Apparatus for recognizing object and method thereof

Abstract

The invention relates to an object recognition device and method. An object recognition apparatus may include a sensor and a processor configured to: identify an object box including a plurality of contour points representing an object based on sensing information of the sensor; determining a plurality of minimum values, wherein each of the plurality of minimum values is the minimum value of the distances between one of the plurality of contour points and a plurality of line segments forming the boundary of the object box; determining a degree of mismatch between the object frame and the plurality of contour points based on at least one of the plurality of minimum values and the number of the plurality of contour points; and outputting a signal for indicating whether the object is a stationary object based on the degree of mismatch.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of priority from Korean Patent Application No. 10-2023-0128421 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 apparatus and method, and more particularly, to a technology for recognizing features of an object based on the distribution of contour points obtained by a sensor (eg, a light detection and ranging (LIDAR) sensor). Background Art

[0004] Autonomous vehicles or vehicles with driver assistance devices can understand the surrounding environment through sensors (such as LiDAR) and collect data for driving.

[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 identify information of an object represented by a point based on the acquired movement information of the point. In particular, a technology for identifying 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 can be used to ensure the stability of autonomous driving or driver assistance driving and reduce the risk of accidents. Summary of the invention

[0007] According to the present invention, a device may include a sensor and a processor, wherein the processor is configured to identify an object frame containing multiple contour points representing an object based on sensing information of the sensor; determine multiple minimum values, wherein each of the multiple minimum values ​​is the minimum value of the distance between a contour point among the multiple contour points and multiple line segments constituting the boundary of the object frame; determine a mismatch between the object frame and the multiple contour points based on at least one of the multiple minimum values ​​and the number of the multiple contour points; and output a signal indicating whether the object is a stationary object based on the mismatch.

[0008] In an embodiment of a device, the processor may be configured to: determine the sum of the distances between the Nth contour point and multiple line segments based on the minimum value of the distances between the Nth contour point and multiple line segments, wherein the Nth contour point is one contour point from the first contour point to the Mth contour point representing the object, and wherein N is a natural number satisfying 1≦N≦M, and M is a natural number greater than 2; determine the second sum of the distances between the N+1th contour point and multiple line segments based on the sum of the distances and the minimum value of the distances between the N+1th contour point and multiple line segments, wherein the N+1th contour point is the next contour point of the Nth contour point; and determine the degree of mismatch based on the sum of the distances between the Mth contour point and multiple line segments.

[0009] In an embodiment of a device, the processor may be configured to: determine a minimum value among distances between an Nth contour point and a plurality of line segments, wherein N is a natural number satisfying 1≦N≦M, and the Nth contour point is a contour point from a first contour point to an Mth contour point representing an object, wherein M is a natural number greater than 2; and determine a degree of mismatch based on a sum of a plurality of minimum values, wherein each of the plurality of minimum values ​​is a minimum value among distances between each contour point from a first contour point to an Mth contour point of the plurality of contour points and a plurality of line segments.

[0010] In an embodiment of a device, the processor may be configured to: determine the sum of distances associated with the Nth layer based on the sum of minimum values ​​of distances between contour points and multiple line segments included in the Nth layer, wherein N is a natural number satisfying 1≦N≦M, and the Nth layer is one of the first layer to the Mth layer included in the object frame, and wherein M is a natural number greater than or equal to 2; determine the sum of distances associated with the Nth layer based on the sum of distances associated with the Nth layer and the sum of minimum values ​​of distances between contour points and multiple line segments included in the N+1th layer, wherein the N+1th layer is the next layer of the Nth layer; and determine the degree of mismatch based on the sum of distances associated with the Mth layer and the number of multiple contour points representing the object.

[0011] In an embodiment of a device, the processor may be configured to: determine the sum of distances associated with the Nth layer based on the sum of a set of minimum values, wherein each of the set of minimum values ​​is the minimum value of distances between a contour point among contour points included in the Nth layer and a plurality of line segments, wherein N is a natural number satisfying 1≦N≦M, and the Nth layer is one of the first layer to the Mth layer included in the object frame, and wherein M is a natural number greater than or equal to 2; and determine the degree of mismatch based on a value obtained by adding the distances associated with the first layer to the Mth layer and the number of contour points representing the object.

[0012] In an embodiment of the device, the processor may be configured to determine a distance between a contour point among the plurality of contour points and the line segment based on coordinates of the contour point and coordinates of two points passing through the line segment among the plurality of line segments.

[0013] In an embodiment of the apparatus, the processor may be configured to determine the degree of mismatch based on a value obtained by dividing a sum of the plurality of minimum values ​​by the number of the plurality of contour points representing the object.

[0014] In an embodiment of a device, the object frame may include more than three line segments.

[0015] In an embodiment of the device, the processor may be configured to: assign a confidence value indicating whether the object is a stationary object based on determining that the mismatch is greater than a mismatch threshold; and determine whether the object is a stationary object based on the confidence value.

[0016] In an embodiment of a device, the processor may be configured to: determine a first credibility value for indicating whether an object is a stationary object based on determining a first mismatch as a mismatch; determine a second credibility value greater than the first credibility value based on determining a second mismatch greater than the first mismatch as a mismatch; and determine whether the object is a stationary object based on the first credibility value or the second credibility value.

[0017] In an embodiment of the device, the processor may be configured to: identify the object frame as a designated graphic; and determine whether the object is an object corresponding to the designated graphic based on the mismatch degree.

[0018] In an embodiment of the apparatus, the processor may be configured to assign an identifier to the object based on determining that the object is a stationary object to indicate that the object is a stationary object.

[0019] In an embodiment of the device, the processor may be configured to: determine a confidence value indicating whether the object is a stationary object based on the mismatch degree; and determine whether the object is a stationary object based on a value obtained by multiplying the confidence value by a weight.

[0020] According to the present invention, a method is provided, which may include: identifying an object frame containing multiple contour points representing an object based on sensing information of a sensor; determining multiple minimum values, wherein each of the multiple minimum values ​​is the minimum value of the distance between a contour point among the multiple contour points and multiple line segments constituting the boundary of the object frame; determining a mismatch between the object frame and the multiple contour points based on at least one of the multiple minimum values ​​and the number of the multiple contour points; and outputting a signal indicating whether the object is a stationary object based on the mismatch.

[0021] In an embodiment of a method, determining the degree of mismatch may include: determining the sum of the distances between the Nth contour point and the multiple line segments based on the minimum value of the distances between the Nth contour point and the multiple line segments, wherein N is a natural number satisfying 1≦N≦M, and the Nth contour point is one contour point from the first contour point to the Mth contour point representing the object, and wherein M is a natural number greater than 2; determining a second sum of the distances between the N+1th contour point and the multiple line segments based on the sum of the distances and the minimum value of the distances between the N+1th contour point and the multiple line segments, wherein the N+1th contour point is next to the Nth contour point; and determining the degree of mismatch based on the sum of the distances between the Mth contour point and the multiple line segments.

[0022] In an embodiment of a method, determining the degree of mismatch may include: determining a minimum value among distances between an Nth contour point and a plurality of line segments, wherein N is a natural number satisfying 1≦N≦M, and the Nth contour point is a contour point from a first contour point to an Mth contour point representing an object, wherein M is a natural number greater than 2; and determining the degree of mismatch based on a sum of a plurality of minimum values, wherein each of the plurality of minimum values ​​is a minimum value among distances between each contour point from a first contour point to an Mth contour point of the plurality of contour points and a plurality of line segments.

