the object in the first frame and a second position of the object in at least one frame before the first frame; determining a trajectory heading of the object frame in the first frame based on the first heading and the second heading of the object frame in at least one frame before the first frame; assigning a different index from a plurality of indexes to each of a plurality of vertices forming the object frame based on the first heading being within a first range, the trajectory heading being within a second range, and the representative moving direction being a specified direction; and further determining a point based on the plurality of indexes.

[0027] Determining the point may include: determining an object frame representing the object in the first frame; and determining the point based on one of: a first line segment having a minimum longitudinal distance to the vehicle among line segments associated with the object frame based on a representative moving direction being parallel to the longitudinal axis of the vehicle, or a second line segment having a minimum lateral distance to the vehicle among line segments associated with the object frame based on a representative moving direction being perpendicular to the longitudinal axis of the vehicle.

[0028] Determining the point may include: determining an object frame representing the object in the first frame; and determining the point based on a center point of one of the line segments constituting the object frame.

[0029] Determining a first moving direction of the object in the first frame may include determining the first moving direction of the object in the first frame based on a value obtained by adding a first distance moved by the vehicle to a second distance moved by the object relative to the vehicle. The second distance may be determined based on a difference between the first position and the second position.

[0030] The object recognition method may further include: further determining whether the object is a moving object or a movable stationary object based on a speed of the object. The speed of the object may be determined based on the first position and the second position.

[0031] The object recognition method may further include assigning to the object one of: a first identifier indicating that the object is a mobile object, or a second identifier indicating that the object is a movable stationary object. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The above and other purposes, features and advantages of the present invention will become more apparent through the following specific embodiments in conjunction with the accompanying drawings:

[0033] Figure 1 is a block diagram showing an object recognition apparatus according to an embodiment of the present invention;

[0034] Figure 2 is a table showing information required for classifying an object in an object recognition device or an object recognition method according to an embodiment of the present invention;

[0035] Figure 3 showing an area where the weight of a feature value varies according to information related to an object recognition apparatus or an object recognition method according to an embodiment of the present invention;

[0036] Figure 4 An example of representative points recognized according to the moving direction of an object in an object recognition device or an object recognition method according to an embodiment of the present invention is shown;

[0037] Figure 5A flowchart showing the operation of an object recognition device in an object recognition device or an object recognition method according to an embodiment of the present invention, for recognizing whether an object is a moving object or an object capable of being in a moving state based on a representative point according to a moving direction of the object;

[0038] Figure 6 An example of a representative moving direction based on credibility recognition in an object recognition device or an object recognition method according to an embodiment of the present invention is shown;

[0039] Figure 7 An example of a representative point identified based on a heading within a first reference range, a track heading, and a representative moving direction in an object recognition device or an object recognition method according to an embodiment of the present invention is shown;

[0040] Figure 8 An example of a representative point identified based on a heading within a second reference range, a track heading, and a representative moving direction in an object recognition device or an object recognition method according to an embodiment of the present invention is shown;

[0041] Fig. 9 An example of representative points recognized according to representative moving directions in an object recognition device or an object recognition method according to an embodiment of the present invention is shown;

[0042] Fig.10 An example of representative points provided by an existing object recognition device or object recognition method and an example of representative points provided by an object recognition device or object recognition method according to an embodiment of the present invention are shown;

[0043] Fig.11 An example of an object path identified by an existing object recognition device or object recognition method and an example of an object path identified by an object recognition device or object recognition method according to an embodiment of the present invention is shown;

[0044] Fig.12 An example of an object path identified by an existing object recognition device or object recognition method and an example of an object path identified by an object recognition device or object recognition method according to an embodiment of the present invention are shown;

[0045] Fig.13 An example of an object path identified by an existing object recognition device or object recognition method and an example of an object path identified by an object recognition device or object recognition method according to an embodiment of the present invention is shown; and

[0046] Fig.14 A computing system related to an object recognition apparatus and an object recognition method according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0047] Hereinafter, some embodiments of the present invention will be described in detail with reference to the schematic drawings. When adding reference numerals to the components of each drawing, it should be noted that the same or equivalent components are represented by the same reference numerals even if they are also shown in other figures. In addition, when describing embodiments of the present invention, detailed descriptions of well-known features or functions will be omitted to avoid unnecessarily obscuring the main points of the present invention.

[0048] When describing components according to embodiments of the present invention, 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 these terms do not limit the nature, order, or sequence of the constituent components. Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by those skilled in the art to which the present invention belongs. Such terms as defined in commonly used dictionaries will be interpreted as having the same meaning as the contextual meaning in the relevant art, and should not be interpreted as having an ideal or overly formal meaning unless explicitly defined in this application.

[0049] In addition, the terms “unit”, “device”, “member”, “body” and the like used hereinafter may indicate at least one shape structure or may indicate a unit for processing a function.

[0050] In addition, the terms “unit”, “device”, “member”, “body” and the like used hereinafter may indicate at least one shape structure or may indicate a unit for processing a function.

[0051] In addition, in the embodiments of the present invention, the expressions "greater than" or "less than" may be used to indicate whether a specific condition is satisfied or possessed, but are only used to indicate embodiments, and do not exclude "greater than or equal to" or "less than or equal to". The case indicating "greater than or equal to" may be replaced by "greater than", the case indicating "less than or equal to" may be replaced by "less than", and the case indicating "greater than or equal to and less than" may be replaced by "greater than and less than or equal to". In addition, "A" to "B" means at least one of the elements from A (including A) to B (including B).

[0052] The following will refer to Figures 1 to 13 Embodiments of the present invention are described in detail.

[0053] Figure 1 is a block diagram showing an object recognition apparatus according to an embodiment of the present invention.

[0054] refer to Figure 1, the object recognition device 101 according to the embodiment of the present invention may be implemented inside a vehicle. In this case, the object recognition device 101 may be integrated with an internal control unit of the vehicle, or may be implemented as a separate device and connected to the control unit of the vehicle through a separate connection means.

[0055] refer to Figure 1 , the object recognition device 101 may include a sensor (eg, LIDAR 103 ) and a processor 105 .

[0056] According to an embodiment, the processor 105 of the object recognition device 101 may obtain a point cloud representing the object through a sensor (eg, LIDAR 103 ).

[0057] According to an embodiment, the processor 105 of the object recognition device 101 may recognize an object in the first frame based on a point cloud included in a specific first frame. The processor 105 of the object recognition device 101 may recognize (e.g., estimate) a first moving direction indicating a moving direction of an object in the first frame and a credibility (e.g., credibility value, credibility score, confidence value, accuracy level, etc.) of the first moving direction. A frame may be a collection of points (e.g., data points) collected by a sensor (e.g., LIDAR 103) at a single time or a single duration. The sensor (e.g., LIDAR 103) may collect data (e.g., points) in one or more frames over time to track any presence or movement of objects around the host vehicle.

[0058] When the estimated confidence of the first moving direction is greater than a specific confidence value, the processor of the object recognition device may identify the first moving direction as a representative moving direction. The processor 105 of the object recognition device 101 may identify a representative point representing the object in the first frame based on the first moving direction.

