DEVICE FOR OBJECT RECOGNITION AND METHOD THEREFOR

Abstract

The present disclosure relates to an apparatus and a method for object recognition and the object recognition apparatus. The apparatus may comprise a sensor and a processor, wherein the processor is configured to identify, based on detection information from the sensor, an object box comprising a plurality of contour points representing an object and determine a plurality of minimum values, each of the plurality of minimum values ​​being a minimum value among the distances between one of the plurality of contour points and line segments forming a boundary of the object box, determine a degree of mismatch between the object box and the plurality of contour points based on at least one of the plurality of minimum values ​​or a number of the plurality of contour points, and output a signal indicating whether the object is a stationary object based on the degree of mismatch.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to an apparatus and method for object recognition, and more particularly to a technique for identifying features of an object from the distribution of contour points obtained by a sensor (e.g., a LIDAR sensor). BACKGROUND

[0002] An autonomous vehicle or a vehicle with activated driver assistance systems can sense the environment using sensors (e.g. LiDAR) and collect data for driving.

[0003] A vehicle can receive data from a LIDAR indicating the position of an object in the vehicle's surroundings. A distance from a LIDAR to an object can be determined by the interval between the time a laser is transmitted by the LIDAR and the time the laser reflected from the object is received. A vehicle can identify the location of a point outside the object in a space where the vehicle is located based on the angle of the transmitted laser and the distance to the object.

[0004] Based on the movement information of the detected points, the autonomous vehicle or the vehicle with an enabled driver assistance device can identify the information of an object represented by the points. Specifically, the technology can be used to detect whether an object is a moving object, an object that can move, or an object that cannot move, ensuring the stability of autonomous driving or driver assistance and reducing the risk of accidents. OVERVIEW

[0005] According to the present disclosure, an apparatus may comprise a sensor and a processor, wherein the processor is configured to identify, based on detection information from the sensor, an object box comprising a plurality of contour points representing an object, determine a plurality of minimum values, each of the plurality of minimum values ​​being a minimum value among the distances between one of the plurality of contour points and line segments forming a boundary of the object box, determine a degree of mismatch between the object box and the plurality of contour points based on at least one of the plurality of minimum values ​​or a number of the plurality of contour points, and output a signal indicating whether the object is a stationary object based on the degree of mismatch.

[0006] The device, wherein the processor is configured to determine a sum of distances between an N-th contour point and the line segments based on a minimum value among the distances between the N-th contour point and the line segments, where the N-th contour point is among a first contour point to the M-th contour point representing the object, and where N is a natural number satisfying 1 ≦ N ≦ M, and M is a natural number of 2 or greater, determine a second sum of distances between an (N+1)-th contour point and the line segments based on the sum of distances and a minimum value among the distances between the (N+1)-th contour point and the line segments, where the (N+1)-th contour point is adjacent to the N-th contour point, and determine the degree of mismatch based on a sum of distances between an M-th contour point and the line segments.

[0007] The apparatus, wherein the processor is configured to determine a minimum value among the distances between an N-th contour point and the line segments, where N is a natural number satisfying 1 ≦ N ≦ M, and the N-th contour points are among a first contour point to an M-th contour point representing the object, where M is a natural number of 2 or greater, and to determine the degree of mismatch based on a sum of the plurality of minimum values, each of the plurality of minimum values ​​being a minimum value among the distances between a respective contour point between the first contour point and the M-th contour point of the plurality of contour points and the line segments.

[0008] The device, wherein the processor is configured to determine a sum of distances associated with an Nth layer based on a sum of minimum values ​​among distances between contour points contained in the Nth layer and the line segments, where N is a natural number satisfying 1 ≦ N ≦ M, and the Nth layer belongs to the first to Mth layers contained in the object box, and where M is a natural number greater than or equal to 2, to determine a sum of distances associated with an (N+1)th layer based on the sum of distances associated with the Nth layer and a sum of minimum values ​​among distances between contour points contained in the (N+1)th layer and the line segments, where the (N+1)th layer is closest to the Nth layer, and to determine the degree of mismatch based on a sum of distances,assigned to the M-th layer, and a number of the plurality of contour points representing the object.

[0009] The apparatus, wherein the processor is configured to determine a sum of distances associated with an N-th layer based on a sum of a set of minimum values, each of the set of minimum values ​​being a minimum value among the distances between one of the contour points included in the N-th layer and the line segments, where N is a natural number satisfying 1 ≦ N ≦ M, and the N-th layer belongs to a first layer to an M-th layer included in the object box, and where M is a natural number greater than or equal to 2, and to determine the degree of mismatch based on a value obtained by summing distances associated with the first layer to the M-th layer and a number of the plurality of contour points representing the object.

[0010] The apparatus wherein the processor is configured to determine a distance between a contour point of the plurality of contour points and a line segment of the line segments based on the coordinates of the one contour point and the coordinates of two points passing through the one line segment.

[0011] The apparatus wherein the processor is configured to determine the degree of mismatch based on a value obtained by dividing a sum of the plurality of minimum values ​​by a number of contour points representing the object.

[0012] The device, wherein the object box may comprise three or more line segments. The device, wherein the processor is configured to assign a confidence value to indicate whether the object is a stationary object based on determining that the degree of mismatch is greater than a threshold degree of mismatch, and to determine whether the object is the stationary object based on the confidence value.

[0013] The apparatus, wherein the processor is configured to determine a first reliability value to indicate whether the object is a stationary object based on determining a first degree of disagreement as the degree of disagreement, determine a second reliability value that is greater than the first reliability value based on determining a second degree of disagreement that is greater than the first degree of disagreement as the degree of disagreement, and determine whether the object is a stationary object based on the first reliability value or the second reliability value.

[0014] The apparatus, wherein the processor is configured to identify the object box as a particular shape and to determine whether the object is an object that conforms to the particular shape based on the degree of mismatch.

[0015] The apparatus wherein the processor is configured to assign to the object an identifier indicating that the object is a stationary object based on a determination that the object is a stationary object.

[0016] The apparatus wherein the processor is configured to determine a reliability value indicating whether the object is a stationary object based on the degree of disagreement and to determine whether the object is stationary based on a value obtained by multiplying the reliability value by a weight.

[0017] According to the present disclosure, a method may include the steps of: identifying an object box comprising a plurality of contour points representing an object based on detection information of the sensor; determining a plurality of minimum values, each of the plurality of minimum values ​​being a minimum value among the distances between one of the plurality of contour points and line segments forming a boundary of the object box; determining a degree of mismatch between the object box and the plurality of contour points based on at least one of: the plurality of minimum values ​​or a number of the plurality of contour points; and outputting a signal indicating whether the object is a stationary object based on the degree of mismatch.

[0018] The method, wherein determining the degree of mismatch may comprise determining a sum of distances between an N-th contour point and the line segments based on a minimum value among the distances between the N-th contour point and the line segments, where N is a natural number satisfying 1 ≦ N ≦ M, and the N-th contour point is among a first contour point to the M-th contour point representing the object, and where M is a natural number of 2 or greater; determining a second sum of distances between an (N+1)-th contour point and the line segments based on the sum of distances and a minimum value among the distances between the (N+1)-th contour point and the line segments, where the (N+1)-th contour point is closest to the N-th contour point; and determining, based on a sum of distances between an M-th contour point and the line segments, the degree of mismatch.

[0019] The method, wherein determining the degree of mismatch may comprise: determining a minimum value among the distances between an N-th contour point and the line segments, where N is a natural number satisfying 1 ≦ N ≦ M, and the N-th contour point is among a first contour point to an M-th contour point representing the object, where M is a natural number of 2 or greater; and determining the degree of mismatch based on a sum of the plurality of minimum values, wherein each of the plurality of minimum values ​​is a minimum value among the distances between a respective contour point between the first contour point and the M-th contour point of the plurality of contour points and the line segments.