[0023] In an embodiment of a method, determining the mismatch may include: determining a sum of distances associated with an Nth layer based on a sum of a set of minimum values, wherein each of the set of minimum values ​​is a minimum value of distances between one contour point included in the Nth layer and a plurality of line segments, wherein N is a natural number satisfying 1≦N≦M, and the Nth layer is one of first to Mth layers included in the object frame, and wherein M is a natural number greater than or equal to 2; determining a sum of distances associated with an N+1th layer based on a sum of distances associated with the Nth layer and a sum of a second set of minimum values, wherein each of the second set of minimum values ​​is a minimum value of distances between one contour point included in the N+1th layer and a plurality of line segments, wherein the N+1th layer is a next layer of the Nth layer; and determining the mismatch based on the sum of distances associated with the Mth layer and the number of contour points representing the object.

[0024] In an embodiment of a method, determining the mismatch may include: determining the sum of distances associated with the Nth layer based on the sum of a set of minimum values, wherein each of the set of minimum values ​​is a minimum value of distances between a contour point included in the Nth layer and a plurality of line segments, wherein N is a natural number satisfying 1≦N≦M, and the Nth layer is one of the first layer to the Mth layer included in the object frame, wherein M is a natural number greater than or equal to 2; and determining the mismatch based on a value obtained by adding the distances associated with the first layer to the Mth layer and the number of the plurality of contour points representing the object.

[0025] In an embodiment of a method, determining the plurality of minimum values ​​may include determining a distance between a contour point among the plurality of contour points and a line segment among the plurality of line segments based on coordinates of the contour point and coordinates of two points passing through the line segment.

[0026] In an embodiment of a method, determining the mismatch degree may include determining the mismatch degree based on a value obtained by dividing a sum of the plurality of minimum values ​​by the number of the plurality of contour points representing the object. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The above and other objects, features and advantages of the present invention will become more apparent through the following detailed description in conjunction with the accompanying drawings:

[0028] Figure 1 An example of an object recognition device according to an embodiment of the present invention is shown;

[0029] Figure 2 An example of information required for classifying an object in an object recognition device or an object recognition method according to an embodiment of the present invention is shown;

[0030] Figure 3 An example showing a region where the weight of a reliability value varies according to information related to an object recognition device or an object recognition method according to an embodiment of the present invention;

[0031] Figure 4 An example of an operation flow chart showing, in an object recognition device or an object recognition method according to an embodiment of the present invention, the object recognition device for recognizing whether an object is an immovable object by recognizing a contour-frame distance;

[0032] Figure 5 An example of distribution of contour points in an object recognition device or an object recognition method according to an embodiment of the present invention is shown;

[0033] Figure 6 An example of distribution of contour points contained in each layer for identifying a matching degree in an object recognition device or an object recognition method according to an embodiment of the present invention is shown;

[0034] Figure 7 An example of a contour-frame distance corresponding to a contour point in an object recognition device or an object recognition method according to an embodiment of the present invention is shown;

[0035] Figure 8 An example of an operation flow chart showing an object recognition device for recognizing whether an object is an immovable object according to a mismatch degree in an object recognition device or an object recognition method according to an embodiment of the present invention;

[0036] Fig. 9 An example of distribution of contour points representing an object that cannot move in an object recognition device or an object recognition method according to an embodiment of the present invention is shown;

[0037] Fig.10 An example of distribution of contour points representing an object that cannot move in an object recognition device or an object recognition method according to an embodiment of the present invention is shown;

[0038] Fig.11 An example of distribution of contour points representing an object that cannot move in an object recognition device or an object recognition method according to an embodiment of the present invention is shown;

[0039] Fig.12 An example of illustrating distribution of contour points representing an object that cannot move in an object recognition device or an object recognition method according to an embodiment of the present invention; and

[0040] Fig.13 An example of a computing system related to an object recognition device or an object recognition method according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0041] 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.

[0042] 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.

[0043] 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.

[0044] 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).

[0045] In the following, reference will be made to Figures 1 to 13 Embodiments of the present invention are described in detail.

[0046] Figure 1 An example of an object recognition device according to an embodiment of the present invention is shown.

[0047] refer to Figure 1 , the object recognition device 101 according to an 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.

[0048] refer to Figure 1 , the object recognition device 101 may include a sensor (eg, LIDAR 103 ) and a processor 105 .

[0049] According to an embodiment, the processor 105 of the object recognition device 101 may acquire a point cloud representing the object through the LIDAR 103. The processor 105 of the object recognition device 101 may recognize contour points among the points included in the point cloud.

[0050] According to an embodiment, the contour point may be acquired by LiDAR. For example, the contour point may be identified in each of the layers formed on the z-axis among the x-axis, the y-axis, and the z-axis. For example, the contour point may be acquired based on the representative point included in the point cloud in each of the layers formed on the z-axis among the x-axis, the y-axis, and the z-axis. For example, the representative point may include all or part of the external points located in the plurality of points included in the point cloud. For example, the point cloud may be acquired by performing clustering based on each of the plurality of points acquired by LIDAR identified within a specified distance.

[0051] According to an embodiment, the object frame may include a virtual frame to which information related to an external object (e.g., a vehicle, a person, a tree, a tire, a road sign, etc.) is assigned. For example, the object frame may be referred to as a contour frame. According to an 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.

[0052] According to an embodiment, the processor 105 of the object recognition apparatus 101 may recognize an object frame including contour points obtained by the LIDAR 103 and representing an object. The processor 105 of the object recognition apparatus 101 may recognize a mismatch indicating the degree to which the distribution of the contour points and the object frame do not match each other.

[0053] According to an embodiment, the processor 105 of the object recognition device 101 may identify a contour-frame distance, which is a minimum value among distances between contour points and line segments constituting the object frame, to identify a mismatch. Figures 4 to 8 Operation of an object recognition device is described, wherein a processor of the object recognition device identifies a mismatch based on a contour-box distance.

[0054] According to an embodiment, the processor 105 of the object recognition device 101 may identify the credibility value based on the mismatching degree according to the frame matching information. Figure 4 Operation of an object recognition device is described that identifies a confidence value from box matching information based on a mismatch.

[0055] According to an embodiment, the processor 105 of the object recognition device 101 may calculate a score representing the probability that the object is an immovable object (e.g., a column of a bridge pier, a guardrail) based on the confidence value according to the frame matching information of the object. Figure 2A score describing the probability of identifying an indicated object as an immovable object (e.g., a stationary object, an immovable object, a fixed object, an immovable object, a non-moving object, a stuck object, a non-moving object, a static object, an object anchored in place, or a rigid object, etc.) based on a confidence value according to box matching information.

[0056] According to an embodiment, the processor 105 of the object recognition device 101 may identify an object as an immovable object based on the score indicating the probability that the object is a moving object (e.g., a moving vehicle) or an object capable of being in a moving state (e.g., a stationary vehicle) is less than the score indicating the probability that the object is an immovable object (e.g., a road sign, a guardrail, a tree, etc.).

[0057] According to an embodiment, the processor 105 of the object recognition device 101 may assign an identifier to the object based on recognizing the object as an immovable object to indicate that the object is an immovable object. The identifier may be referred to as a tag, but is not limited thereto.

[0058] Figure 2 An example of information required for classifying an object in an object recognition apparatus or an object recognition method according to an embodiment of the present invention is shown.