[0059] When the estimated confidence of the first moving direction is less than or equal to a specific confidence value, the processor 105 of the object recognition device 101 may identify the second direction as a representative moving direction, the second direction indicating the moving direction of the object in the second frame immediately before (e.g., before) the first frame. The representative moving direction in the first frame may include the second moving direction. The processor 105 of the object recognition device 101 may identify a representative point representing the object in the first frame based on the second moving direction.

[0060] The processor 105 of the object recognition device 101 can identify whether the object is a moving object or a stationary object capable of being in a moving state (also called a movable stationary object) based on the position of a representative point representing the object in the first frame and the position of a representative point representing the object in at least one frame before the first frame.

[0061] According to an embodiment, the processor 105 of the object recognition device 101 may assign an index to each of the multiple vertices forming the object frame, and identify one of the line segments forming the object frame based on the indexes of the multiple vertices and the representative moving direction. The processor 105 of the object recognition device 101 may identify a representative point representing the object based on one line segment. Fig. 9 A method for identifying representative points based on representative movement directions is described.

[0062] However, when the heading is included in the reference range, the track heading is included in the specified range, and the representative moving direction is the specified direction, the processor 105 of the object recognition device 101 may recognize a representative point representing the object based on the heading, the track heading, and the representative moving direction.

[0063] In other words, the processor 105 of the object recognition device 101 may identify the heading of the object frame based on the position of the object in the first frame and the position of the object in at least one frame before the first frame. The processor 105 of the object recognition device 101 may identify the trajectory heading of the object frame in the first frame based on the heading and the heading of the object frame in at least one frame before the first frame. The processor 105 of the object recognition device 101 may assign an index to each of a plurality of vertices forming the vertices of the object frame based on the heading being within a reference range, the trajectory heading being within a specified range, and the representative moving direction being a specified direction, identify one of the line segments constituting the object frame based on the indices of the plurality of vertices and the representative moving direction, and identify a representative point representing the object based on the line segment.

[0064] The following will refer to Figure 7 and Figure 8 A method for identifying representative points based on heading, representative movement direction, and trajectory heading is described.

[0065] According to one embodiment, 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. According to one embodiment, the contour frame may be created (or formed) based on contour points. For example, the contour frame may correspond to the external object. For example, the contour frame may include a virtual frame to which information related to the external object is assigned. For example, the information related to the external object may include at least one of the type of the external object, the speed of the external object, the moving direction of the external object, or the position of the external object, or any combination thereof.

[0066] According to an embodiment, the heading may indicate the direction of movement of an external object contained in an object frame. The track heading may be identified based on the direction of the object frame contained in a previous frame. For example, the track heading of a specific frame may be identified based on at least one heading of an object frame contained in at least one frame before the specific frame.

[0067] According to an embodiment, the processor 105 of the object recognition device 101 may recognize the feature value according to the tracking information. Figure 5 The operation of an object recognition device that recognizes a feature value based on tracking information based on recognized representative points is described.

[0068] According to an embodiment, the processor 105 of the object recognition device 101 may calculate a score indicating the probability that the object is a moving object (e.g., a moving vehicle) or a movable stationary object (e.g., a stationary vehicle) based on the feature value according to the object tracking information. Figure 2 A score indicating the probability that an object is a moving object or a movable stationary object is described based on identifying feature values ​​according to tracking information.

[0069] According to an embodiment, the processor 105 of the object recognition device 101 may identify an object as a mobile object or a movable stationary object based on a score indicating the probability that the object is a mobile object (e.g., a moving vehicle) or a movable stationary object (e.g., a stationary vehicle) is greater than a score indicating the probability that the object is a stationary object that cannot be in a moving state (also called an immovable stationary object) (e.g., a pole with a traffic sign).

[0070] According to an embodiment, the processor 105 of the object recognition device 101 may assign an identifier indicating that the object is a moving object or a movable stationary object to the object based on recognizing that the object is a moving object or a movable stationary object. The identifier may be called a tag, but is not limited thereto.

[0071] Figure 2 : is a table showing information required to classify an object in the object recognition apparatus or the object recognition method according to an embodiment of the present invention.

[0072] refer to Figure 2 , table 201 may indicate the type of information used to calculate a score for identifying whether an object is a mobile object or a movable stationary object. A fixity score 203 may indicate a score for identifying whether an object is an immovable stationary object. The fixity score 203 may be identified based on information such as out-of-lane information 211, frame size information 213, and frame matching information 215. A mobility score 205 may indicate a score for identifying whether an object is a mobile object or a movable stationary object. The mobility score 205 may be identified based on information such as in-lane information 217, tracking information 219, other in-lane object information 221, speed information 223, contour point distribution information 225, and boundary object information 227.

[0073] According to an embodiment, the first feature value may be a feature for determining whether the object is an immovable stationary object. The processor of the object recognition device may identify the fixedness score 203 by summing up the values ​​obtained by multiplying the first feature value indicated by the plurality of information by the weight. The second feature value may be a feature for determining whether the object is a mobile object or a movable stationary object. The processor of the object recognition device may identify the mobility score 205 by summing up the values ​​obtained by multiplying the second feature value indicated by the plurality of information by the weight.

[0074] According to an embodiment, the lane-outside information 211 for identifying the fixity score 203 may represent a first feature value assigned based on whether an object is recognized outside a lane. The frame size information 213 for identifying the fixity score 203 may represent a first feature value assigned based on whether the size of the object frame is greater than or equal to a reference size. The frame matching information 215 for identifying the fixity score 203 may represent a feature value assigned based on the distribution of contour points and the matching degree of the object frame.

[0075] According to an embodiment, contour points may be obtained by a sensor (e.g., LIDAR). For example, contour points may be identified in each layer formed based on the z-axis among the x-axis, the y-axis, and the z-axis. For example, contour points may be obtained based on external points included in a point cloud in each layer formed on the z-axis among the x-axis, the y-axis, and the z-axis. For example, the external points may include all or part of the points located outside of the plurality of points included in the point cloud. For example, a point cloud may be obtained by performing clustering based on each of the plurality of points acquired by a sensor (e.g., LIDAR) identified within a specified distance.

[0076] According to an embodiment, the in-lane information 217 may represent a second feature value assigned based on whether an object is recognized in the lane. The tracking information 219 may represent a second feature value assigned based on whether the object is moving. The speed information 223 may represent a second feature value assigned based on the speed of the object. The boundary object information 227 may represent a second feature value assigned based on whether the object is seen and not blocked at the boundary of the field of view.

[0077] According to an embodiment, the fixedness score 203 may be identified by summing the values ​​obtained by multiplying the first feature values ​​represented by the plurality of pieces of information by the weight. For example, the fixedness score 203 may be identified by summing the values ​​obtained by multiplying the first feature values ​​according to the lane-outside information 211 by the weight corresponding to the lane-outside information 211 (e.g., the weight S1 ), multiplying the first feature value according to the frame size information 213 by a weight corresponding to the frame size information 213 (for example, a weight S2 ), multiplying the first feature value according to the frame matching information 215 by the weight corresponding to the frame matching information 215 (for example, the weight S3) or any combination thereof to identify the fixity score 203. However, the embodiments of the present invention are not limited thereto. According to the embodiments, the fixity score 203 may be identified not only by adding values ​​obtained by multiplying the information listed in the table 201 by the weights, but also by adding values ​​obtained by multiplying the information not listed in the table 201 by the weights.