[0020] The method, wherein determining the degree of mismatch may comprise: determining a sum of distances associated with an N-th layer based on a sum of a set of minimum values, each of the set of minimum values ​​being a minimum value among the distances between one of the contour points included in the N-th layer and the line segments, where N is a natural number satisfying 1 ≦ N ≦ M, and the N-th layer is among a first layer to an M-th layer included in the object box, and where M is a natural number greater than or equal to 2;Determining a sum of distances associated with an (N+1)th layer based on the sum of distances associated with the Nth layer and a sum of a second set of minimum values, each of the second set of minimum values ​​being a minimum value among the distances between one of the contour points included in the (N+1)th layer and the line segments, the (N+1)th layer being closest to the Nth layer; and determining the degree of mismatch based on a sum of distances associated with the Mth layer and a number of the plurality of contour points representing the object;

[0021] The method, wherein determining the degree of mismatch may comprise: determining a sum of distances associated with an N-th layer based on a sum of a set of minimum values, each of the set of minimum values ​​being a minimum value among the distances between one of the contour points included in the N-th layer and the line segments, where N is a natural number satisfying 1 ≦ N ≦ M, and the N-th layer is among a first layer to an M-th layer included in the object box, where M is a natural number greater than or equal to 2; and determining the degree of mismatch based on a value obtained by summing distances associated with the first layer to the M-th layer and a number of the plurality of contour points representing the object.

[0022] The method wherein determining the plurality of minimum values ​​may comprise determining a distance between a contour point of the plurality of contour points and a line segment of the line segments based on the coordinates of the one contour point and the coordinates of two points passing through the one line segment.

[0023] The method wherein determining the degree of mismatch may comprise determining the degree of mismatch based on a value obtained by dividing a sum of the plurality of minimum values ​​by a number of contour points representing the object. BRIEF DESCRIPTION OF THE DRAWING FIGURES

[0024] The above and other objects, features and advantages of the present disclosure will become more apparent from the following detailed description taken in conjunction with the accompanying drawing figures: Fig. 1 shows an example of an object recognition device according to an example of the present disclosure; Fig. 2 shows an example of information required for classifying objects in an object recognition device or method according to an example of the present disclosure; Fig. 3 shows an example of a range in which the weights of the reliability values ​​vary according to the information related to an object recognition device or an object recognition method according to an example of the present disclosure; Fig. 4 shows an example of a flowchart for the operation of an object recognition device for identifying whether an object is an object that cannot be in a moving state by identifying a contour box distance in an object recognition device or an object recognition method according to an example of the present disclosure; Fig. 5 shows an example of the distribution of contour points in an object recognition device or an object recognition method according to an example of the present disclosure; Fig. 6 shows an example of the distribution of contour points included in a single layer for identifying a matching degree in an object recognition apparatus or method according to an example of the present disclosure; Fig. 7 shows an example of a contour box distance corresponding to a contour point in an object recognition device or an object recognition method according to an example of the present disclosure; Fig. 8 shows an example of a flowchart for the operation of an object recognition device for identifying whether an object is an object incapable of being in a moving state according to a degree of disagreement in an object recognition device or an object recognition method according to an example of the present disclosure; Fig. 9 shows an example of the distribution of contour points representing an object that cannot be in a moving state in an object recognition device or an object recognition method according to an example of the present disclosure; Fig. 10 shows an example of the distribution of contour points representing an object that cannot be in a moving state in an object recognition device or an object recognition method according to an example of the present disclosure; Fig. 11 shows an example of the distribution of contour points representing an object that cannot be in a moving state in an object recognition device or an object recognition method according to an example of the present disclosure; Fig. 12 shows an example of the distribution of contour points representing an object that cannot be in a moving state in an object recognition device or an object recognition method according to an example of the present disclosure; and Fig. 13 shows an example of a computer system in communication with an object recognition device or method according to an example of the present disclosure. DETAILED DESCRIPTION

[0025] Below, some examples of the present disclosure will be described in detail with reference to the exemplary drawing figures. When adding reference numerals to the components in each drawing figure, it should be noted that the same or equivalent component will be designated by the same numeral even if they are shown in different drawing figures. Furthermore, in describing the example of the present disclosure, a detailed description of well-known features or functions will be omitted so as not to unnecessarily obscure the essence of the present disclosure.

[0026] In describing the components of the example according to the present disclosure, terms such as first, second, "A," "B," (a), (b), and the like may be used. These terms are used solely to distinguish one component from another, and the terms do not limit the nature, order, or arrangement of the individual components. Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure relates.Such terms, as defined in a commonly used dictionary, are to be construed as having a meaning consistent with contextual meanings in the relevant field of art and not as having an ideal or overly formal meaning, unless they are clearly defined as such in the present application.

[0027] Furthermore, the terms “unit”, “device”, “element”, “body” or the like used below may refer to at least one shape structure or a unit for performing a function.

[0028] Furthermore, in the examples of the present disclosure, the terms "greater than" or "less than" may be used to indicate whether a certain condition is met or not, but they are only for illustrative purposes and do not exclude "greater than or equal to" or "less than or equal to." A condition indicating "greater than or equal to" may be replaced with "greater than," a condition indicating "less than or equal to" may be replaced with "less than," and a condition indicating "greater than or equal to and less than" may be replaced with "greater than and less than or equal to." Furthermore, "A" to "B" means at least one of A (including A) to B (including B).

[0029] Hereinafter, examples of the present disclosure are described with reference to the Fig. 1 to 13 described in detail.

[0030] Fig. 1 shows an example of an object recognition device according to an example of the present disclosure.

[0031] Referring to Fig. 1, an object detection device 101 according to an example of the present disclosure may be implemented in a vehicle. In this case, the object detection device 101 may be formed integrally with internal control units of the vehicle or implemented as a separate device and connected to the control units of the vehicle by separate connection means.

[0032] As in Fig. 1, the object detection device 101 may include a sensor (e.g., a LIDAR 103) and a processor 105.

[0033] According to one example, the processor 105 of the object recognition device 101 may acquire a point cloud representing the object by the LIDAR 103. The processor 105 of the object recognition device 101 may identify contour points among the points included in the point cloud.

[0034] For example, contour points can be acquired using a LiDAR. For example, contour points can be identified in each of the layers formed on the z-axis between the x-axis, the y-axis, and the z-axis. For example, contour points can be identified from representative points in a point cloud in each of the layers formed on the z-axis between the x-axis, the y-axis, and the z-axis. For example, representative points can include all or part of the points located outside of a plurality of points in the point cloud. For example, a point cloud can be obtained by clustering based on identifying each individual point from a plurality of points acquired by a LiDAR within a certain distance.

[0035] For example, an object box may be a virtual box to which information about an external object (e.g., a vehicle, a person, a tree, a tire, a road sign, etc.) is assigned. The object box may, for example, be called a contour box. According to one example, the contour box may be created (or formed) using contour points. The contour box may, for example, correspond to an external object. The contour box may, for example, contain the virtual box to which the information about the external object is assigned. The information about the external object may, for example, include the type of the external object, the speed of the external object, the direction of movement of the external object, or the position of the external object, or any combination thereof.

[0036] According to one example, the processor 105 of the object detection device 101 may identify an object box containing contour points obtained by the LIDAR 103 that represent an object. The processor 105 of the object detection device 101 may determine a degree of mismatch indicating the degree of mismatch between the distribution of the contour points and the object box.

[0037] According to one example, the processor 105 of the object recognition device 101 may identify a contour box distance, which is the minimum value among the distances between the contour point and the line segments forming the object box, to identify the degree of mismatch. The operation of the object recognition device in which the processor of the object recognition device identifies the degree of mismatch based on the contour box distances will be described with reference to the following Fig. 4 to 8 described.

[0038] According to one example, the processor 105 of the object recognition device 101 may identify a reliability value according to the box matching information based on the degree of mismatch. The operation of the object recognition device that identifies a reliability value according to the box matching information based on the degree of mismatch is described below with reference to Fig. 4 described.