[0059] refer to Figure 2 , table 201 may display the type of information used to calculate a score for identifying whether an object is an object that cannot be moved. A stationary score 203 may represent a score for identifying whether an object is an object that cannot be moved (e.g., a stationary object, an immovable object, a fixed object, an immovable object, a non-moving object, a stuck object, an object that does not move, a static object, an object anchored in place, or a rigid object, etc.). The stationary 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 mobile score 205 may represent a score for identifying whether an object is a mobile object or an object that can be in a mobile state. The mobile 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.

[0060] According to an embodiment, the first credibility may be a credibility for determining whether the object is an object that cannot be moved. The processor of the object recognition device may identify the static score 203 by summing the values ​​obtained by multiplying the first credibility indicated by the plurality of information by the weight. The second credibility may be a credibility for determining whether the object is a mobile object or an object that can be in a mobile state. The processor of the object recognition device may identify the mobile score 205 by summing the values ​​obtained by multiplying the second credibility indicated by the plurality of information by the weight.

[0061] According to an embodiment, the lane-outside information 211 for identifying the stationary score 203 may indicate a first confidence level assigned based on whether the object is identified outside the lane. The frame size information 213 for identifying the stationary score 203 may indicate a first confidence level 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 stationary score 203 may indicate a first confidence level assigned based on the distribution of contour points and the matching degree of the object frame.

[0062] According to an embodiment, the in-lane information 217 may indicate the second confidence level assigned based on whether the object is recognized in the lane. The tracking information 219 may indicate the second confidence level assigned based on whether the object is moving. The speed information 223 may indicate the second confidence level assigned based on the speed of the object. The boundary object information 227 may indicate the second confidence level assigned based on whether the object is seen and not blocked at the boundary of the field of view.

[0063] According to an embodiment, the stationary score 203 may be identified by summing the values ​​obtained by multiplying the first credibility represented by the plurality of pieces of information by a weight. For example, the stationary score 203 may be identified by summing the values ​​obtained by multiplying the first credibility represented by the plurality of pieces of information by a weight. s1 ), multiplying the first credibility according to the frame size information 213 by the weight corresponding to the frame size information 213 (for example, the weight s2 ), multiplying the first credibility of the frame matching information 215 by the weight corresponding to the frame matching information 215 (eg, weight s3 ) or any combination thereof to identify the static score 203. However, the embodiment of the present invention is not limited thereto. According to an embodiment, the static score 203 may be identified by adding up not only the values ​​obtained by multiplying the information listed in the table 201 by the weights, but also the values ​​obtained by multiplying the information not listed in the table 201 by the weights.

[0064] According to an embodiment, the movement score 205 may be identified by summing the values ​​obtained by multiplying the second credibility represented by the plurality of pieces of information by a weight. For example, the movement score 205 may be identified by summing the values ​​obtained by multiplying the second credibility according to the in-lane information 217 by a weight corresponding to the in-lane information 217 (e.g., a weight D1 ), a value obtained by multiplying the second credibility according to the tracking information 219 by a weight corresponding to the tracking information 219 (eg, a weight D2 ), a value obtained by multiplying the second credibility according to the other lane object information 221 by a weight corresponding to the other lane object information 221 (for example, a weight D3 ), a value obtained by multiplying the second reliability according to the speed information 223 by a weight corresponding to the speed information 223 (for example, a weightD4 ), a value obtained by multiplying the second credibility according to the contour point distribution information 225 by a weight corresponding to the contour point distribution information 225 (for example, a weight D5 ), a value obtained by multiplying the second credibility according to the boundary object information 227 by a weight corresponding to the boundary object information 227 (eg, a weight D6 ) or at least one of the sum of the values ​​obtained by multiplying the weights or any combination thereof to identify the moving score 205. However, the embodiment of the present invention is not limited thereto. According to an embodiment, the moving score 205 can be identified by adding up not only the values ​​obtained by multiplying the information listed in the table 201 by the weights, but also the values ​​obtained by multiplying the information not listed in the table 201 by the weights.

[0065] According to an embodiment, if the movement score 205 of the specific object is higher than the static score 203 of the specific object, the processor of the object recognition device may recognize the specific object as a moving object or an object capable of being in a moving state. According to an embodiment, if the static score 203 of the specific object is higher than the movement score 205 of the specific object, the processor of the object recognition device may recognize the specific object as an object that cannot be moved (e.g., a road sign, a wall, a power line, etc.).

[0066] According to an embodiment of the present invention, the processor of the object recognition device may identify the first credibility indicated by the frame matching information 215. Figures 4 to 8 A method for identifying the first reliability represented by the frame matching information 215 according to an embodiment is described. Hereinafter, the value of the first reliability represented by the frame matching information 215 may be referred to as a reliability value.

[0067] Figure 3 An example of a region where the weight of a reliability value varies according to information related to an object recognition apparatus or an object recognition method according to an embodiment of the present invention is shown.

[0068] refer to Figure 3 , box 301 may represent a first area 305, a second area 307, and a third area 309, which depends on the distance from the main vehicle 303 including the object recognition device. The first area 305 may include an area within the field of view. The second area 307 (e.g., an area in front of the main vehicle 303 but not in the field of view) may include an area for classifying objects of interest. The third area 309 (e.g., an area in front of the main vehicle 303 and farther than the second area 307) may include an area other than the first area 305 and the second area 307.

[0069] 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 classification 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.

[0070] According to an embodiment, the processor of the object recognition device may assign different weights (e.g., Figure 2 ) to identify the static score or the moving 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.

[0071] Figure 4 An example of an operation flow chart of an object recognition device for recognizing whether an object is an immovable object by recognizing a contour-frame distance in an object recognition device or an object recognition method according to an embodiment of the present invention is shown.

[0072] In the following, it is assumed Figure 1 The object recognition device 101 performs Figure 4 In addition, Figure 4 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.

[0073] refer to Figure 4 In a first step 401, a processor of an object recognition apparatus according to an embodiment may recognize a contour-frame distance among distances between a contour point and a plurality of line segments constituting an object frame.

[0074] To identify the contour-box distance, the processor of the object recognition device may identify an object box including a plurality of contour points acquired by a sensor (eg, a LIDAR sensor) and representing the object.

[0075] The processor of the object recognition device can identify the contour-frame distance, which is the minimum value of the distance between each contour point and the multiple line segments that constitute the object frame. Figure 6 to Figure 7 Describes a method for identifying contour-box distances.

[0076] In the second step 403, the processor of the object recognition device according to the embodiment may identify whether there are contour points whose contour-frame distance is not recognized in the layer. If there are contour points whose contour-frame distance is not recognized in the layer, the processor of the object recognition device may perform the first step 401. If there are no contour points whose contour-frame distance is not recognized in the layer, the processor of the object recognition device may identify the third step 405.

[0077] According to an embodiment, the processor of the object recognition apparatus may recognize a contour-frame distance of a contour point in one of a plurality of layers respectively containing the contour points.

[0078] In the third step 405, the processor of the object recognition device according to the embodiment may identify whether there is a layer having contour points with unrecognized contour-frame distance. If there is a layer having contour points with unrecognized contour-frame distance, the processor of the object recognition device may execute the first step 401. If there is no layer having contour points with unrecognized contour-frame distance, the processor of the object recognition device may execute the fourth step 407.

[0079] According to an embodiment, the processor of the object recognition device may recognize the contour-frame distance of the contour point contained in a layer other than the layer for which the contour-frame distance is recognized.

[0080] In a fourth step 407, the processor of the object recognition device according to an embodiment may identify a mismatch degree. The mismatch degree may represent a mismatch degree between the object frame and the contour point based on at least one of a contour-frame distance of the contour point, a number of contour points representing the object, or any combination thereof.