[0078] According to an embodiment, the mobility score 205 may be identified by summing up the values ​​obtained by multiplying the second feature values ​​represented by the plurality of pieces of information by the weight. For example, the mobility score 205 may be identified by summing up the values ​​obtained by multiplying the second feature values ​​according to the in-lane information 217 by the weight corresponding to the in-lane information 217 (e.g., the weight D1 ), by multiplying the second feature value according to the tracking information 219 by the weight corresponding to the tracking information 219 (for example, the weight D2 ), by multiplying the second feature value according to the other lane object information 221 by the weight corresponding to the other lane object information 221 (for example, the weight D3 ), by multiplying the second eigenvalue according to the speed information 223 by a weight corresponding to the speed information 223 (for example, a weight D4 ), by multiplying the second eigenvalue according to the contour point distribution information 225 by a weight corresponding to the contour point distribution information 225 (for example, a weight D5 ), by multiplying the second feature value according to the boundary object information 227 by the weight corresponding to the boundary object information 227 (for example, the weight D6 ), or the sum of at least one of the values ​​obtained by multiplying the information listed in the table 201 by the weight, or any combination thereof to identify the mobility score 205. However, the embodiments of the present invention are not limited thereto. According to an embodiment, the mobility score 205 may be identified not only by adding values ​​obtained by multiplying the information listed in the table 201 by the weight, but also by adding values ​​obtained by multiplying the information not listed in the table 201 by the weight.

[0079] According to an embodiment, when the mobility score 205 of the specific object is higher than the fixity score 203 of the specific object, the processor of the object recognition device may recognize the specific object as a mobile object or a movable stationary object. According to an embodiment, when the fixity score 203 of the specific object is higher than the mobility score 205 of the specific object, the processor of the object recognition device may recognize the specific object as an immovable stationary object.

[0080] According to an embodiment of the present invention, the processor of the object recognition device may recognize the second feature value indicated by the tracking information 219. Figure 5 A method for identifying a second feature value indicated by tracking information 219 according to an embodiment is described in FIG.

[0081] According to an embodiment of the present invention, the processor of the object recognition device may identify representative points for obtaining tracking information 219. Figures 4 to 9 A method for identifying a representative point according to an embodiment is described.

[0082] Hereinafter, the second feature value indicated by the tracking information 219 may be referred to as a feature value.

[0083] Figure 3 The area where the weight of the feature value changes according to the information related to the object recognition device or the object recognition method according to the embodiment of the present invention is shown.

[0084] refer to Figure 3 , frame 301 may represent a first area 305, a second area 307, and a third area 309 distinguished depending on the distance from a host vehicle 303 including an object recognition device. The first area 305 may include an area within the field of view. The second area 307 may include an area for classifying an object of interest. The third area 309 may include an area other than the first area 305 and the second area 307.

[0085] According to an embodiment, the first region 305 may be referred to as a field of view (FoV) region, but is not limited thereto. The second region 307 may be referred to as a category region of interest (classification ROI), but is not limited thereto. The third region 309 may be referred to as a default region, but is not limited thereto.

[0086] According to an embodiment, the processor of the object recognition device may assign different weights according to the area including the object (e.g., Figure 2 ) to identify the fixedness score or the mobility score. This is because the information with high importance may be different depending on the location of the object. For example, the processor of the object recognition device may only include the contour point distribution information (e.g., Figure 2 The weight of the contour point distribution information 225 in the first area 305 is set to a value greater than 0. For example, the processor of the object recognition device may Figure 2 The weight of the boundary object information 227 in the second area 307 is set to be greater than the weight of the boundary object information in the second area 307 and the weight of the boundary object information in the third area 309.

[0087] Figure 4 An example of representative points recognized according to the moving direction of an object in the object recognition apparatus or the object recognition method according to an embodiment of the present invention is shown.

[0088] refer to Figure 4 In the first case 401 , the moving direction of the object at the first position 403 is changed to the moving direction of the object at the second position 405 , or the moving direction of the object at the third position 407 .

[0089] In the second case 411 , the moving direction of the first object 413 may be different from the moving direction of the second object 415 .

[0090] According to an embodiment, the processor of the object recognition device included in the host vehicle may acquire a point cloud through a sensor (e.g., LIDAR). The processor of the object recognition device may identify the position of the object based on the acquired point cloud. For example, the object may be located at the first position 403, the second position 405, or the third position 407.

[0091] According to an embodiment, the object located at the first position 403 may move to the second position 405. Therefore, as the moving direction of the object located at the first position 403 changes to the second position 405, the representative point for identifying the feature value according to the tracking information of the object may change.

[0092] The object located at the first position 403 may move to the third position 407. Therefore, as the moving direction of the object located at the first position 403 changes to the third position 407, the representative point for identifying the feature value according to the tracking information of the object may change.

[0093] Unlike the existing object recognition apparatus, the object recognition apparatus according to the embodiment may continuously select specific points of the object frame as representative points even if the moving direction of the object changes.

[0094] In the second case 411, the processor of the object recognition device may assign an index to each of multiple vertices that form the vertices of an object box representing an object (e.g., the first object 413 or the second object 415), and identify a representative point representing the object based on the indices of the multiple vertices and the moving direction of the object.

[0095] In the second case 411, the moving direction of the first object 413 may be approximately 45 degrees to the +x-axis direction. The moving direction of the second object 415 may be approximately -45 degrees to the +x-axis direction.

[0096] The indices (3, 0, 1, and 2) assigned to the vertices of the object box representing the first object 413 when the moving direction of the first object 413 is less than approximately 45 degrees from the +x-axis direction may be different from the indices (0, 1, 2, and 3) assigned to the vertices of the object box representing the first object 413 when the moving direction of the first object 413 is greater than approximately 45 degrees from the +x-axis direction.

[0097] Therefore, the existing object recognition apparatus may recognize the representative points of the first object 413 (which are determined based on the indices respectively assigned to the vertices of the object frame and the movement direction) as discontinuous points according to the movement of the first object 413 .

[0098] The object recognition apparatus according to the embodiment may recognize representative points of the first object 413 as continuous points according to the moving direction of the first object 413 .

[0099] The indices (0, 1, 2, 3) assigned to the vertices of the object box representing the second object 415 when the moving direction of the second object 415 is less than approximately -45 degrees to the +x-axis direction may be different from the indices (3, 0, 1, 2) assigned to the vertices of the object box representing the second object 415 when the moving direction of the second object 415 is greater than approximately -45 degrees to the +x-axis direction.

[0100] Therefore, the existing object recognition apparatus may recognize the representative points of the second object 415 (which are determined based on the indices respectively assigned to the vertices of the object frame and the movement direction) as discontinuous points according to the movement of the second object 415 .

[0101] The object recognition apparatus according to the embodiment may recognize representative points of the second object 415 as continuous points according to the movement of the second object 415 .