[0039] According to one example, the processor 105 of the object recognition device 101 may calculate a score representing a probability that an object is an object that cannot be in a moving state (e.g., a bridge pier, a guardrail) based on a confidence value according to the object's box matching information. Identifying a score indicating the probability that an object is an object that cannot be in a moving state (e.g., a stationary object, an immovable object, a fixed object, an immobile object, a stuck object, a motionless object, a non-moving object, a static object, an object anchored in place, or a rigid object, etc.)), based on the reliability value according to the box-matching information, is described below with reference to . Fig. 2 described.

[0040] According to one example, the processor 105 of the object recognition device 101 may identify that an object is an object that cannot be in a moving state based on a score indicating the probability that an object is a moving object (e.g., a moving vehicle) or an object that can be in a moving state (e.g., a stationary vehicle) that is less than a score indicating the probability that an object is an object that cannot be in a moving state (e.g., a road sign, a guardrail, a tree, etc.).

[0041] According to one example, the processor 105 of the object recognition device 101 may assign an identifier to the object indicating that the object is an object that cannot be in a movable state based on the identification that the object is an object that cannot be in a movable state. The identifier may be referred to as a flag, but is not limited to this.

[0042] Fig. 2 shows an example of information required for classifying objects in an object recognition device or method according to an example of the present disclosure.

[0043] Referring to Fig. 2, the table 201 may represent types of information for calculating a value to determine whether an object is an object incapable of being in a moving state. An immobility score 203 may represent a value for determining whether an object is an object incapable of moving (e.g., a stationary object, an immobile object, a fixed object, a non-moving object, a stuck object, an immobile object, a static object, an object anchored in place, or a rigid object, etc.). The immobility score 203 may be determined from information such as the tracking information 211, the box size information 213, and the box conformity information 215.A mobility value 205 may represent a value that can be used to determine whether an object is a moving object or an object that may be in a moving state. The mobility value 205 may be determined based on information such as in-lane information 217, tracking information 219, other information about other objects in the lane ("in-lane object information") 221, speed information 223, contour point distribution information 225, and boundary object information 227.

[0044] According to one example, a first reliability may be the reliability for determining whether an object is an object incapable of being in a moving state. The processor of the object recognition device may determine the immobility value 203 by the sum of the values ​​obtained by multiplying the first reliabilities indicated by the information by a weight. A second reliability may be the reliability for determining whether an object is a moving object or an object capable of being in a moving state. The processor of the object recognition device may determine the mobility value 205 by the sum of the values ​​resulting from multiplying the second reliabilities indicated by the information by a weight.

[0045] According to one example, the information 211 for identifying the immobility value 203 may represent a first reliability based on whether an object outside a lane is identified. The information about the box size 213 for identifying the immobility value 203 may represent a first reliability based on whether the size of an object box is greater than or equal to a reference size. The information 215 about the match of the boxes for identifying the immobility value 203 may represent a first reliability assigned based on the distribution of the contour points and the degree of match of the object box.

[0046] According to one example, lane information 217 may represent a second reliability based on whether an object is identified within a lane. Tracking information 219 may represent a second reliability based on whether an object is moving. Speed ​​information 223 may represent a second reliability assigned based on the speed of an object. Edge object information 227 may represent a second reliability based on whether an object is seen at the edge of a field of view without being occluded.

[0047] According to one example, the immobility value 203 can be determined by the sum of the values ​​obtained by multiplying the first reliability represented by information by a weight. For example, the immobility value 203 can be identified by the sum of a value obtained by multiplying the first reliability according to the lane keeping information 211 by a weight (e.g., weight S1 ) according to the tracking accuracy information 211, a value obtained by multiplying the first reliability according to the box size information 213 by a weight (e.g., (e.g., weight S2 ) according to the box size information 213, a value obtained by multiplying the first reliability according to the box matching information 215 by a weight (e.g., weight S3) obtained according to the box matching information 215, or at least any combination thereof. However, the examples within the scope of the present disclosure are not limited thereto. According to one example, the immobility value 203 may be determined by adding not only a value obtained by multiplying information listed in the table 201 by a weight, but also a value obtained by multiplying information not listed in the table 201 by the weight.

[0048] According to one example, the mobility value 205 may be determined by the sum of the values ​​obtained by multiplying the second reliability represented by the information by a weight. For example, the mobility rating 205 may be identified by the sum of at least one value obtained by multiplying the second reliability according to the lane information 217 by a weight (e.g., weight D1 ) according to the lane information 217, a value obtained by multiplying the second reliability according to the tracking information 219 by a weight (e.g. weight D2 ) according to the tracking information 219, a value obtained by multiplying the second reliability according to the information 221 about another object in the lane by a weight (e.g., weight D3) according to the information 221 about another object in the lane, a value obtained by multiplying the second reliability according to the speed information 223 by a weight (e.g. weight D4 ) according to the speed information 223, a value obtained by multiplying the second reliability according to the contour point distribution information 225 by a weight (e.g., weight D5 ) corresponding to the contour point distribution information 225, a value obtained by multiplying the second reliability according to the boundary object information 227 by a weight (e.g., weight D6) corresponding to the boundary object information 227, or any combination thereof. However, the examples within the scope of the present disclosure are not limited thereto. According to one example, the mobility score 205 may be determined by adding not only a value obtained by multiplying information listed in the table 201 by a weight, but also a value obtained by multiplying information not listed in the table 201 by the weight.

[0049] For example, if the mobility value (mobility score) 205 for a particular object is higher than the immobility value (immobility score) 203 for the particular object, the processor of the object recognition device may determine that the particular object is a moving object or an object capable of being in a moving state. For example, if the immobility value 203 for a particular object is higher than the mobility value 205 for the particular object, the processor of the object recognition device may determine that the particular object is an object (e.g., a street sign, a wall, a power line, etc.) that is incapable of being in motion.

[0050] According to one example of the present disclosure, the processor of the object recognition device may identify a first reliability indicated by the box matching information 215. A method for identifying the first reliability represented by the box matching information 215 is described below with reference to Fig. 4 to 8. Hereinafter, a value of the first reliability represented by the box matching information 215 may be referred to as a reliability value.

[0051] Fig. 3 shows an example of a range in which the weights of the reliability values ​​vary according to the information related to an object recognition device or an object recognition method according to an example of the present disclosure.

[0052] Referring to Fig. 3, a frame 301 may represent a first region 305, a second region 307, and a third region 309 separated by a distance from a host vehicle 303 having the object detection device. The first region 305 may include a region within a field of view. The second region 307 (e.g., an area in front of the host vehicle 303 but not within the field of view) may include a region for classifying objects of interest. The third region 309 (e.g., an area in front of the host vehicle 303 and farther away from the second region 307) may include regions other than the first region 305 and the second region 307.

[0053] According to one example, the first region 305 may be referred to as a field of view (FoV), but is not limited thereto. The second region 307 may be referred to as a class region of interest (Class ROI), but need not be limited thereto. The third region 309 may be referred to as a standard region, but need not be limited thereto.

[0054] According to one example, the processor of the object recognition device may apply different weights (e.g., weights in Fig. 2) for identifying an immobility value or a mobility value depending on an area in which an object is located. The reason for this is that information of high importance can change depending on the position of an object. For example, the processor of the object recognition device can weight the contour point distribution information (e.g., the contour point distribution information 225 in Fig. 2) only in the second area 307 to a value greater than 0. For example, the processor of the object recognition device may weight the boundary object information (e.g., the boundary object information 227 in Fig. 2) in the first area 305 so that it is greater than a weighting of the boundary object information in the second area 307 and a weighting of the boundary object information in the third area 309.

[0055] Fig. 4 shows an example of a flowchart for the operation of an object recognition device for identifying whether an object is an object that cannot be in a moving state by identifying a contour box distance in an object recognition device or an object recognition method according to an example of the present disclosure.

[0056] In the following, it is assumed that the object recognition device 101 of Fig. 1 the process of Fig. 4. In addition, the description of Fig. 4, the operations described as being performed by the device are understood to be controlled by the processor 105 of the object recognition device 101.