[0081] According to an embodiment, the mismatch may be identified based on the contour-box distance of all contour points representing the object.

[0082] For example, the mismatch degree identified in the first manner may be identified based on the sum of the contour-frame distances from the first contour point to the Mth contour point (where M is a natural number greater than 2). The specific layer may consist of the first to Mth contour points.

[0083] The processor of the object recognition device may identify the degree of mismatch based on a value obtained by dividing the sum of the contour-frame distances of the contour points by the number of contour points representing the object.

[0084] For example, the mismatch degree identified in the second manner may be identified based on the sum of the distances of the last contour point among all contour points representing the object. In other words, the mismatch degree identified in the second manner may be identified based on the sum of the distances of the Mth contour point. The sum of the distances of the (N+1)th contour point may represent a value obtained by adding the sum of the distances of the Nth contour point to the contour-frame distance of the (N+1)th contour point.

[0085] The processor of the object recognition device may identify the degree of mismatch based on a value obtained by dividing the sum of the distances of the Mth contour points by the number of contour points representing the object.

[0086] According to an embodiment, the mismatch may be identified based on contour-box distances of contour points included in a plurality of layers.

[0087] For example, the mismatch degree identified in the third manner may be identified based on the sum of distances from the first layer to the Mth layer. The sum of distances of a specific layer may represent the sum of contour-box distances of contour points included in the specific layer.

[0088] The processor of the object recognition device may identify the degree of mismatch based on a value obtained by dividing the sum of the distances from the first layer to the Mth layer by the number of contour points representing the object.

[0089] For example, the mismatch degree identified in the fourth manner may be identified based on the sum of the distances of the Mth layer.

[0090] The processor of the object recognition device may identify the sum of distances of the Nth layer based on the sum of contour-frame distances of contour points contained in the Nth layer (N is a natural number satisfying 1≦N≦M) of the first to Mth layers (where M is a natural number greater than or equal to 2) contained in the object frame. The processor of the object recognition device may identify the sum of distances of the (N+1)th layer based on the sum of distances of the Nth layer and the sum of contour-frame distances of contour points contained in the (N+1)th layer which is the next layer after the Nth layer.

[0091] In the fifth step 409 , the processor of the object recognition apparatus according to the embodiment may recognize whether the object is an object that cannot be moved based on the mismatch degree.

[0092] According to an embodiment, the processor of the object recognition device may assign a specified value (e.g., 1) as a confidence level indicating whether the object is an object that cannot be moved based on recognizing that the mismatch is greater than a specified mismatch threshold. The processor of the object recognition device may recognize whether the object is an object that cannot be moved based on the confidence level.

[0093] According to an embodiment, the greater the mismatch, the greater the credibility value assigned to the object by the processor of the object recognition device. In other words, the processor of the object recognition device may recognize the first credibility value as the credibility for indicating whether the object is an object that cannot be moved based on recognizing the first mismatch value as the mismatch, and recognize the second credibility value greater than the first credibility value as the credibility based on recognizing the second mismatch value greater than the first mismatch value as the mismatch. The processor of the object recognition device may recognize whether the object is an object that cannot be moved based on the first credibility value or the second credibility value.

[0094] Figure 5 An example of distribution of contour points in an object recognition apparatus or an object recognition method according to an embodiment of the present invention is shown.

[0095] refer to Figure 5 In the first case 501, the processor of the object recognition device may obtain a point cluster (e.g., a first point cloud 505) representing a first object 503. The processor of the object recognition device may recognize a first contour line 507 based on the first point cloud 505. The first contour line 507 may include a line connecting contour points recognized based on the first point cloud 505.

[0096] In the second case 511, the processor of the object recognition device may obtain a second point cluster (e.g., a second point cloud 515) representing a second object 513. The processor of the object recognition device may recognize a second contour line 517 based on the second point cloud 515. The second contour line 517 may include a line connecting contour points recognized based on the second point cloud 515.

[0097] According to an embodiment, the processor of the object recognition device may recognize the mismatch degree of the first object 503 based on the distribution of contour points included in the first contour line 507 .

[0098] The processor of the object recognition device may identify the first object 503 based on the mismatch degree of the first object 503 according to the frame matching information of the first object 503 (for example, Figure 2 The processor of the object recognition device can identify the first object 503 as a moving object (e.g., a moving vehicle) or an object capable of being in a moving state (e.g., a stopped vehicle) based on the frame matching information 215 in the first object 503 according to the frame matching information of the first object 503.

[0099] According to an embodiment, the processor of the object recognition device may recognize the mismatch degree of the second object 513 based on the distribution of contour points included in the second contour line 517. The mismatch degree of the first object 503 may be lower than the mismatch degree of the second object 513.

[0100] The processor of the object recognition device may identify the credibility value as being greater than or equal to the reference value based on the mismatch of the second object 513 according to the frame matching information of the second object 513. The processor of the object recognition device may identify the second object 513 (e.g., a tree) as an object that cannot be moved based on the credibility based on the frame matching information of the second object 513.

[0101] exist Figure 5 In the figure, the object frame is shown as a rectangle, but the embodiment of the present invention is not limited thereto. According to an embodiment, the object frame may be composed of more than three line segments. For example, the object frame may be a triangle or a rectangle. For example, the object frame may be a pentagon, a hexagon, or an octagon.

[0102] Despite Figure 5 The object frame is displayed as identifying whether the object is an object that cannot be moved (e.g., a tree, a road sign, a wall, a power line, etc.) according to the mismatch degree, but the embodiment of the present invention is not limited thereto. According to an embodiment, the processor of the object recognition device may identify the object frame as a specified graphic (e.g., a triangle, a rectangle, a pentagon, a hexagon, an octagon, a circle, an ellipse, a trapezoid, etc.) and identify the mismatch degree representing the degree of mismatch between the object frame and the contour points based on at least one of the contour-frame distance of the contour points representing the object, the number of contour points representing the object, or any combination thereof. The processor of the object recognition device may identify whether the object is an object corresponding to the graphic based on the mismatch degree.

[0103] Figure 6 An example of distribution of contour points included in each layer for identifying a matching degree in an object recognition device or an object recognition method according to an embodiment of the present invention is shown.

[0104] refer to Figure 6 , the first group 601 may include contour points included in a first layer (e.g., layer 0). The second group 603 may include contour points included in a second layer (e.g., layer 1). The third group 605 may include contour points included in a third layer (e.g., layer 2). The fourth group 607 may include contour points included in a fourth layer (e.g., layer 3).

[0105] According to an embodiment, the processor of the object recognition device may identify the degree of mismatch based on the sum of the distances of each layer.

[0106] For example, through the third way (for example, in Figure 4 The mismatch degree identified in the third way given in the description of step 407 of FIG. 4 may be identified based on the sum of the distances from the first layer to the Mth layer. The sum of the distances of a specific layer may represent the sum of the contour-box distances of the contour points included in the specific layer.

[0107] In other words, the sum of the distances of the first layer may include a value obtained by adding the contour-frame distances of the contour points included in the first layer. The sum of the distances of the second layer may include a value obtained by adding the contour-frame distances of the contour points included in the second layer. The processor of the object recognition device may identify the degree of mismatch based on a value obtained by dividing a value obtained by summing the sum of the distances from the first layer to the Mth layer by the number of contour points representing the object.