[0102] Figure 5 A flowchart showing the operation of an object recognition device for recognizing whether an object is a moving object or a movable stationary object based on representative points according to a moving direction of the object in an object recognition device or an object recognition method according to an embodiment of the present invention.

[0103] In the following, it is assumed Figure 1 The object recognition device 101 performs Figure 5 In addition, Figure 5 In the description, the operations described as being performed by the device may be understood to be controlled by the processor 105 of the object recognition device 101.

[0104] refer to Figure 5 In a first step 501, a processor of an object recognition device according to an embodiment may recognize a first moving direction of an object in a specific first frame and a credibility (eg, confidence) of the first moving direction.

[0105] According to an embodiment, the moving direction of the object may be identified based on the road the object and the host vehicle are located in. For example, the processor of the object recognition device may identify the moving distance of the object in the first frame based on the sum of the moving distance of the object relative to the host vehicle and the moving distance of the host vehicle according to the difference between the position of the object in the first frame and the position of the object in at least one frame before the first frame.

[0106] According to an embodiment, one value among 1 point, 2 points, and 3 points may be used as the credibility of the moving direction of the object.

[0107] In the second step 503, the processor of the object recognition device according to the embodiment may identify whether the reliability is greater than a specific reliability value. When the reliability is greater than the specific reliability value, the processor of the object recognition device may execute the third step 505. When the reliability is not greater than the specific reliability value, the processor of the object recognition device may execute the fourth step 507.

[0108] The reason for this is that the moving direction of the object whose confidence is greater than a certain confidence value is identified as a representative moving direction. Due to an error caused by the operation of the host vehicle, the moving direction of the object may be different from the actual moving direction of the object.

[0109] In a third step 505 , the processor of the object recognition device according to an embodiment may recognize a representative point representing the object in the first frame based on the first moving direction.

[0110] According to an embodiment, the processor of the object recognition device may recognize the first moving direction as a representative moving direction, and recognize a representative point representing the object in the first frame based on the first moving direction, the first moving direction being the representative moving direction.

[0111] In a fourth step 507 , the processor of the object recognition device according to the embodiment may recognize a representative point representing the object in the first frame based on the second moving direction.

[0112] According to an embodiment, the second moving direction may indicate a moving direction of an object in a second frame immediately before (e.g., before) the first frame. According to an embodiment, the processor of the object recognition device may recognize the second moving direction as a representative moving direction, and recognize a representative point representing the object in the first frame based on the second moving direction as the representative moving direction.

[0113] In the fifth step 509, the processor of the object recognition device may recognize whether the object is a mobile object or a movable stationary object. For example, the processor of the object recognition device may output a signal indicating whether the object is a mobile object or a movable stationary object so that the vehicle can recognize the object and react accordingly by controlling the vehicle (e.g., avoiding the object).

[0114] According to an embodiment, the processor of the object recognition device may recognize whether the object is a moving object or a movable stationary object based on the speed of the object, wherein the speed of the object is recognized according to the position of a representative point representing the object in a first frame and the position of a representative point representing the object in at least one frame before the first frame.

[0115] According to an embodiment, the processor of the object recognition device may assign an identifier indicating that the object is a moving object or a movable stationary object to the object based on recognizing that the object is a moving object or a movable stationary object.

[0116] Figure 6An example of a representative moving direction recognized according to a feature in an object recognition apparatus or an object recognition method according to an embodiment of the present invention is shown.

[0117] refer to Figure 6 , the first table 601 may indicate the moving direction of the object and the credibility of the moving direction. The second table 611 may indicate the representative moving direction of the object. The first row 603 may indicate the moving direction and credibility of the object included in the frame after the first time point and before the second time point. The second row 605 may indicate the moving direction and credibility of the object included in the frame after the second time point and before the third time point. The third row 607 may indicate the moving direction and credibility of the object included in the frame after the third time point and before the fourth time point. The fourth row 609 may indicate the moving direction and credibility of the object included in the frame after the fourth time point and before the fifth time point.

[0118] According to an embodiment, the processor of the object recognition device may recognize the moving direction of the object in the first frame based on the difference between the position of the object in the first frame and the position of the object in at least one frame before the first frame. The F / B direction may refer to the direction opposite to the moving direction of the host vehicle. The RR direction may refer to the right direction relative to the moving direction of the host vehicle.

[0119] According to an embodiment, in the frame corresponding to RR(1) in the first row 603, the moving direction is identified as the RR direction to the right of the moving direction of the host vehicle, but the reliability of the moving direction of the object may be 1.

[0120] The processor of the object recognition device may identify a confidence value greater than a specific confidence value as a representative moving direction. Because when the specific confidence value is 2, the object corresponding to RR(1) has a confidence value of the object moving direction not exceeding a specific confidence value (e.g., approximately 2), the F / B direction, which is the moving direction of the object in the previous frame, may be identified as the representative moving direction of the object.

[0121] According to an embodiment, in the frame corresponding to RR(3) in the fourth row 609, the moving direction may be identified as the RR direction to the right of the moving direction of the host vehicle, and the reliability of the moving direction of the object may be 3.

[0122] According to an embodiment, since the credibility of the moving direction of the object corresponding to RR (3) is greater than a specific credibility value (e.g., approximately 2) when a specific credibility value is 2, the RR direction (i.e., the moving direction in the relevant frame) rather than the moving direction of the object in the previous frame can be identified as the representative moving direction of the object.

[0123] Figure 7An example of a representative point identified based on a heading, a track heading, and a representative moving direction in an object recognition apparatus or an object recognition method according to an embodiment of the present invention is shown.

[0124] refer to Figure 7 , the heading of the object frame of the object 701 may be included in the first reference range. The first case 703, the second case 705, the third case 707 and the fourth case 709 may represent objects whose track heading is included in the specified range and whose moving direction is the specified direction.

[0125] According to an embodiment, the heading of the object frame of the object 701 may be included in a range of values ​​obtained by adding or subtracting a specified value to +45 degrees in the +x-axis direction. For example, the heading of the object frame of the object 701 may be included in a range of greater than or equal to about 43 degrees and less than about 47 degrees in the +x-axis direction, but embodiments of the present invention are not limited thereto.

[0126] According to an embodiment, when the representative moving direction of the object 701 is the F / B direction opposite to the moving direction of the host vehicle, the representative point may correspond to the center point of the line segment of the object frame to which the indexes 0 and 3 are assigned.

[0127] According to an embodiment, when the direction of the object is included in the first reference range, the processor of the object recognition device may recognize a representative point different from the representative point recognized based on the representative moving direction and the index before the change in four cases. The four cases may include a first case 703, a second case 705, a third case 707, and a fourth case 709.

[0128] According to an embodiment, the processor of the object recognition device may identify the trajectory heading of the object frame in the first frame based on the heading of the first frame and the heading of the object frame in at least one frame before the first frame. For example, in two or more frames before the first frame, when the heading of the object frame is about -45 degrees with the +x-axis direction, the trajectory heading of the object frame may be about -45 degrees.