[0057] Referring to Fig. 4, according to one example, in a first operation 401, the processor of the object recognition device may identify contour box distances among the distances between a contour point and line segments forming an object box.

[0058] To determine the contour box distances, the processor of the object detection device may identify an object box with contour points obtained by a sensor (e.g., LIDAR sensor) that represent an object.

[0059] The processor of the object recognition device can determine a contour box distance that is the minimum value among the distances between one of the contour points and the line segments forming the object box. A method for identifying contour box distances is described below with reference to Fig. 6 to 7.

[0060] In a second operation 403, according to one example, the processor of the object recognition device may determine whether there is a contour point in a layer for which a contour-box distance has not been identified. If there is a contour point in the layer for which the contour-box distance has not been identified, the processor of the object recognition device may perform a first operation 401. If there is no contour point in the layer for which the contour-box distance has not been identified, the processor of the object recognition device may perform a third operation 405.

[0061] According to one example, the processor of the object recognition device may determine a contour box distance for contour points in one layer of a plurality of layers each containing contour points.

[0062] In the third operation 405, according to one example, the processor of the object recognition device may determine whether there is a layer with a contour point for which the contour-box distance is not identified. If there is a layer with a contour point for which the contour-box distance is not identified, the processor of the object recognition device may perform the first operation 401. If there is no layer with a contour point for which the contour-box distance is not identified, the processor of the object recognition device may perform a fourth operation 407.

[0063] According to one example, the processor of the object recognition device may identify the contour box distance for contour points contained in a different layer than the layer in which the contour box distance is identified.

[0064] In the fourth operation 407, according to one example, the processor of the object recognition device may identify a degree of mismatch. The degree of mismatch may indicate the degree of mismatch between an object box and contour points based on at least one of the contour box distances of the contour points or the number of contour points representing an object, or any combination thereof.

[0065] In one example, the degree of mismatch can be determined based on the contour box distances of all contour points representing the object.

[0066] For example, the degree of mismatch determined in a first method can be determined based on the sum of the contour box distances from a first contour point to an Mth contour point (where M is a natural number of 2 or more). A particular layer can be composed of the first to Mth contour points.

[0067] The processor of the object recognition device may determine the degree of mismatch based on a value obtained by dividing the sum of the contour box distances of the contour points by the number of contour points representing the object.

[0068] For example, the degree of mismatch determined in a second method may be determined from the sum of the distances of the last contour point among all contour points representing the object. In other words, the degree of mismatch determined in the second method may be determined from 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 contour box distance of the (N+1)th contour point to the sum of the distances of the Nth contour point.

[0069] The processor of the object recognition device may determine the degree of mismatch based on a value of which is obtained when the sum of the distances of the Mth contour point is divided by the number of contour points representing the object.

[0070] In one example, the degree of mismatch can be determined based on the contour box distances of contour points in a plurality of layers.

[0071] For example, the degree of mismatch determined in a third method can be determined based on the sum of the distances from the first layer to the Mth layer. The sum of the distances of a given layer can represent the sum of the contour box distances of the contour points contained in the given layer.

[0072] The processor of the object recognition device can determine the degree of mismatch based on a value of which is obtained when the sum of the distances of the first layer to the Mth layer is divided by the number of contour points representing the object.

[0073] For example, the degree of mismatch determined in a fourth method can be determined from the sum of the distances of the Mth layer.

[0074] The processor of the object recognition device may identify the sum of the distances of the Nth layer from the sum of the contour box distances of the contour points included in an Nth layer (N is a natural number satisfying 1 ≦ N ≦ M) among the first to Mth layers (where M is a natural number greater than or equal to 2) included in an object box. The processor of the object recognition device may determine the sum of the distances of the (N+1)th layer from the sum of the distances of the Nth layer and the sum of the contour box distances of the contour points included in the (N+1)th layer, which is the next layer after the Nth layer.

[0075] In a fifth operation 409, according to one example, the processor of the object recognition device may determine whether the object is an object that cannot be in a moving state based on the degree of mismatch.

[0076] According to one example, the processor of the object recognition device may assign a certain value (e.g., 1) as a reliability indicating whether an object is an object that cannot be in a moving state based on determining that the degree of mismatch is greater than a certain threshold for the degree of mismatch. The processor of the object recognition device may determine whether the object is an object that cannot be in a moving state based on the reliability.

[0077] According to one example, the processor of the object recognition device may assign a larger reliability value to an object as the degree of mismatch increases. In other words, the processor of the object recognition device may identify a first reliability value as the reliability to indicate whether an object is an object incapable of being in a moving state based on identifying a first mismatch value as the degree of mismatch, and identify a second reliability value greater than the first reliability value as the reliability based on identifying a second mismatch value greater than the first mismatch value as the degree of mismatch.The processor of the object recognition device may identify whether the object is an object that cannot be in a moving state based on the first reliability value or the second reliability value.

[0078] Fig. 5 shows an example of the distribution of contour points in an object recognition apparatus or an object recognition method according to an example of the present disclosure.

[0079] Referring to Fig. 5, in a first situation 501, the processor of an object recognition device may receive a collection of points (e.g., a first point cloud 505) representing a first object 503. The processor of the object recognition device may identify a first contour line 507 based on the first point cloud 505. The first contour line 507 may comprise a line connecting contour points identified from the first point cloud 505.

[0080] In a second situation 511, the processor of the object recognition device may receive a second group of points (e.g., a second point cloud 515) representing a second object 513. The processor of the object recognition device may identify a second contour line 517 from the second point cloud 515. The second contour line 517 may comprise a line connecting contour points identified from the second point cloud 515.

[0081] In one example, the processor of the object recognition device may determine the degree of mismatch of the first object 503 based on the distribution of the contour points included in the first contour line 507.

[0082] The processor of the object recognition device may determine a reliability value according to the box matching information (e.g., box matching information 215 in Fig. 2) of the first object 503 as less than a reference value based on the degree of mismatch of the first object 503. The processor of the object recognition device may identify the first object 503 as a moving object (e.g., a moving vehicle) or as an object that may be in a moving state (e.g., a parked vehicle) based on the reliability according to the box matching information of the first object 503.

[0083] In one example, the processor of the object recognition device may determine the degree of mismatch of the second object 513 based on the distribution of the contour points included in the second contour line 517. The degree of mismatch of the first object 503 may be lower than that of the second object 513.

[0084] The processor of the object recognition device may identify the reliability value according to the box matching information of the second object 513 as greater than or equal to the reference value based on the degree of non-conformity 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 unable to be in a moving state based on the reliability based on the box matching information of the second object 513.

[0085] In Fig. 5, the object box is depicted as rectangular, however, the examples of the present disclosure are not limited thereto. According to one example, the object box may consist of three or more line segments. For example, the object box may have the shape of a triangle or a rectangle. The object field may have the shape of a pentagon, a hexagon, or an octagon, for example.

[0086] Although in Fig. 5, the object box detects whether an object is an object (e.g., a tree, a street sign, a wall, a power line, etc.) that cannot be in a moving state, the examples of the present disclosure are not limited thereto. According to one example, the processor of the object recognition device may identify the object box as a particular shape (e.g., a triangle, a rectangle, a pentagon, a hexagon, an octagon, a circle, an oval, a trapezoid, etc.) and identify a degree of mismatch indicating the degree of mismatch between the object box and the contour points based on at least one of the contour box distances from contour points representing an object, the number of contour points representing the object, or any combination thereof. The processor of the object recognition device may determine whether an object conforms to a shape based on the degree of mismatch.

[0087] Fig. 6 shows an example of the distribution of contour points included in a single layer for identifying a matching degree in an object recognition apparatus or an object recognition method according to an example of the present disclosure.

[0088] As in Fig. 6, a first set 601 may include contour points contained in a first layer (e.g., layer 0). A second set 603 may include contour points contained in a second layer (e.g., layer 1). A third set 605 may include contour points contained in a third layer (e.g., layer 2). A fourth set 607 may include contour points contained in a fourth layer (e.g., layer 3).