[0108] For example, through the fourth method (for example, in Figure 4 The mismatch degree identified in the fourth manner given in the description of step 407) can be identified based on the sum of the distances of the Mth layer which is the last layer.

[0109] In other words, the processor of the object recognition device can identify the sum of the distances of the Nth layer based on the sum of the contour-frame distances of the contour points contained in the Nth layer (N is a natural number satisfying 1≦N≦M) of the first to Mth layers (where M is a natural number greater than or equal to 2) contained in the object frame. The processor of the object recognition device can identify the sum of the distances of the (N+1)th layer based on the sum of the distances of the Nth layer and the sum of the contour-frame distances of the contour points contained in the (N+1)th layer which is the next layer after the Nth layer. The processor of the object recognition device can identify the degree of mismatch based on a value obtained by dividing the sum of the distances of the Mth layer by the number of contour points representing the object.

[0110] Figure 7 An example of contour-frame distances corresponding to contour points in an object recognition apparatus or an object recognition method according to an embodiment of the present invention is shown.

[0111] refer to Figure 7 , in a first state 701, the processor of the object recognition device may identify a contour-frame distance of a first contour point. In a second state 703, the processor of the object recognition device may identify a contour-frame distance of a second contour point. In a third state 705, the processor of the object recognition device may identify a contour-frame distance of a third contour point. In a fourth state 707, the processor of the object recognition device may identify a contour-frame distance of a fourth contour point. In a fifth state 709, the processor of the object recognition device may identify a contour-frame distance of a fifth contour point. In a sixth state 711, the processor of the object recognition device may identify a contour-frame distance of a sixth contour point. The first contour point, the second contour point, the third contour point, the fourth contour point, the fifth contour point, and the sixth contour point may be included in a specific layer.

[0112] According to an embodiment, in order to obtain the distance between a contour point (such as the first contour point, the second contour point, the third contour point, the fourth contour point, the fifth contour point, or the sixth contour point) and a line segment (such as a boundary line) constituting an object frame, the processor of the object recognition device may identify the distance between the contour point and the line segment based on the coordinates of the contour point and the coordinates of two points passing through the line segment. For example, if the coordinates of a contour point are (y, x) and the coordinates of two points passing through a line segment are (y 1 , x 1 ) and (y 2 , x 2 ), then the distance (d) between the contour point and the line segment can be identified according to the following equation 1.

[0113] [Equation 1]

[0114]

[0115] According to an embodiment, in the first state 701, the processor of the object recognition device may identify the minimum value of the distance between the first contour point and the line segment constituting the object frame, that is, the contour-frame distance. For example, the distance between the first contour point and the line segment constituting the object frame may be 0.16m, 3.28m, 1.36m, and 0.06m, respectively. Among them, the minimum value 0.06m may be identified as the contour-frame distance. Accordingly, the contour-frame distance of the first contour point may be 0.06m.

[0116] According to an embodiment, in the second state 703, the distances between the second contour point and the line segment constituting the object frame may be 0.53m, 3.32m, 1.0m, and 0.02m, respectively. Among them, the minimum value 0.02m may be identified as the contour-frame distance. Accordingly, the contour-frame distance of the second contour point may be 0.02m.

[0117] According to one embodiment, in the third state 705, the distances between the third contour point and the line segment constituting the object frame may be 1.49 m, 3.32 m, 0.04 m, and 0.02 m, respectively. Among them, the minimum value 0.02 m may be identified as the contour-frame distance. Accordingly, the contour-frame distance of the third contour point may be 0.02 m.

[0118] According to one embodiment, in the fourth state 707, the distances between the fourth contour point and the line segment constituting the object frame may be 1.53m, 2.81m, 0.0m, and 0.53m, respectively. Among them, the minimum value 0.0m may be identified as the contour-frame distance. Accordingly, the contour-frame distance of the fourth contour point may be 0.0m.

[0119] According to an embodiment, in the fifth state 709, the distances between the fifth contour point and the line segments constituting the object frame may be 1.55 m, 0.16 m, 0.02 m, and 3.18 m, respectively. Among them, the minimum value 0.02 m may be identified as the contour-frame distance. Accordingly, the contour-frame distance of the fifth contour point may be 0.02 m.

[0120] According to one embodiment, in the sixth state 711, the distances between the sixth contour point and the line segments constituting the object frame may be 1.52 m, 0.01 m, 0.0 m, and 3.33 m, respectively. Among them, the minimum value 0.0 m may be identified as the contour-frame distance. Accordingly, the contour-frame distance of the sixth contour point may be 0.0 m.

[0121] Figure 8 An example of an operation flow chart of an object recognition device for recognizing whether an object is an immovable object according to a mismatch degree in an object recognition device or an object recognition method according to an embodiment of the present invention is shown.

[0122] In the following, it is assumed Figure 1 The object recognition device 101 performs Figure 8 In addition, Figure 8 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.

[0123] refer to Figure 8 In a first step 801, a processor of an object recognition device according to an embodiment may identify an object frame including a plurality of contour points acquired by LIDAR and representing an object.

[0124] In a second step 803 , the processor of the object recognition apparatus according to an embodiment may identify a contour-frame distance, that is, a minimum value among distances between one contour point among a plurality of contour points and a plurality of line segments constituting the object frame.

[0125] In a third step 805, the processor of the object recognition device according to the embodiment may identify a mismatch degree indicating a mismatch degree between the object frame and the contour points based on at least one of the contour-frame distance of the contour points, the number of contour points, or any combination thereof. The number of contour points may be the number of contour points representing the object.

[0126] In a fourth step 807 , the processor of the object recognition device according to an embodiment may recognize whether the object is an object that cannot be moved based on the mismatch degree.

[0127] Fig. 9 An example of distribution of contour points representing an object that cannot move in an object recognition apparatus or an object recognition method according to an embodiment of the present invention is shown.

[0128] refer to Fig. 9 , in a first situation 901 , an object recognition device included in a host vehicle 903 may recognize a first object 905 (eg, a vehicle) and a second object 907 (eg, a tree).

[0129] The first object 905 may correspond to a front vehicle of the host vehicle 903. The mismatch degree of the first object 905 may be less than or equal to a mismatch degree threshold. Accordingly, the first object 905 may be identified as a moving object or an object capable of being in a moving state.

[0130] The second object 907 may correspond to a group of trees in front of the host vehicle. The mismatch degree of the second object 907 may be greater than the mismatch degree threshold. Accordingly, the second object 907 may be identified as an object that cannot be moved.

[0131] Fig.10 An example of distribution of contour points representing an object that cannot move in an object recognition apparatus or an object recognition method according to an embodiment of the present invention is shown.

[0132] refer to Fig.10 In the first case 1001, the object recognition device included in the host vehicle 1003 may recognize a first object 1005 (e.g., a power line). In the second case 1011, the object recognition device included in the host vehicle 1003 may recognize a second object 1013 (e.g., a road sign). In the third case 1021, the object recognition device included in the host vehicle 1003 may recognize a third object 1023.

[0133] According to an embodiment, the first object 1005 may correspond to a plant in a flower bed located beside a road on which the host vehicle 1003 is traveling. The mismatch degree of the first object 1005 may be greater than a specified mismatch degree threshold. Accordingly, the first object 1005 may be identified as an object that cannot be moved.

[0134] According to an embodiment, the second object 1013 may correspond to a group of trees located beside the road on which the host vehicle 1003 is traveling. The mismatch degree of the second object 1013 may be greater than a specified mismatch degree threshold. Accordingly, the second object 1013 may be identified as an object that cannot be moved.