[0129] According to an embodiment, in the first case 703, the heading of the object frame of the object may be included in the range of values ​​obtained by adding or subtracting a specified value at about +45 degrees in the +x-axis direction (e.g., greater than or equal to about 43 degrees and less than about 47 degrees).

[0130] When the trajectory heading of the object frame is included in a specified range (e.g., a range between about -43 degrees and about -47 degrees from the +x-axis direction) and the representative moving direction corresponds to a specified direction (e.g., the same F / F direction as the moving direction of the host vehicle), the processor of the object recognition device may re-recognize the index corresponding to the object frame. In addition, the processor of the object recognition device may recognize the center point between the vertex with the index 0 after the change and the vertex with the index 3 after the change as the representative point, rather than the center point between the vertex with the index 0 before the change (e.g., the vertex with the index 3 after the change) and the vertex with the index 3 before the change (e.g., the vertex with the index 2 after the change).

[0131] According to an embodiment, in the second case 705, the heading of the object frame of the object may be included in the range of values ​​obtained by adding or subtracting a specified value at about +45 degrees in the +x-axis direction (e.g., greater than or equal to about 43 degrees and less than about 47 degrees).

[0132] When the trajectory heading of the object frame is included in a specified range (e.g., a range between about 133 degrees and about 137 degrees from the +x-axis direction) and the representative moving direction corresponds to a specified direction (e.g., an F / B direction opposite to the moving direction of the host vehicle), the processor of the object recognition device may re-recognize the index corresponding to the object frame. In addition, the processor of the object recognition device may recognize the center point between the vertex with the index 0 after the change and the vertex with the index 3 after the change as the representative point, rather than the center point between the vertex with the index 3 before the change (e.g., the vertex with the index 2 after the change) and the vertex with the index 0 before the change (e.g., the vertex with the index 3 after the change).

[0133] According to an embodiment, in the third case 707, the heading of the object frame of the object may be included in the range of values ​​obtained by adding or subtracting a specified value at approximately +45 degrees in the +x-axis direction (e.g., greater than or equal to approximately 43 degrees and less than approximately 47 degrees).

[0134] When the trajectory heading of the object frame is included in the specified range (e.g., a range between about 43 degrees and about 47 degrees from the +x-axis direction) and the representative moving direction corresponds to the specified direction (e.g., the RR direction located to the right of the moving direction of the host vehicle), the processor of the object recognition device may re-recognize the index corresponding to the object frame. In addition, the processor of the object recognition device may recognize the center point between the vertex with the index 2 after the change and the vertex with the index 3 after the change as the representative point, rather than the center point between the vertex with the index 2 before the change (e.g., the vertex with the index 1 after the change) and the vertex with the index 3 before the change (e.g., the vertex with the index 2 after the change).

[0135] According to an embodiment, in the fourth case 709, the heading of the object frame of the object may be included in the range of values ​​obtained by adding or subtracting a specified value at approximately +45 degrees in the +x-axis direction (e.g., greater than or equal to approximately 43 degrees and less than approximately 47 degrees).

[0136] When the trajectory heading of the object frame is included in a specified range (e.g., a range between about -133 degrees and about -137 degrees from the +x-axis direction) and the representative moving direction corresponds to a specified direction (e.g., the LL direction located to the left of the moving direction of the host vehicle), the processor of the object recognition device may re-recognize the index corresponding to the object frame. In addition, the processor of the object recognition device may recognize the center point between the vertex with the index 1 after the change and the vertex with the index 0 after the change as the representative point, rather than the center point between the vertex with the index 1 before the change (e.g., the vertex with the index 0 after the change) and the vertex with the index 0 before the change (e.g., the vertex with the index 3 after the change).

[0137] Figure 8 An example of a representative point identified based on a heading, a track heading, and a representative moving direction in an object recognition apparatus or an object recognition method according to an embodiment of the present invention is shown.

[0138] refer to Figure 8 , the heading of the object frame of the object 801 may be included in the second reference range. The first case 803, the second case 805, the third case 807, and the fourth case 809 may represent an object whose track heading is included in the specified range and whose moving direction is the specified direction.

[0139] According to an embodiment, the heading of the object frame of the object 801 may be included in a range of values ​​obtained by adding or subtracting a specified value at about -45 degrees in the +x-axis direction. For example, the heading of the object frame of the object 801 may be included in a range greater than or equal to about -43 degrees and less than about -47 degrees along the +x-axis direction, but embodiments of the present invention are not limited thereto.

[0140] According to an embodiment, when the representative moving direction of the object 801 is the F / B direction opposite to the moving direction of the host vehicle, the representative point may correspond to the center point of a line segment formed by vertices of the object frame with index 0 and vertices of the object frame with index 3.

[0141] According to an embodiment, when the heading of the object is included in the second reference range, the processor of the object recognition device may recognize a representative point different from the representative point recognized according to the representative moving direction and the index in four cases. The four cases may include a first case 803, a second case 805, a third case 807, and a fourth case 809.

[0142] According to an embodiment, the processor of the object recognition device may identify the trajectory heading of the object frame in the first frame based on the heading of the first frame and the heading of the object frame in at least one frame before the first frame. For example, in two or more frames before the first frame, when the heading of the object frame is about -45 degrees with the +x-axis direction, the trajectory heading of the object frame may be about -45 degrees.

[0143] According to an embodiment, in the first case 803, the heading of the object frame of the object may be included in the range of values ​​obtained by adding or subtracting a specified value at about -45 degrees in the +x-axis direction (e.g., greater than or equal to about -43 degrees and less than about -47 degrees).

[0144] When the trajectory heading of the object frame is included in a specified range (e.g., a range between about 43 degrees and about 47 degrees from the +x-axis direction) and the representative moving direction corresponds to a specified direction (e.g., the same F / F direction as the moving direction of the host vehicle), the processor of the object recognition device may re-recognize the index corresponding to the object frame. In addition, the processor of the object recognition device may recognize the center point between the vertex with the index 0 after the change and the vertex with the index 3 after the change as the representative point, rather than the center point between the vertex with the index 3 before the change (e.g., the vertex with the index 0 after the change) and the vertex with the index 0 before the change (e.g., the vertex with the index 1 after the change).

[0145] According to an embodiment, in the second case 805, the heading of the object frame of the object may be included in the range of values ​​obtained by adding or subtracting a specified value at about -45 degrees in the +x-axis direction (e.g., greater than or equal to about -43 degrees and less than about -47 degrees).

[0146] When the trajectory heading of the object frame is included in a specified range (e.g., a range between about -133 degrees and about -137 degrees from the +x-axis direction) and the representative moving direction corresponds to a specified direction (e.g., an F / B direction opposite to the moving direction of the host vehicle), the processor of the object recognition device may re-recognize the index corresponding to the object frame. In addition, the processor of the object recognition device may recognize the center point between the vertex with the index 0 after the change and the vertex with the index 3 after the change as the representative point, rather than the center point between the vertex with the index 0 before the change (e.g., the vertex with the index 1 after the change) and the vertex with the index 3 before the change (e.g., the vertex with the index 0 after the change).