[0089] In one example, the processor of the object recognition device may determine the degree of mismatch based on the sum of the distances of the individual layers.

[0090] For example, the degree of disagreement achieved by the third method (e.g., the third method described in the description of operation 407 of Fig. 4) can be identified by the sum of the distances from the first layer to the Mth layer. The sum of the distances of a given layer can be the sum of the contour box distances of the contour points contained in the given layer.

[0091] In other words, the sum of the distances of the first layer may include a value obtained by summing the contour box 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 summing the contour box distances of the contour points included in the second layer. The processor of the object recognition device may determine the degree of mismatch based on a value obtained by dividing a value obtained by adding the sums of the distances of the first layer to the Mth layer by the number of contour points representing the object.

[0092] For example, the degree of mismatch determined by the fourth method (e.g., the fourth method described in the description of operation 407 of Fig. 4) is identified by the sum of the distances of the M-th layer, which is the last layer.

[0093] In other words, the processor of the object recognition device may determine the sum of the distances of the Nth layer from the sum of the contour box distances of the contour points included in an Nth layer (N is a natural number satisfying 1 ≦ N ≦ M) among the first to Mth layers (where M is a natural number greater than or equal to 2) included in an object box. The processor of the object recognition device may determine the sum of the distances of the (N+1)th layer from the sum of the distances of the Nth layer and the sum of the contour box distances of the contour points included in the (N+1)th layer, which is the next layer after the Nth layer. The processor of the object recognition device may determine the degree of mismatch from a value of obtained by dividing the sum of the distances of the Mth layer by the number of contour points representing the object.

[0094] Fig. 7 shows an example of a contour box distance corresponding to a contour point in an object recognition device or an object recognition method according to an example of the present disclosure.

[0095] Referring to Fig. 7, in a first state 701, the processor of an object recognition device may identify the contour box distance of a first contour point. In a second state 703, the processor of the object recognition device may identify the contour box distance of a second contour point. In a third state 705, the processor of the object recognition device may identify the contour box distance of a third contour point. In a fourth state 707, the processor of the object recognition device may identify the contour box distance of a fourth contour point. In a fifth state 709, the processor of the object recognition device may identify the contour box distance of a fifth contour point. In a sixth state 711, the processor of the object recognition device may identify the contour box 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 can be contained in a certain layer.

[0096] For example, to determine a 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 of line segments (e.g., boundary lines) that form an object box, the processor of the object recognition device may identify a distance between the one contour point and the one line segment based on the coordinates of the contour point and the coordinates of two points that pass through the line segment. For example, if the coordinates of the one contour point are (y,x) and the coordinates of the two points that pass through the one line segment are (y 1 , x 1) and (y 2 , x 2 ), the distance (d) between the one contour point and the one line segment can be determined according to equation 1 below. d=|(x2−x1)y+(y1−y2)x+(y2x1−y1x2)|(x2−x1)2+(y1−y2)2

[0097] According to one example, in the first state 701, the processor of the object recognition device may identify the contour box distance, which is the minimum value among the distances between the first contour point and the line segments forming the object box. For example, the distances between the first contour point and the line segments forming the object box may be 0.16 m, 3.28 m, 1.36 m, and 0.06 m, respectively. Among these, the minimum value of 0.06 m may be identified as the contour box distance. Accordingly, the contour box distance of the first contour point may be 0.06 m.

[0098] In one example, in the second state 703, the distances between the second contour point and the line segments forming the object box may be 0.53 m, 3.32 m, 1.0 m, and 0.02 m, respectively. Among these, the minimum value of 0.02 m may be set as the contour box distance. Accordingly, the contour-to-box distance of the second contour point may be 0.02 m.

[0099] According to one example, in the third state 705, the distances between the third contour point and the line segments forming the object box may be 1.49 m, 3.32 m, 0.04 m, and 0.02 m, respectively. Among these, the minimum value of 0.02 m may be determined as the contour box distance. Accordingly, the contour-to-box distance of the third contour point may be 0.02 m.

[0100] According to one example, in the fourth state 707, the distances between the fourth contour point and the line segments forming the object box may be 1.53 m, 2.81 m, 0.0 m, and 0.53 m, respectively. Among these, the minimum value of 0.0 m may be set as the contour box distance. Accordingly, the contour-box distance of the fourth contour point may be 0.0 m.

[0101] In one example, in the fourth state 709, the distances between the fourth contour point and the line segments forming the object box may be 1.55 m, 0.16 m, 0.02 m, and 3.18 m, respectively. Among these, the minimum value of 0.02 m may be determined as the contour box distance. Accordingly, the contour-box distance of the fifth contour point may be 0.02 m.

[0102] According to one example, in the sixth state 711, the distances between the sixth contour point and the line segments forming the object box may be 1.52 m, 0.01 m, 0.0 m, and 3.33 m, respectively. Among these, the minimum value of 0.0 m may be set as the contour box distance. Accordingly, the contour-box distance of the sixth contour point may be 0.0 m.

[0103] Fig. 8 shows an example of a flowchart for the operation of an object recognition device for identifying whether an object is an object incapable of being in a moving state according to a degree of disagreement in an object recognition device or an object recognition method according to an example of the present disclosure.

[0104] In the following, it is assumed that the object recognition device 101 of Fig. 1 the process of Fig. 8. In addition, the description of Fig. 8, the operations described as being performed by the device are understood to be controlled by the processor 105 of the object recognition device 101.

[0105] Referring to Fig. 8, according to one example, the processor of an object recognition device may, in a first operation 801, identify an object box containing contour points obtained by a LIDAR and representing an object.

[0106] In a second operation 803, according to one example, the processor of the object recognition device may identify a contour box distance that is the minimum value among the distances between one of the contour points and the line segments that form the object box.

[0107] In a third operation 805, according to one example, the processor of the object recognition device may identify a degree of mismatch indicating the degree of mismatch between the object box and the contour points based on at least one of the contour box distances of the contour points or the number of contour points, or any combination thereof. The number of contour points may indicate the number of contour points representing an object.

[0108] In a fourth operation 807, according to one example, the processor of the object recognition device may determine whether the object is an object that cannot be in a moving state based on the degree of mismatch.

[0109] Fig. 9 shows an example of the distribution of contour points representing an object that cannot be in a moving state in an object recognition apparatus or an object recognition method according to an example of the present disclosure.

[0110] Referring to Fig. 9, in a first situation 901, an object recognition device included in a host vehicle 903 may identify a first object 905 (e.g., a vehicle) and a second object 907 (e.g., a tree).

[0111] The first object 905 may correspond to a preceding vehicle of the host vehicle 903. The degree of mismatch of the first object 905 may be less than or equal to a threshold degree of mismatch. Accordingly, the first object 905 may be identified as a moving object or an object that may be in a moving state.

[0112] The second object 907 may correspond to a bush in front of the host vehicle. The degree of mismatch of the second object 907 may be greater than the threshold for the degree of mismatch. Accordingly, the second object 907 may be identified as an object that cannot be in a moving state.

[0113] Fig. 10 shows an example of the distribution of contour points representing an object that cannot be in a moving state in an object recognition device or an object recognition method according to an example of the present disclosure.

[0114] Referring to Fig. 10, in a first situation 1001, an object detection device included in a host vehicle 1003 may identify a first object 1005 (e.g., a power line). In a second situation 1011, the object detection device in the host vehicle 1003 may identify a second object 1013 (e.g., a road sign). In a third situation 1021, the object detection device in the host vehicle 1003 may identify a third object 1023.

[0115] According to one example, the first object 1005 may correspond to a plant in a flowerbed located adjacent to a road on which the host vehicle 1003 is traveling. The degree of mismatch of the first object 1005 may be greater than a certain threshold for the degree of mismatch. Accordingly, the first object 1005 may be identified as an object that is unable to move.