[0135] According to an embodiment, the third object 1023 may correspond to a tree located beside the road on which the host vehicle 1003 is traveling. The mismatch degree of the third object 1023 may be greater than a specified mismatch degree threshold. Accordingly, the third object 1023 (eg, a tree) may be identified as an object that cannot be moved.

[0136] Fig.11 An example of distribution of contour points representing an object that cannot move in an object recognition apparatus or an object recognition method according to an embodiment of the present invention is shown.

[0137] refer to Fig.11 In the first case 1101, the object recognition device included in the host vehicle 1103 according to the embodiment may recognize the first object 1105 (e.g., the first pillar of the first bridge). In the second case 1111, the object recognition device included in the host vehicle 1103 may recognize the second object 1113 (e.g., the first pillar of the second bridge) and the third object 1115 (e.g., the second pillar of the second bridge). In the third case 1121, the object recognition device included in the host vehicle 1103 may recognize the fourth object 1123 and the fifth object 1125.

[0138] According to an embodiment, in the first case 1101, the first object 1105 may correspond to a pillar 1107 of a first bridge pier located beside a road on which the host vehicle 1103 is traveling. The mismatch degree of the first object 1105 may be greater than a specified mismatch degree threshold. Accordingly, the first object 1105 may be identified as an object that cannot be moved.

[0139] According to an embodiment, in the second case 1111, the second object 1113 and the third object 1115 may correspond to a column 1117 of a second bridge pier located beside the road on which the host vehicle 1103 is traveling. The mismatch degree of the second object 1113 and the mismatch degree of the third object 1115 may be greater than a specified mismatch degree threshold. Accordingly, the second object 1113 and the third object 1115 may be identified as objects that cannot be moved.

[0140] According to an embodiment, in the third case 1121, the fourth object 1123 and the fifth object 1125 may correspond to a pillar 1127 of a third bridge pier located beside the road on which the host vehicle 1103 is traveling. The mismatch degree of the fourth object 1123 and the mismatch degree of the fifth object 1125 may be greater than a specified mismatch degree threshold. Accordingly, the fourth object 1123 and the fifth object 1125 may be identified as objects that cannot be moved.

[0141] Fig.12 An example of distribution of contour points representing an object that cannot move in an object recognition apparatus or an object recognition method according to an embodiment of the present invention is shown.

[0142] refer to Fig.12 According to an embodiment, in a first case 1201, an object recognition device included in a host vehicle 1203 may recognize a first object 1205 (e.g., a temporary guardrail). In a second case 1211, an object recognition device included in the host vehicle 1203 may recognize a second object 1213 (e.g., a first shock absorbing facility). In a third case 1221, an object recognition device included in the host vehicle 1203 may recognize a third object 1223 (a second shock absorbing facility).

[0143] According to an embodiment, in the first case 1201, the first object 1205 may correspond to a temporary guardrail 1207 located beside the road on which the host vehicle 1203 is traveling. The mismatch degree of the first object 1205 may be greater than a specified mismatch degree threshold. Accordingly, the first object 1205 may be identified as an object that cannot be moved.

[0144] According to an embodiment, in the second case 1211, the second object 1213 may correspond to a shock absorbing facility 1215 of a highway located beside the road on which the host vehicle 1203 is traveling. The mismatch degree of the second object 1213 may be greater than a specified mismatch degree threshold. Accordingly, the second object 1213 may be identified as an object that cannot be moved.

[0145] According to an embodiment, in the third case 1221, the third object 1223 may correspond to a shock absorbing facility 1225 of a highway located beside the road on which the host vehicle 1203 is traveling. The mismatch degree of the third object 1223 may be greater than a specified mismatch degree threshold. Accordingly, the third object 1223 may be identified as an object that cannot be moved.

[0146] Fig.13 An example of a computing system related to an object recognition device or an object recognition method according to an embodiment of the present invention is shown.

[0147] refer to Fig.13 The computing system 1300 may include at least one processor 1310 , a memory 1330 , a user interface input device 1340 , a user interface output device 1350 , a storage device 1360 , and a network interface 1370 , which are interconnected via a bus 1320 .

[0148] The processor 1310 may be a central processing unit (CPU) or a semiconductor device that processes instructions stored in the memory 1330 and / or the storage device 1360. The memory 1330 and the storage device 1360 may include various types of volatile or non-volatile storage media. For example, the memory 1330 may include a ROM (Read Only Memory) 1331 and a RAM (Random Access Memory) 1332.

[0149] Thus, the operations of the methods or algorithms described in conjunction with the examples disclosed herein may be directly embodied in a hardware or software module, or a combination thereof, executed by the processor 1310. The software module may reside on a storage medium (i.e., memory 1330 and / or storage device 1360) such as RAM, flash memory, ROM, EPROM, EEPROM, registers, hard disk, removable disk, and CD-ROM.

[0150] The storage medium of the embodiment may be connected to the processor 1310, and the processor 1310 may read information from the storage medium, and may record information in the storage medium. Alternatively, the storage medium may be integrated with the processor 1310. 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.

[0151] The present invention aims to solve the above-mentioned problems in the prior art while maintaining the advantages of the prior art.

[0152] An embodiment of the present invention provides an object recognition device and method, which are used to recognize whether an object is an immovable object.

[0153] An embodiment of the present invention provides an object recognition device and method, which are used to recognize whether an object is an immovable object according to a mismatch degree recognized based on contour points and an object frame.

[0154] Embodiments of the present invention provide an object recognition device and method for improving the accuracy of determining whether an object is an immovable object.

[0155] 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.

[0156] According to an embodiment of the present invention, an object recognition device includes a LIDAR and a processor.

[0157] According to an embodiment, the processor may identify an object frame obtained by LIDAR and including contour points representing an object, identify a contour-frame distance, which is a minimum value of distances between one of the contour points and a line segment constituting the object frame, identify a mismatch degree representing a degree of mismatch between the object frame and the contour points based on at least one of the contour-frame distances of the contour points, the number of contour points representing the object, or any combination thereof, and identify whether the object is an immovable object based on the mismatch degree.

[0158] According to an embodiment, the processor may identify the sum of the distances of the Nth contour point based on the contour-frame distance of the Nth contour point (N is a natural number satisfying 1≦N≦M) among the first to Mth contour points (where M is a natural number greater than 2) representing the object, identify the sum of the distances of the (N+1)th contour point based on the sum of the distance of the Nth contour point and the contour-frame distance of the (N+1)th contour point which is the next contour point of the Nth contour point, and identify the degree of mismatch based on the sum of the distances of the Mth contour point.

[0159] According to an embodiment, the processor may identify the contour-frame distance of the Nth contour point (N is a natural number satisfying 1≦N≦M) among the first to Mth contour points (where M is a natural number greater than 2) representing the object, and identify the degree of mismatch based on the sum of the contour-frame distances from the first contour point to the Mth contour point.

[0160] According to an embodiment, the processor may identify the sum of distances of the Nth layer based on the sum of contour-frame distances of contour points contained in the Nth layer (N is a natural number satisfying 1≦N≦M) of the first to Mth layers (where M is a natural number greater than or equal to 2) included in the object frame, identify the sum of distances of the (N+1)th layer based on the sum of distances of the Nth layer and the sum of contour-frame distances of contour points contained in the (N+1)th layer which is the next layer after the Nth layer, and identify the degree of mismatch based on the sum of contour-frame distances from the first contour point to the Mth contour point.