[0147] According to an embodiment, in the third case 807, the heading of the object frame of the object may be included in the range of values ​​obtained by adding or subtracting a specified value at about -45 degrees in the +x-axis direction (e.g., greater than or equal to about -43 degrees and less than about -47 degrees).

[0148] When the trajectory heading of the object frame is included in a specified range (e.g., a range between about -43 degrees and about -47 degrees from the +x-axis direction) and the representative moving direction corresponds to a specified direction (e.g., the RR direction located to the right of the moving direction of the host vehicle), the processor of the object recognition device may re-recognize the index corresponding to the object frame. In addition, the processor of the object recognition device may recognize the center point between the vertex with the index 2 after the change and the vertex with the index 3 after the change as the representative point, rather than the center point between the vertex with the index 3 before the change (e.g., the vertex with the index 0 after the change) and the vertex with the index 2 before the change (e.g., the vertex with the index 3 after the change).

[0149] According to an embodiment, in the fourth case 809, the title of the object frame of the object may be included in the range of values ​​obtained by increasing or decreasing the specified value of approximately -45 degrees in the +x-axis direction (e.g., greater than or equal to approximately -43 degrees and less than approximately -47 degrees).

[0150] When the trajectory heading of the object frame is included in a specified range (e.g., a range between about 133 degrees and about 137 degrees from the +x-axis direction) and the representative moving direction corresponds to a specified direction (e.g., the LL direction to the left of the moving direction of the host vehicle), the processor of the object recognition device may re-recognize the index corresponding to the object frame. In addition, the processor of the object recognition device may recognize the center point between the vertex with the index 0 after the change and the vertex with the index 1 after the change as the representative point, rather than the center point between the vertex with the index 0 before the change (e.g., the vertex with the index 1 after the change) and the vertex with the index 1 before the change (e.g., the vertex with the index 2 after the change).

[0151] Fig. 9 An example of representative points recognized according to representative moving directions in the object recognition apparatus or the object recognition method according to the embodiment of the present invention is shown.

[0152] refer to Fig. 9 , in screen 901, the first object 903 may cross the y-axis (e.g., intersect). The second object 905 may have a representative moving direction opposite to the moving direction of the main vehicle. The third object 907 may have a representative moving direction to the left from the moving direction of the main vehicle (e.g., 90 degrees counterclockwise relative to the moving direction of the main vehicle). The fourth object 909 may have a representative moving direction to the right from the moving direction of the main vehicle (e.g., 90 degrees clockwise relative to the moving direction of the main vehicle). The fifth object 911 may have a representative moving direction that is the same as the moving direction of the main vehicle. According to an embodiment, the direction that is the same as the moving direction of the main vehicle may be referred to as a direction away from the main vehicle in the longitudinal direction (also referred to as a forward longitudinal direction). The second object 905 and the fifth object 911 may each have a longitudinal axis parallel to the main vehicle (e.g., as Fig. 9The third object 907 and the fourth object 909 may each have a direction perpendicular to the longitudinal axis of the host vehicle or parallel to the lateral axis of the host vehicle (e.g., Fig. 9 Representative movement direction (Y-axis shown).

[0153] According to an embodiment, the processor of the object recognition device may recognize a representative point of the first object 903 based on a line segment that leads in a direction away from the host vehicle in the longitudinal direction and is one of the line segments constituting the object frame. In other words, the representative point of the first object 903 may be recognized based on a line segment that leads in a moving direction of the host vehicle. The representative point of the first object 903 may be located between a vertex assigned with index 1 and a vertex assigned with index 2.

[0154] According to an embodiment, the processor of the object recognition device may recognize the representative point of the second object 905 based on a line segment that leads in a direction close to the main vehicle in the longitudinal direction and is one of the line segments constituting the object frame. In other words, the representative point of the second object 905 may be recognized based on a line segment that leads in a direction opposite to the moving direction of the main vehicle (also referred to as a rear longitudinal direction). Among the line segments constituting the object frame, the line segment may have the smallest longitudinal distance to the main vehicle (e.g., the line segment that is closest to the lateral axis of the main vehicle, e.g., Fig. 9 The representative point of the second object 905 may be located between the vertex assigned with index 0 and the vertex assigned with index 3.

[0155] According to an embodiment, the processor of the object recognition device may recognize the representative point of the third object 907 based on a line segment that leads in the left direction relative to the moving direction of the host vehicle and is one of the line segments constituting the object frame. Among the line segments constituting the object frame, the line segment may have the smallest lateral distance to the host vehicle (e.g., the line segment closest to the longitudinal axis of the host vehicle, e.g., Fig. 9 The representative point of the third object 907 may be located between the vertex assigned with index 1 and the vertex assigned with index 0.

[0156] According to an embodiment, the processor of the object recognition device may recognize a representative point of the fourth object 909 based on a line segment that leads in the right direction relative to the moving direction of the host vehicle and is one of the line segments constituting the object frame. Among the line segments constituting the object frame, the line segment may have the smallest lateral distance to the host vehicle (e.g., the line segment closest to the longitudinal axis of the host vehicle, e.g., Fig. 9 The representative point of the fourth object 909 may be located between the vertex assigned with index 2 and the vertex assigned with index 3.

[0157] According to an embodiment, the processor of the object recognition device may recognize the representative point of the fifth object 911 based on a line segment that is along a direction away from the main vehicle in the longitudinal direction and is one of the line segments constituting the object frame. In other words, the representative point of the fifth object 911 may be recognized based on a line segment along a direction opposite to the moving direction of the main vehicle. Among the line segments constituting the object frame, the line segment may have the smallest longitudinal distance to the main vehicle (e.g., the line segment closest to the lateral axis of the main vehicle, e.g., Fig. 9 The representative point of the fifth object 911 may be located between the vertex assigned with index 0 and the vertex assigned with index 3.

[0158] The processor of the object recognition device may recognize a representative point representing an object (eg, first object 903 , second object 905 , third object 907 , fourth object 909 , or fifth object 911 ) based on a center point of one of the line segments constituting the object frame or any other point of the line segment.

[0159] According to one embodiment, the processor of the object recognition device may assign a specified value (e.g., 0) to a vertex in one direction (e.g., left) of a line segment that is closest to the y-axis and forms a small angle with the y-axis among the line segments constituting the object frame. The processor of the object recognition device may sequentially assign indexes in a rotation direction (e.g., clockwise) starting from the vertex indexed with the specified value (e.g., 0).

[0160] Fig.10 An example of representative points provided by an existing object recognition device or object recognition method and an example of representative points provided by an object recognition device or object recognition method according to an embodiment of the present invention are shown.

[0161] refer to Fig.10 , the first path 1001 may represent a path of a representative point changing over time, which is recognized by the object recognition device according to the embodiment. The second path 1011 may represent a path of a representative point changing over time recognized by an existing object recognition device.

[0162] According to an embodiment, the processor of the object recognition device may recognize the representative point of the object according to the moving direction of the object and the indexes assigned to the vertices of the object frame.

[0163] In the first path 1001, the processor of the object recognition device can identify a representative moving direction based on the credibility of the moving direction, and since the index assigned to the object frame vertex also changes when the specified condition is met, a representative point different from the representative point in the second path 1011 can be identified.