[0116] According to one example, the second object 1013 may correspond to a bush located next to a road on which the host vehicle 1003 is traveling. The degree of mismatch of the second object 1013 may be greater than the specified threshold for the degree of mismatch. Accordingly, the second object 1013 may be identified as an object that is unable to move.

[0117] According to one example, the third object 1023 may correspond to a tree located adjacent to the road on which the host vehicle 1003 is traveling. The degree of mismatch of the third object 1023 may be greater than the specified threshold for the degree of mismatch. Accordingly, the third object 1023 (e.g., a tree) may be identified as an object that cannot be in a moving state.

[0118] Fig. 11 shows an example of the distribution of contour points representing an object that cannot be in a moving state in an object recognition device or an object recognition method according to an example of the present disclosure.

[0119] Referring to Fig. 11, in a first situation 1101, an object recognition device included in a host vehicle 1103 may, according to one example, identify a first object 1105 (e.g., a first pier of a first bridge). In a second situation 1111, the object recognition device in the host vehicle 1103 may identify a second object 1113 (e.g., a first pier of a second bridge) and a third object 1115 (e.g., a second pier of the second bridge). In the third situation 1121, the object recognition device in the host vehicle 1103 may identify a fourth object 1123 and a fifth object 1125.

[0120] By way of example, the first object 1105 in the first situation 1101 may correspond to a pier 1107 of a first bridge pier located adjacent to a road on which the host vehicle 1103 is traveling. The degree of mismatch of the first object 1105 may be greater than a certain threshold for the degree of mismatch. Accordingly, the first object 1105 may be identified as an object unable to move.

[0121] According to one example, in a second situation 1111, the second object 1113 and the third object 1115 may correspond to a pier 1117 of a second bridge pier located adjacent to the road on which the host vehicle 1103 is traveling. The degree of mismatch of the second object 1113 and the degree of mismatch of the third object 1115 may be greater than a certain threshold for the degree of mismatch. Accordingly, the second object 1113 and the third object 1115 may be identified as objects that cannot be in a moving state.

[0122] According to one example, in a third situation 1121, the fourth object 1123 and the fifth object 1125 may correspond to a pier 1127 of a third bridge pier located adjacent to the road on which the host vehicle 1103 is traveling. The degree of mismatch of the fourth object 1123 and the degree of mismatch of the fifth object 1125 may be greater than the specified threshold for the degree of mismatch. Accordingly, the fourth object 1123 and the fifth object 1125 may be identified as objects that cannot be in a traveling state.

[0123] Fig. 12 shows an example of the distribution of contour points representing an object that cannot move in an object recognition apparatus or an object recognition method according to an example of the present disclosure.

[0124] Referring to Fig. 12, in a first situation 1201, an object detection device included in a host vehicle 1203, according to one example, may identify a first object 1205 (e.g., a temporary guardrail). In a second situation 1211, the object detection device in the host vehicle 1203 may identify a second object 1213 (e.g., a first impact beam). In a third situation 1221, the object detection device in the host vehicle 1203 may identify a third object 1223 (a second impact beam).

[0125] According to one example, the first object 1205 in the first situation 1201 may correspond to a temporary guardrail 1207 located adjacent to a road on which the host vehicle 1203 is traveling. The degree of mismatch of the first object 1205 may be greater than a certain threshold for the degree of mismatch. Accordingly, the first object 1205 may be identified as an object unable to move.

[0126] According to one example, the second object 1213 in the second situation 1211 may correspond to an impact beam 1215 of a highway located adjacent to the road on which the host vehicle 1203 is traveling. The degree of mismatch of the second object 1213 may be greater than a certain threshold for the degree of mismatch. Accordingly, the second object 1213 may be identified as an object that is unable to move.

[0127] According to one example, the third object 1223 in the third situation 1221 may correspond to an impact beam 1225 of a highway located adjacent to the road on which the host vehicle 1203 is traveling. The degree of mismatch of the third object 1223 may be greater than the specified threshold for the degree of mismatch. Accordingly, the third object 1223 may be identified as an object incapable of being in a moving state.

[0128] Fig. 13 shows an example of a computer system in communication with an object recognition device or method according to an example of the present disclosure.

[0129] As in Fig. 13, a computer system 1300 may include at least a processor 1310, a memory 1330, a user interface input device 1340, a user interface output device 1350, a memory 1360, and a network interface 1370 interconnected via a bus 1320.

[0130] Processor 1310 may be a central processing unit (CPU) or a semiconductor device that processes instructions stored in memory 1330 and / or storage 1360. Memory 1330 and storage 1360 may include various types of volatile or non-volatile storage media. For example, memory 1330 may include a ROM (Read Only Memory) 1331 and a RAM (Random Access Memory) 1332.

[0131] Thus, the operations of the method or algorithm described in connection with the examples disclosed herein may be embodied directly in hardware or in a software module executed by processor 1310, or in a combination thereof. The software module may be located on a storage medium (i.e., memory 1330 and / or storage 1360) such as RAM, flash memory, ROM, EPROM, EEPROM, register, hard disk, removable disk, and CD-ROM.

[0132] The example storage medium may be connected to processor 1310, and processor 1310 may read information from the storage medium and record information to the storage medium. Alternatively, the storage medium may be integrated into processor 1310. The processor and storage medium may be housed in an application-specific integrated circuit (ASIC). The ASIC may be housed in a user terminal. Alternatively, the processor and storage medium may be housed as separate components in the user terminal.

[0133] The present disclosure has been made to solve the above-mentioned problems encountered in the prior art while maintaining the advantages of the prior art.

[0134] An example of the present disclosure provides an object detection apparatus and method for identifying whether an object is an object that cannot be in a moving state.

[0135] An example of the present disclosure provides an object recognition apparatus and method for identifying whether an object is an object that cannot be in a moving state according to the degree of mismatch identified from contour points and an object box.

[0136] An example of the present disclosure provides an object recognition apparatus and method for improving the accuracy of determining whether an object is an object that cannot be in a moving state.

[0137] The technical problems to be solved by the present disclosure are not limited to the above-mentioned problems, and any other technical problems not mentioned here will be clearly apparent from the following description to those skilled in the art to which the present disclosure refers.

[0138] According to an example of the present disclosure, an object detection device includes a LIDAR and a processor.

[0139] According to one example, the processor may identify an object box obtained by the LIDAR and containing contour points representing an object, identify a contour box distance that is a minimum value among the distances between any of the contour points and line segments forming the object box, identify a mismatch degree that indicates a degree of mismatch between the object box and the contour points based on at least one of the contour box distances of the contour points or a number of contour points representing the object, or any combination thereof, and identify whether the object is an object incapable of being in a moving state based on the mismatch degree.

[0140] According to one example, the processor may identify a sum of distances of an N-th contour point based on the contour box distance of the N-th contour point (N is a natural number satisfying 1 ≦ N ≦ M) among the first to M-th contour points (where M is a natural number of 2 or more) that represent the object, identify a sum of distances of an (N+1)-th contour point based on the sum of distances of the N-th contour point and a contour box distance of the (N+1)-th contour point that is a nearest contour point of the N-th contour point, and identify the degree of mismatch based on a sum of distances of an M-th contour point.

[0141] According to one example, the processor may identify the contour box distance of an 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 of 2 or more) representing the object, and identify the degree of mismatch based on a sum of the contour box distances of the first contour point to the Mth contour point.

[0142] According to one example, the processor may identify a sum of distances of an N-th layer based on a sum of contour box distances of the contour points included in the N-th layer (N is a natural number satisfying 1 ≦ N ≦ M) among the first to M-th layers (where M is a natural number greater than or equal to 2) included in the object box, identify a sum of distances of an (N+1)-th layer based on a sum of distances of the N-th layer and a sum of contour box distances of the contour points included in the (N+1)-th layer, which is the next layer after the N-th layer, and identify the degree of mismatch based on the sum of contour box distances of the first contour point to the M-th contour point.