[0161] According to an embodiment, the processor may identify the sum of distances of the Nth layer based on the sum of contour-frame distances of contour points contained in the Nth layer (N is a natural number satisfying 1≦N≦M) of the first to Mth layers (where M is a natural number greater than or equal to 2) included in the object frame, and identify the degree of mismatch based on a value obtained by adding the sum of distances from the first layer to the Mth layer and the number of contour points representing the object.

[0162] According to an embodiment, the processor may identify the distance between one contour point and the line segment based on the coordinates of the one contour point and the coordinates of two points passing through one of the line segments constituting the object frame.

[0163] According to an embodiment, the processor may identify the degree of mismatch based on a value obtained by dividing the sum of contour-box distances of contour points by the number of contour points representing the object.

[0164] According to an embodiment, the processor may recognize an object frame composed of more than three line segments.

[0165] According to an embodiment, the processor may assign a specified value as the credibility indicating whether the object is an object that cannot be moved based on recognizing that the mismatch is greater than a specified mismatch threshold, and identify whether the object is an object that cannot be moved based on the credibility.

[0166] According to an embodiment, the processor may identify a first credibility value as a credibility for indicating whether the object is an object that cannot be moved based on identifying a first mismatch as a mismatch, identify a second credibility value greater than the first credibility value as a credibility based on identifying a second mismatch greater than the first mismatch as a mismatch, and identify whether the object is an object that cannot be moved based on the first credibility value or the second credibility value.

[0167] According to an embodiment, the processor may identify the object frame as a specified graphic, identify a mismatch degree indicating the degree of mismatch between the object frame and the contour points based on at least one of the contour-frame distance of the contour points, the number of contour points representing the object, or any combination thereof, and identify whether the object is an object corresponding to the shape based on the mismatch degree.

[0168] According to an embodiment, the processor may assign an identifier indicating that the object is an unmovable object to the object based on recognizing the object as an unmovable object.

[0169] According to an embodiment, the processor may identify the credibility indicating whether the object is an object that cannot be moved based on the mismatch degree, and identify whether the object is an object that cannot be moved based on a value obtained by multiplying the credibility by a weight.

[0170] According to an embodiment of the present invention, an object recognition method includes identifying an object frame obtained by LIDAR and containing contour points representing an object, identifying a contour-frame distance, the contour-frame distance being the minimum value of the distances between one of the contour points and a line segment constituting the object frame, identifying a mismatch degree indicating the degree of mismatch between the object frame and the contour points based on at least one of the contour-frame distances of the contour points, the number of contour points representing the object, or any combination thereof, and identifying whether the object is an immovable object based on the mismatch degree.

[0171] According to an embodiment, identifying a mismatch degree indicating a degree of mismatch between an object frame and a contour point based on at least one of a contour-frame distance of a contour point, the number of contour points representing an object, or any combination thereof may include identifying a sum of distances of an Nth contour point based on a contour-frame distance of an Nth contour point (N is a natural number satisfying 1≦N≦M) among first to Mth contour points (where M is a natural number greater than 2) representing the object, identifying a sum of distances of an (N+1)th contour point based on a sum of the distance of the Nth contour point and a contour-frame distance of an (N+1)th contour point which is a contour point next to the Nth contour point, and identifying a mismatch degree based on the sum of the distances of the Mth contour point.

[0172] According to an embodiment, identifying a mismatch degree indicating the degree of mismatch between an object frame and a contour point based on at least one of a contour-frame distance of a contour point, the number of contour points representing an object, or any combination thereof may include identifying a contour-frame distance of an Nth contour point (N is a natural number satisfying 1≦N≦M) among first to Mth contour points (where M is a natural number greater than 2) representing the object, and identifying the mismatch degree based on the sum of the contour-frame distances from the first contour point to the Mth contour point.

[0173] According to an embodiment, identifying a mismatch degree indicating a degree of mismatch between an object frame and a contour point based on at least one of a contour-frame distance of a contour point, a number of contour points representing an object, or any combination thereof may include identifying a sum of distances of the Nth layer based on a sum of contour-frame distances of contour points contained in an Nth layer (N is a natural number satisfying 1≦N≦M) among first to Mth layers (where M is a natural number greater than or equal to 2) included in the object frame, identifying a sum of distances of the (N+1)th layer based on a sum of distances of the Nth layer and a sum of contour-frame distances of contour points contained in an (N+1)th layer which is a next layer after the Nth layer, and identifying a mismatch degree based on a sum of distances of the Mth layer and the number of contour points representing the object.

[0174] According to an embodiment, identifying a mismatch degree indicating the degree of mismatch between an object frame and a contour point based on at least one of the contour-frame distance of the contour point, the number of contour points representing the object, or any combination thereof may include identifying the sum of distances of the Nth layer based on the sum of contour-frame distances of contour points contained in an Nth layer (N is a natural number satisfying 1≦N≦M) from first to Mth layers (where M is a natural number greater than or equal to 2) included in the object frame, and identifying the mismatch degree based on a value obtained by adding the sum of distances from the first layer to the Mth layer and the number of contour points representing the object.

[0175] According to an embodiment, identifying the minimum value of the distance between one of the contour points and the line segments constituting the object frame, i.e., the contour-frame distance, may include identifying the distance between the contour point and the line segment based on the coordinates of the contour point and the coordinates of two points passing through one of the line segments constituting the object frame.

[0176] According to an embodiment, identifying a mismatch degree indicating a degree of mismatch between an object frame and a contour point based on at least one of a contour-frame distance of the contour point, a number of contour points representing an object, or any combination thereof may include identifying the mismatch degree based on a value obtained by dividing a sum of the contour-frame distances of the contour points by the number of contour points representing the object.

[0177] The above description merely illustrates the technical idea 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.

[0178] Therefore, the examples disclosed in the present invention are not intended to limit the technical ideas of the present invention, but to describe the present invention, and the scope of the technical ideas 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.

[0179] The present technology can improve the accuracy of determination of whether or not an object is an object that cannot be moved based on the mismatch degree.

[0180] In addition, the present technology can identify whether the object is an object that cannot be moved based on the distribution of contour points by identifying whether the distribution of contour points matches the object frame.

[0181] In addition, the present technology may enhance user experience by improving the accuracy of determining whether an object is an object that cannot be moved.

[0182] In addition, the present technology can improve the performance of autonomous driving or driver-assisted driving by improving the determination accuracy of whether an object is an immovable object.

[0183] Furthermore, various effects directly or indirectly understood by the present invention can be provided.

[0184] Although the present invention has been described above with reference to the embodiments and the accompanying drawings, the present invention is not limited thereto but may be variously modified and altered by those skilled in the art without departing from the spirit and scope of the present invention as claimed in the following claims.

Claims

1. A device comprising: sensor; and processor, Wherein, the processor is configured to: identifying an object box including a plurality of contour points representing an object based on the sensed information of the sensor; Determine a plurality of minimum values, wherein each of the plurality of minimum values ​​is respectively: a minimum value of distances between a contour point among the plurality of contour points and a plurality of line segments constituting a boundary of the object frame; determining a mismatch between the object frame and the plurality of contour points based on at least one of the plurality of minimum values ​​and the number of the plurality of contour points; and A signal indicating whether the object is a stationary object is output based on the mismatch degree.