[0164] Unlike the representative points in the second path 1011 , the representative points in the first path 1001 may be continuously identified.

[0165] Fig.11 Examples of an object path identified by an existing object recognition device or object recognition method and an object path identified by an object recognition device or object recognition method according to an embodiment of the present invention are shown.

[0166] refer to Fig.11 , the first screen 1101 may display the path of the representative points recognized by the existing object recognition device over time. The first screen 1101 may include a first path 1105 and a second path 1107, the first path 1105 being the path of the representative points recognized for the first object, and the second path 1107 being the path of the representative points recognized for the second object. The first path 1105 and the second path 1107 may be recognized by the object recognition device of the host vehicle 1103.

[0167] The second screen 1111 may display a path of representative points recognized by the object recognition device according to the embodiment over a period of time. The second screen 1111 may include a third path 1113 and a fourth path 1115, the third path 1113 being a path of representative points recognized for the first object, and the fourth path 1115 being a path of representative points recognized for the second object. The third path 1113 and the fourth path 1115 may be recognized by the object recognition device of the host vehicle 1103.

[0168] In the third path 1113, the processor of the object recognition device can identify a representative moving direction based on the credibility of the moving direction, and since the index assigned to the object frame vertices also changes when the specified conditions are met, a representative point different from the representative point in the first path 1015 can be identified.

[0169] In the fourth path 1115, the processor of the object recognition device can identify the representative movement direction based on the credibility of the movement direction, and since the index assigned to the object frame vertex also changes when the specified conditions are met, a representative point different from the representative point in the second path 1107 can be identified.

[0170] Unlike the representative points in the first path 1105 and the representative points in the second path 1107 , the representative points in the third path 1113 and the representative points in the fourth path 1115 are identified in sequence.

[0171] Fig.12 An example of an object path recognized by a conventional object recognition device or object recognition method and an example of an object path recognized by an object recognition device or object recognition method according to an embodiment of the present invention are shown.

[0172] refer to Fig.12, the first screen 1201 may display the path of the representative points recognized by the existing object recognition device over time. The first screen 1201 may include a first path 1205, which is the path of the representative points recognized for the first object, and a second path 1207, which is the path of the representative points recognized for the second object. The first path 1205 and the second path 1207 may be recognized by the object recognition device of the host vehicle 1203.

[0173] The second screen 1211 may display a path of representative points recognized by the object recognition device according to the embodiment over a period of time. The second screen 1211 may include a third path 1213 and a fourth path 1215, the third path 1213 being a path of representative points recognized for the first object, and the fourth path 1215 being a path of representative points recognized for the second object. The third path 1213 and the fourth path 1215 may be recognized by the object recognition device of the host vehicle 1203.

[0174] In the third path 1213, the processor of the object recognition device can identify a representative moving direction based on the credibility of the moving direction, and since the index assigned to the object frame vertex also changes when the specified conditions are met, a representative point different from the representative point in the first path 1205 can be identified.

[0175] In the fourth path 1215, the processor of the object recognition device can identify the representative movement direction based on the credibility of the movement direction, and since the index assigned to the object frame vertex also changes when the specified condition is met, a representative point different from the representative point in the second path 1207 can be identified.

[0176] Unlike the representative points in the first path 1205 and the representative points in the second path 1207 , the representative points in the third path 1213 and the representative points in the fourth path 1215 are identified in sequence.

[0177] Fig.13 Examples of an object path identified by an existing object recognition device or object recognition method and an object path identified by an object recognition device or object recognition method according to an embodiment of the present invention are shown.

[0178] refer to Fig.13 , the first screen 1301 may display a path of a representative point recognized by an existing object recognition device over time. The first screen 1301 may include a first path 1305, which is a path of a representative point recognized for a first object. The first path 1305 may be recognized by an object recognition device of a host vehicle 1303.

[0179] The second screen 1311 may display a path of representative points recognized by the object recognition device according to the embodiment over a period of time. The second screen 1311 may include a second path 1313, which is a path of representative points recognized for the first object. The second path 1313 may be recognized by the object recognition device of the host vehicle 1303.

[0180] In the second path 1313, the processor of the object recognition device can identify a representative moving direction based on the credibility of the moving direction, and since the index assigned to the object frame vertex also changes when the specified condition is met, a representative point different from the representative point in the first path 1305 can be identified.

[0181] Unlike the representative points in the first path 1305 , the representative points in the second path 1313 may be continuously identified.

[0182] Fig.14 A computing system related to an object recognition apparatus and an object recognition method according to an embodiment of the present invention is shown.

[0183] refer to Fig.14 , the computing system 1400 may include at least one processor 1410 , a memory 1430 , a user interface input device 1440 , a user interface output device 1450 , a storage device 1460 , and a network interface 1470 , which are connected to each other via a bus 1420 .

[0184] The processor 1410 may be a central processing unit (CPU) or a semiconductor device for processing instructions stored in the memory 1430 and / or the storage device 1460. The memory 1430 and the storage device 1460 may include various types of volatile or non-volatile storage media. For example, the memory 1430 may include a ROM (Read Only Memory) 1431 and a RAM (Random Access Memory) 1432.

[0185] Therefore, the operations of the methods or algorithms described in conjunction with the embodiments disclosed herein may be directly embodied in a hardware or software module, or a combination thereof, executed by the processor 1410. The software module may reside on a storage medium (i.e., memory 1430 and / or storage device 1460), such as RAM, flash memory, ROM, EPROM, EEPROM, registers, hard disk, removable disk, and CD-ROM.

[0186] An exemplary storage medium may be coupled to the processor 1410, and the processor 1410 may read information from the storage medium and record the information in the storage medium. Alternatively, the storage medium may be integrated with the processor 1410. 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.

[0187] The above description merely illustrates the technical concept of the present invention, and those skilled in the art to which the present invention belongs may make various modifications and changes without departing from the basic features of the present invention.

[0188] Therefore, the embodiments disclosed in the present invention are not intended to limit the technical concept of the present invention, but to describe the present invention, and the scope of the technical concept of the present invention is not limited by the embodiments. The protection scope of the present invention should be interpreted by the following claims, and all technical ideas within the scope equivalent to them should be interpreted as included in the scope of the present invention.

[0189] This technology can improve the accuracy of determining whether the identified object is a moving object or a movable stationary object by using the object's moving direction that is greater than a certain confidence value.

[0190] This technology can identify whether an object is a moving object or a movable stationary object by tracking representative points that represent specific points of the object.

[0191] In addition, this technology can enhance the user experience by improving the accuracy of determining whether an object is a moving object or a movable stationary object.

[0192] In addition, this technology can improve the performance of autonomous driving or driver-assisted driving by improving the accuracy of determining whether an object is a moving object or a movable stationary object.

[0193] Furthermore, various effects can be provided, which can be directly or indirectly realized by the present invention.

[0194] Although the present invention has been described above with reference to exemplary embodiments and the accompanying drawings, the present invention is not limited thereto and those skilled in the art may make various modifications and changes to the present invention without departing from the spirit and scope of the invention as claimed in the appended claims.