[0143] According to one example, the processor may identify a sum of distances of an N-th layer from a sum of contour box distances of contour points included in the N-th layer (N is a natural number satisfying 1 ≦ N ≦ M) among the first to M-th layers (where M is a natural number greater than or equal to 2) included in the object box, and identify the degree of mismatch from a value obtained by summing the sums of the distances of the first layer to the M-th layer and a number of contour points representing the object.

[0144] In one example, the processor may determine a distance between a contour point and a line segment based on the coordinates of the one contour point and the coordinates of two points passing through the one line segment among the line segments that form the object box.

[0145] In one example, the processor may determine the degree of mismatch using a value resulting from dividing the sum of the contour box distances of the contour points by the number of contour points representing the object.

[0146] In one example, the processor may identify the object box consisting of three or more line segments.

[0147] According to one example, the processor may assign a specified value as reliability to indicate whether the object is an object incapable of being in a movable state based on determining that the degree of mismatch is greater than a specified threshold for the degree of mismatch, and determine whether the object is an object incapable of being in a movable state based on the reliability.

[0148] According to one example, the processor may identify a first reliability value as reliability to indicate whether the object is an object incapable of being in a movable state based on identifying a first degree of mismatch as the degree of mismatch, identify a second reliability value greater than the first reliability value as reliability based on identifying a second degree of mismatch greater than the first degree of mismatch as the degree of mismatch, and identify whether the object is an object incapable of being in a movable state based on the first reliability value or the second reliability value.

[0149] According to one example, the processor may identify the object box as a particular shape, identify a degree of mismatch indicating a degree of mismatch between the object box and the contour points based on at least one of the contour-box distances of the contour points or a number of the contour points representing the object, or any combination thereof, and identify whether the object is an object that conforms to the shape based on the degree of mismatch.

[0150] According to one example, the processor may assign an identifier to the object indicating that the object is an object that cannot move based on the identification that the object is an object that cannot move.

[0151] According to one example, the processor may identify a reliability indicating whether the object is an object incapable of being in a movable state based on the degree of mismatch, and identify whether the object is an object incapable of being in a movable state based on a value obtained by multiplying the reliability by a weight.

[0152] According to an example of the present disclosure, an object detection method includes identifying an object box obtained by a LIDAR and including contour points representing an object, identifying a contour box distance that is a minimum value among distances between any of the contour points and line segments constituting the object box, identifying a mismatch degree that indicates a degree of mismatch between the object box and the contour points based on at least one of the contour box distances of the contour points or a number of contour points representing the object, or any combination thereof, and identifying whether the object is an object incapable of being in a moving state based on the mismatch degree.

[0153] According to one example, identifying the degree of mismatch indicating the degree of mismatch between the object box and the contour points may be performed based on at least one of the contour box distances of the contour points or the number of contour points representing the object, or any combination thereof, may include identifying a sum of distances of an Nth contour point based on the contour box 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 of 2 or more) representing the object, identifying a sum of distances of an (N+1)th contour point based on the sum of distances of the Nth contour point and a contour box distance of the (N+1)th contour point that is a nearest contour point to the Nth contour point,and identifying the degree of mismatch based on a sum of distances of an Mth contour point.,

[0154] According to one example, identifying the degree of mismatch indicating the degree of mismatch between the object box and the contour points may be performed based on at least one of the contour box distances of the contour points or the number of contour points representing the object, or any combination thereof, may include identifying the contour box distance of an 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 of 2 or more) representing the object, and identifying the degree of mismatch based on a sum of the contour box distances of the first contour point to the Mth contour point.

[0155] According to one example, identifying the degree of mismatch, which indicates the degree of mismatch between the object box and the contour points, may be performed based on at least one of the contour box distances of the contour points or the number of contour points representing the object, or any combination thereof, may include identifying a sum of distances of an Nth layer based on a sum of contour box distances of contour points included in the Nth layer (N is a natural number satisfying 1 ≦ N ≦ M) among the first to Mth layers (where M is a natural number greater than or equal to 2) included in the object box, identifying a sum of distances of an (N+1)th layer based on a sum of distances of the Nth layer and a sum of the contour box distances of the contour points included in the (N+1)th layer, which is the next layer after the Nth layer,and identifying the degree of mismatch based on the sum of distances of the M-th layer and a number of contour points representing the object.,

[0156] According to one example, identifying the degree of mismatch indicating the degree of mismatch between the object box and the contour points may be performed based on at least one of the contour box distances of the contour points or the number of contour points representing the object, or any combination thereof, may include identifying a sum of distances of an N-th layer based on a sum of contour box distances of contour points included in the N-th layer (N is a natural number satisfying 1 ≦ N ≦ M) among first to M-th layers (where M is a natural number greater than or equal to 2) included in the object box, and identifying the degree of mismatch based on a value obtained by summing the sums of the distances of the first layer to the M-th layer and a number of contour points representing the object.

[0157] According to one example, identifying the contour box distance, which is the minimum value among the distances between one of the contour points and the line segments forming the object box, may include identifying a distance between a contour point and a line segment based on the coordinates of the one contour point and the coordinates of two points passing through the one line segment among the line segments forming the object box.

[0158] According to one example, identifying the degree of mismatch indicating the degree of mismatch between the object box and the contour points based on at least one of the contour box distances of the contour points or the number of contour points representing the object, or any combination thereof, may include identifying the degree of mismatch based on a value obtained by dividing a sum of the contour box distances of the contour points by a number of contour points representing the object.

[0159] The above description is merely illustrative of the technical idea of ​​the present disclosure, and various modifications and variations can be made by those skilled in the art of the present disclosure without departing from the essential features of the present disclosure.

[0160] Accordingly, the example disclosed in the present disclosure is not intended to limit the technical idea of ​​the present disclosure, but to describe the present disclosure, and the scope of the technical idea of ​​the present disclosure is not limited by the example. The scope of the present disclosure should be interpreted by the following claims, and all technical ideas within their corresponding scope should be considered to be included in the scope of the present disclosure.

[0161] Current technology can increase the accuracy of determining whether an object is an object incapable of being in a moving state based on the degree of mismatch.

[0162] In addition, the present technology can detect whether an object is a non-moving object based on the distribution of contour points by determining whether the distribution of contour points corresponds to an object box.

[0163] Furthermore, the present technology can improve the user experience by increasing the accuracy of determining whether an object is an object that cannot move.

[0164] Furthermore, the present technology can improve the performance of autonomous driving or driver assistance by increasing the accuracy of determining whether an object is an object that is unable to move.

[0165] In addition, various effects can be provided that are understood directly or indirectly through the disclosure.

[0166] Although the present disclosure has been described with reference to examples and the accompanying drawing figures, the present disclosure is not limited thereto, but can be modified and changed in various ways by those skilled in the art to which the present disclosure pertains without departing from the spirit and scope of the present disclosure as claimed in the following claims. REPRESENTATIVE FIGURE: FIG. 1

[0167] Fig. 2. 201 Table Fig. 3. 301 frames 303 Host vehicle 305 first area 307 second area 309 third area Fig. 5. 501 first situation 503 first object 505 first point cloud 507 first contour line 511 second situation 513 second object 515 second point cloud 517 second contour line Fig. 6. 601 first sentence 603 second movement 605 third movement 607 Fourth movement Fig. 7. 701 first state 703 second state 705 Third State 707 Fourth State 709 Fifth State 711 Sixth State Fig. 9. 901 first situation 903 host vehicle 905 first object 907 second object Fig. 10. 1001 first situations 1003 Host vehicle 1005 first object 1011 second situation 1013 second object 1021 Third Situation 1023 third object Fig. 11. 1101 first situation 1103 Host Vehicle 1105 first object 1107 Column 1111 second situation 1113 second object 1115 third object 1117 Column 1121 Third Situation 1123 fourth object 1125 fifth object 1127 Column Fig. 12. 1201 first situation 1203 Host vehicle 1205 first object 1207 temporary guardrail 1211 second situation 1213 second object 1215 impact beams 1221 Third Situation 1223 third object 1225 impact beams Fig. 13. 1300 computer system