2. The device according to claim 1, wherein: The processor is configured to: Determining a sum of distances between an Nth contour point and the plurality of line segments based on a minimum value of distances between the Nth contour point and the plurality of line segments, wherein the Nth contour point is one of first to Mth contour points representing the object, and wherein N is a natural number satisfying 1≦N≦M, and M is a natural number greater than or equal to 2; determining a second sum of distances between an N+1th contour point and the plurality of line segments based on the sum of the distances and a minimum value of distances between an N+1th contour point and the plurality of line segments, wherein the N+1th contour point is a contour point next to the Nth contour point; and The mismatch degree is determined based on a sum of distances between the Mth contour point and the plurality of line segments.

3. The device according to claim 1, wherein: The processor is configured to: Determining a minimum value of distances between an Nth contour point and the plurality of line segments, wherein N is a natural number satisfying 1≦N≦M, and the Nth contour point is one of the first to Mth contour points representing the object, wherein M is a natural number greater than or equal to 2; and The mismatch degree is determined based on the sum of the multiple minimum values, wherein each of the multiple minimum values ​​is: a minimum value of distances between each contour point from the first contour point to the Mth contour point of the multiple contour points and the multiple line segments.

4. The device according to claim 1, wherein: The processor is configured to: Determining a sum of distances associated with the Nth layer based on a sum of minimum values ​​of distances between contour points included in the Nth layer and the plurality of line segments, wherein N is a natural number satisfying 1≦N≦M, and the Nth layer is one of the first layer to the Mth layer included in the object frame, and wherein M is a natural number greater than or equal to 2; determining a sum of distances associated with an N+1th layer based on a sum of distances associated with the Nth layer and a sum of minimum values ​​of distances between contour points included in an N+1th layer and the plurality of line segments, wherein the N+1th layer is a next layer of the Nth layer; and The mismatch degree is determined based on a sum of distances associated with the Mth layer and a number of the plurality of contour points representing the object.

5. The device according to claim 1, wherein: The processor is configured to: determining a sum of distances associated with an N-th layer based on a sum of a set of minimum values, wherein each of the set of minimum values ​​is a minimum value of distances between one of the contour points included in the N-th layer and the plurality of line segments, wherein N is a natural number satisfying 1≦N≦M, and the N-th layer is one of the first layer to the M-th layer included in the object frame, and wherein M is a natural number greater than or equal to 2; and The mismatch degree is determined based on a value obtained by adding distances associated with the first layer to the M-th layer and the number of the plurality of contour points representing the object.

6. The device according to claim 1, wherein: The processor is configured to determine a distance between one of the plurality of contour points and one of the line segments based on coordinates of the one contour point and coordinates of two points passing through the one of the plurality of line segments.

7. The device according to claim 1, wherein: The processor is configured to determine the mismatch degree based on a value obtained by dividing a sum of the plurality of minimum values ​​by the number of the plurality of contour points representing the object.

8. The device according to claim 1, wherein: The object frame includes more than three line segments.

9. The device according to claim 1, wherein: The processor is configured to: Based on determining that the mismatch is greater than a mismatch threshold, assigning a confidence value indicating whether the object is a stationary object; and It is determined whether the object is a stationary object based on the confidence value.

10. The device according to claim 1, wherein: The processor is configured to: determining a first confidence value indicating whether the object is a stationary object based on determining the first mismatch degree as the mismatch degree; determining a second confidence value greater than the first confidence value based on determining a second mismatch degree greater than the first mismatch degree as the mismatch degree; as well as Whether the object is a stationary object is determined based on the first confidence value or the second confidence value.

11. The device according to claim 1, wherein: The processor is configured to: Identify the object frame as a specified graphic; and It is determined whether the object is an object corresponding to the designated graphic based on the mismatch degree.

12. The device according to claim 1, wherein: The processor is configured to assign an identifier to the object to indicate that the object is a stationary object based on determining that the object is a stationary object.

13. The apparatus according to claim 1, wherein: The processor is configured to: determining a confidence value indicating whether the object is a stationary object based on the mismatch degree; and Whether the object is a stationary object is determined based on a value obtained by multiplying the reliability value by a weight.

14. A method comprising the steps of: identifying an object box including a plurality of contour points representing the object based on the sensed information of the sensor; Determine a plurality of minimum values, wherein each of the plurality of minimum values ​​is a minimum value of distances between a contour point among the plurality of contour points and a plurality of line segments constituting a boundary of the object frame; determining a mismatch between the object frame and the plurality of contour points based on at least one of the plurality of minimum values ​​and the number of the plurality of contour points: and A signal indicating whether the object is a stationary object is output based on the mismatch degree.

15. The method according to claim 14, wherein: The step of determining the mismatch degree comprises: determining a sum of distances between an Nth contour point and the plurality of line segments based on a minimum value of distances between the Nth contour point and the plurality of line segments, wherein N is a natural number satisfying 1≦N≦M, and the Nth contour point is one of the first to Mth contour points representing the object, and wherein M is a natural number greater than or equal to 2; determining a second sum of distances between an N+1th contour point and the plurality of line segments based on a minimum value of the sum of the distances and the distances between an N+1th contour point and the plurality of line segments, wherein the N+1th contour point is next to the Nth contour point; and The mismatch degree is determined based on a sum of distances between the Mth contour point and the plurality of line segments.

16. The method according to claim 14, wherein: The step of determining the mismatch degree comprises: Determining a minimum value of distances between an Nth contour point and the plurality of line segments, wherein N is a natural number satisfying 1≦N≦M, and the Nth contour point is one of the first to Mth contour points representing the object, wherein M is a natural number greater than or equal to 2; and The mismatch degree is determined based on the sum of the multiple minimum values, wherein each of the multiple minimum values ​​is a minimum value of distances between each contour point from the first contour point to the Mth contour point of the multiple contour points and the multiple line segments.

17. The method according to claim 14, wherein: The step of determining the mismatch degree comprises: Determining a sum of distances associated with an N-th layer based on a sum of a set of minimum values, wherein each of the set of minimum values ​​is a minimum value of distances between one of the contour points included in the N-th layer and the plurality of line segments, wherein N is a natural number satisfying 1≦N≦M, and the N-th layer is one of the first layer to the M-th layer included in the object frame, and wherein M is a natural number greater than or equal to 2; determining a sum of distances associated with an N+1th layer based on a sum of distances associated with the Nth layer and a sum of a second set of minimum values, wherein each of the second set of minimum values ​​is a minimum value of distances between one of the contour points included in the N+1th layer and the plurality of line segments, wherein the N+1th layer is a next layer to the Nth layer; and The mismatch degree is determined based on a sum of distances associated with the Mth layer and a number of the plurality of contour points representing the object.

18. The method according to claim 14, wherein: The step of determining the mismatch degree comprises: determining a sum of distances associated with an N-th layer based on a sum of a set of minimum values, wherein each of the set of minimum values ​​is a minimum value of distances between one of the contour points included in the N-th layer and the plurality of line segments, wherein N is a natural number satisfying 1≦N≦M, and the N-th layer is one of the first layer to the M-th layer included in the object frame, wherein M is a natural number greater than or equal to 2; and The mismatch degree is determined based on a value obtained by adding distances associated with the first layer to the M-th layer and the number of the plurality of contour points representing the object.

19. The method according to claim 14, wherein: The step of determining the plurality of minimum values ​​includes determining a distance between one of the plurality of contour points and one of the line segments based on coordinates of the one contour point and coordinates of two points passing through the one of the plurality of line segments.

20. The method according to claim 14, wherein: The step of determining the mismatching degree includes determining the mismatching degree based on a value obtained by dividing a sum of the plurality of minimum values ​​by the number of the plurality of contour points representing the object.

Images