Claims

1. An object recognition device, comprising: Sensors associated with the vehicle; and A processor configured to: determining, via the sensor, an object in a first frame; determining a first moving direction of the object in the first frame; determining a credibility value of the first moving direction; A point representing the object in the first frame is determined based on a representative moving direction of the object, wherein: If the confidence value is greater than a threshold, the representative moving direction is the first moving direction, or If the confidence value is less than or equal to the threshold, the representative movement direction is a second movement direction of the object in a second frame preceding the first frame; as well as Based on a first position of the point in the first frame and a second position of the point in at least one frame before the first frame, a signal indicating that the object is a moving object or a movable stationary object is output.

2. The object recognition device according to claim 1, wherein: The processor is configured to determine the point by: determining an object frame representing the object in the first frame; determining a line segment associated with the object frame based on the representative movement direction; as well as The point is further determined based on the line segment.

3. The object recognition device according to claim 1, wherein: The processor is configured to determine the point by: determining an object frame representing the object in the first frame; determining a first heading of the object frame in the first frame based on a third position of the object in the first frame and a fourth position of the object in at least one frame before the first frame; Determine a trajectory heading of the object frame in the first frame based on the first heading and a second heading of the object frame in at least one frame before the first frame; determining a line segment associated with the object frame based on the first heading being within a first range, the trajectory heading being within a second range, and the representative moving direction being a specified direction; as well as The point is further determined based on the line segment.

4. The object recognition device according to claim 1, wherein: The processor is configured to determine the point by: determining an object frame representing the object in the first frame; assigning a different index from a plurality of indices to each vertex from a plurality of vertices forming the object box; as well as The point is determined further based on the plurality of indices.

5. The object recognition device according to claim 1, wherein: The processor is configured to determine the point by: determining an object frame representing the object in the first frame; determining a first heading of the object frame based on a first position of the object in the first frame and a second position of the object in at least one frame before the first frame; Determine a trajectory heading of the object frame in the first frame based on the first heading and a second heading of the object frame in at least one frame before the first frame; assigning a different index from a plurality of indexes to each of a plurality of vertices forming the object frame based on the first heading being within a first range, the trajectory heading being within a second range, and the representative moving direction being a specified direction; as well as The point is determined further based on the plurality of indices.

6. The object recognition device according to claim 1, wherein: The processor is configured to determine the point by: determining an object frame representing the object in the first frame; and The point is determined based on one of the following: a first line segment, which has a minimum longitudinal distance to the vehicle among the line segments associated with the object frame based on the representative movement direction being parallel to the longitudinal axis of the vehicle, or A second line segment is perpendicular to a longitudinal axis of the vehicle based on the representative moving direction, and the second line segment has a minimum lateral distance to the vehicle among the line segments associated with the object frame.

7. The object recognition device according to claim 1, wherein: The processor is configured to determine the point by: determining an object frame representing the object in the first frame; and The point is determined based on a center point of one of the line segments constituting the object frame.

8. The object recognition device according to claim 1, wherein: The processor is configured to determine the first moving direction of the object in the first frame based on a value obtained by adding a first distance moved by the vehicle and a second distance moved by the object relative to the vehicle, wherein the second distance is determined based on a difference between the first position and the second position.

9. The object recognition device according to claim 1, wherein: The processor is configured to determine whether the object is a moving object or a movable stationary object further based on a velocity of the object, wherein the velocity of the object is determined based on the first position and the second position.

10. The object recognition device according to claim 1, wherein: The processor is further configured to assign to the object one of: a first identifier indicating that the object is a mobile object, or A second identifier indicating that the object is a movable stationary object.

11. A method for object recognition executed by a processor, the method comprising the following steps: determining, via a sensor associated with the vehicle, an object in a first frame; determining a first moving direction of the object in the first frame; determining a credibility value of the first moving direction; A point representing the object in the first frame is determined based on a representative moving direction of the object, wherein: If the confidence value is greater than a threshold, the representative moving direction is the first moving direction, or If the confidence value is less than or equal to the threshold, the representative movement direction is a second movement direction of the object in a second frame preceding the first frame; as well as Based on a first position of the point in the first frame and a second position of the point in at least one frame before the first frame, a signal indicating that the object is a moving object or a movable stationary object is output.

12. The object recognition method according to claim 11, wherein: The steps of determining the point include: determining an object frame representing the object in the first frame; determining a line segment associated with the object frame based on the representative movement direction; and The point is further determined based on the line segment.

13. The object recognition method according to claim 11, wherein: The steps of determining the point include: determining an object frame representing the object in the first frame; determining a first heading of the object frame in the first frame based on a third position of the object in the first frame and a fourth position of the object in at least one frame before the first frame; Determine a trajectory heading of the object frame in the first frame based on the first heading and a second heading of the object frame in at least one frame before the first frame; determining a line segment associated with the object frame based on that the first heading is within a first range, the track heading is within a second range, and the representative moving direction is a specified direction; and The point is further determined based on the line segment.

14. The object recognition method according to claim 11, wherein: The steps of determining the point include: determining an object frame representing the object in the first frame; assigning a different index from among a plurality of indices to each vertex from among a plurality of vertices forming the object box; and The point is determined further based on the plurality of indices.

15. The object recognition method according to claim 11, wherein: The steps of determining the point include: determining an object frame representing the object in the first frame; determining a first heading of the object frame based on a first position of the object in the first frame and a second position of the object in at least one frame before the first frame; Determine a trajectory heading of the object frame in the first frame based on the first heading and a second heading of the object frame in at least one frame before the first frame; assigning a different index from a plurality of indexes to each of a plurality of vertices forming the object frame based on the first heading being within a first range, the trajectory heading being within a second range, and the representative moving direction being a specified direction; and The point is determined further based on the plurality of indices.

16. The object recognition method according to claim 11, wherein: The steps of determining the point include: determining an object frame representing the object in the first frame; and The point is determined based on one of the following: a first line segment, which has a minimum longitudinal distance to the vehicle among the line segments associated with the object frame based on the representative movement direction being parallel to the longitudinal axis of the vehicle, or A second line segment is perpendicular to a lateral axis of the vehicle based on the representative moving direction, and the second line segment has a minimum lateral distance to the vehicle among the line segments associated with the object frame.

17. The object recognition method according to claim 11, wherein: The steps of determining the point include: determining an object frame representing the object in the first frame; and The point is determined based on a center point of one of the line segments constituting the object frame.

18. The object recognition method according to claim 11, wherein: The step of determining the first moving direction of the object in the first frame comprises: The first moving direction of the object in the first frame is determined based on a value obtained by adding a first distance moved by the vehicle and a second distance moved by the object relative to the vehicle, wherein the second distance is determined based on a difference between the first position and the second position.

19. The object recognition method according to claim 11, further comprising the following steps: Whether the object is a moving object or a movable stationary object is further determined based on a velocity of the object, wherein the velocity of the object is determined based on the first position and the second position.

20. The object recognition method according to claim 11, further comprising the following steps: Assign one of the following to the object: a first identifier indicating that the object is a mobile object, or A second identifier indicating that the object is a movable stationary object.

Images