Claims

[1] Device comprising: a sensor; and a processor, wherein the processor is configured to identify an object box comprising a plurality of contour points representing an object based on detection information from the sensor; determining a plurality of minimum values, each of the plurality of minimum values ​​being a minimum value among the distances between one of the plurality of contour points and line segments forming a boundary of the object box; to determine a degree of mismatch between the object box and the plurality of contour points based on at least one of the following: the majority of the minimum values ​​or a number of the plurality of contour points; and to output a signal based on the degree of mismatch indicating whether the object is a stationary object. [2] The apparatus of claim 1, wherein the processor is configured to determine a sum of distances between an N-th contour point and the line segments based on a minimum value among the distances between the N-th contour point and the line segments, the N-th contour point being located between a first contour point and an M-th contour point representing the object, and where N is a natural number satisfying 1 ≦ N ≦ M and M is a natural number of 2 or greater; to determine a second sum of distances between an (N+1)th contour point and the line segments based on the sum of distances and a minimum value among the distances between the (N+1)th contour point and the line segments, where the (N+1)th contour point is closest to the Nth contour point; and to determine the degree of mismatch based on a sum of distances between an Mth contour point and the line segments. [3] The apparatus of claim 1, wherein the processor is configured to: to determine a minimum value among the distances between an Nth contour point and the line segments, where N is a natural number satisfying 1 ≦ N ≦ M, and the Nth contour point is among a first contour point to the Mth contour point representing the object, where M is a natural number of 2 or greater; and to determine the degree of mismatch based on a sum of the plurality of minimum values, each of the plurality of minimum values ​​being a minimum value among the distances between a respective contour point between the first contour point and the M-th contour point of the plurality of contour points and the line segments. [4] The apparatus of claim 1, wherein the processor is configured to: to determine a sum of distances associated with an Nth layer based on a sum of minimum values ​​among the distances between contour points contained in the Nth layer and the line segments, where N is a natural number satisfying 1 ≦ N ≦ M, and the Nth layer belongs to the first to Mth layers contained in the object box, and where M is a natural number greater than or equal to 2; to determine a sum of distances associated with an (N+1)th layer based on the sum of distances associated with the Nth layer and a sum of minimum values ​​among the distances between contour points contained in the (N+1)th layer and the line segments, where the (N+1)th layer is adjacent to the Nth layer; and to determine the degree of mismatch 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 apparatus of claim 1, wherein the processor is configured to: to determine a sum of distances associated with an Nth layer based on a sum of a set of minimum values, each of the set of minimum values ​​being a minimum value among the distances between one of the contour points contained in the Nth layer and the line segments, where N is a natural number satisfying 1 ≦ N ≦ M, and the Nth layer is among a first layer to an Mth layer contained in the object box, and where M is a natural number greater than or equal to 2; and to determine the degree of mismatch based on a value obtained by summing distances associated with the first layer to the M-th layer and a number of the plurality of contour points representing the object. [6] The apparatus of claim 1, wherein the processor is configured to determine a distance between a contour point of the plurality of contour points and a line segment of the line segments based on the coordinates of the one contour point and the coordinates of two points passing through the one line segment. [7] The apparatus of claim 1, wherein the processor is configured to determine the degree of mismatch based on a value obtained by dividing a sum of the plurality of minimum values ​​by a number of the plurality of contour points representing the object. [8] The apparatus of claim 1, wherein the object box comprises three or more line segments. [9] The apparatus of claim 1, wherein the processor is configured to: assign a confidence value to indicate whether the object is a stationary object based on determining that the degree of deviation is greater than a threshold value for the degree of deviation; and to determine whether the object is the stationary object based on the reliability value. [10] The apparatus of claim 1, wherein the processor is configured to: to determine a first reliability value to indicate whether the object is a stationary object by determining a first degree of disagreement as the degree of disagreement; determine a second reliability value that is greater than the first reliability value based on determining a second degree of disagreement that is greater than the first degree of disagreement as the degree of disagreement; and to determine whether the object is a stationary object based on the first reliability value or the second reliability value. [11] The apparatus of claim 1, wherein the processor is configured to: to identify the object box as a specific shape; and to determine whether the object is an object that conforms to the specified shape based on the degree of mismatch. [12] The apparatus of claim 1, wherein the processor is configured to assign to the object an identifier indicating whether the object is a stationary object based on determining whether the object is a stationary object. [13] The apparatus of claim 1, wherein the processor is configured to: to determine a reliability value based on the degree of disagreement that indicates whether the object is a stationary object; and to determine whether the object is a stationary object using a value obtained by multiplying the reliability value by a weight. [14] Method comprising: Identifying an object box comprising a plurality of contour points representing an object based on detection information from the sensor; Determining a plurality of minimum values, each of the plurality of minimum values ​​being a minimum value among the distances between one of the plurality of contour points and line segments forming a boundary of the object box; Determining a degree of mismatch between the object box and the plurality of contour points based on at least one of the following: the majority of the minimum values ​​or a number of the plurality of contour points; and Output a signal based on the degree of mismatch that indicates whether the object is a stationary object. [15] The method of claim 14, wherein determining the degree of mismatch comprises: Determining a sum of distances between an N-th contour point and the line segments based on a minimum value among the distances between the N-th contour point and the line segments, where N is a natural number satisfying 1 ≦ N ≦ M, and the N-th contour point is among a first contour point to M-th contour point representing the object, and where M is a natural number of 2 or greater; Determining a second sum of distances between an (N+1)th contour point and the line segments based on the sum of distances and a minimum value among the distances between the (N+1)th contour point and the line segments, wherein the (N+1)th contour point is closest to the Nth contour point; and Determine the degree of mismatch based on a sum of distances between an Mth contour point and the line segments. [16] The method of claim 14, wherein determining the degree of mismatch comprises: Determining a minimum value among the distances between an Nth contour point and the line segments, where N is a natural number satisfying 1 ≦ N ≦ M, and the Nth contour point lies between a first contour point and an Mth contour point representing the object, where M is a natural number of 2 or greater; and Determining the degree of mismatch based on a sum of the plurality of minimum values, wherein each of the plurality of minimum values ​​is a minimum value among the distances between a respective contour point between the first contour point and the M-th contour point of the plurality of contour points and the line segments. [17] The method of claim 14, wherein determining the degree of mismatch comprises: Determining a sum of distances associated with an N-th layer based on a sum of a set of minimum values, each of the set of minimum values ​​being a minimum value among the distances between one of the contour points included in the N-th layer and the line segments, where N is a natural number satisfying 1 ≦ N ≦ M, and the N-th layer is among a first layer to an M-th layer included in the object box, and where M is a natural number greater than or equal to 2; Determining a sum of distances associated with an (N+1)th layer based on the sum of distances associated with the Nth layer and a sum of a second set of minimum values, each of the second set of minimum values ​​being a minimum value among the distances between one of the contour points included in the (N+1)th layer and the line segments, the (N+1)th layer being adjacent to the Nth layer; and Determining the degree of mismatch 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 of claim 14, wherein determining the degree of mismatch comprises: Determining a sum of distances associated with an N-th layer based on a sum of a set of minimum values, each of the set of minimum values ​​being a minimum value among the distances between one of the contour points contained in the N-th layer and the line segments, where N is a natural number satisfying 1 ≦ N ≦ M, and the N-th layer is among a first layer to an M-th layer contained in the object box, where M is a natural number greater than or equal to 2; and Determining the degree of mismatch based on a value obtained by summing distances associated with the first layer to the M-th layer and a number of the plurality of contour points representing the object. [19] The method of claim 14, wherein determining the plurality of minimum values ​​comprises determining a distance between a contour point of the plurality of contour points and a line segment of the line segments based on the coordinates of the one contour point and the coordinates of two points passing through the one line segment. [20] The method of claim 14, wherein determining the degree of mismatch comprises determining the degree of mismatch based on a value obtained by dividing a sum of the plurality of minimum values ​​by a number of contour points representing the object.

Images