Lane detection method and apparatus using the moving path of the moving object

Abstract

The present invention provides a lane detection method in which a plurality of points are obtained in which at least one moving body is detected in a plurality of image frames, a lane is detected using the plurality of points, and a lane is detected using the lane.

Description

Technology Field

[0001] The present invention relates to a method and apparatus for detecting a lane using the movement path of a moving body, and more specifically, to a method and apparatus for detecting a lane using the path traveled by a moving body. Background Technology

[0002] In relation to vehicle automation and / or autonomous driving, various methods for creating digital road maps have been developed in the past.

[0003] In the paper “Generating Lane Level Road Data from Vehicle Trajectories Using Kernel Density Estimation” by Uruwaragoda et al. (2013), published in the minutes of the 16th International IEEE Annual Meeting on Intelligent Transportation Systems (ITSC 2013), Issue 201, a method for estimating, for example, the number and width of lanes on a road is disclosed. To this end, based on a roadway-accurate road map, the center lines of the road are intersected at right angles at discrete intervals. For each of these vertical lines, intersection points with vehicle trajectories on the road are calculated, and kernel density estimation is performed for each. Thus, along the road, supporting points are generated that contain information on the number and width of lanes and can be finally computed.

[0004] In the paper “Mining GPS Traces for Map Refinement” by Schroedel et al., published in Data Mining and Knowledge Discovery (September 2004), information regarding the number of lanes and lane widths is derived by an algorithm that identifies center lines by first dividing the vehicle trajectories into segments without prior knowledge of the map. Then, in order to make statements about the lanes, vertical intervals to the vehicle trajectories along the center lines are classified by a density estimation method.

[0005] In the paper “Creating Enhanced Maps for Lane-Level Vehicle Navigation” by Betaille et al., published in IEEE Transaction on Intelligent Transportation Systems (April 2010, October 2010), a model-based approach is pursued in which models described through clothoids are matched to the measured trajectory data of vehicles. Prior art literature

[0006] Korean Patent Publication No. 10-2020-0012960 (February 5, 2020) The problem to be solved

[0007] The technical problem that the present invention aims to solve is to provide a method and apparatus for detecting multiple moving objects in an image frame, estimating their movement paths, and using this to detect lanes. means of solving the problem

[0008] One aspect of the present invention may include: a step of acquiring a plurality of points in which at least one moving body is detected in a plurality of image frames; a step of detecting a lane using the plurality of points; and a step of detecting a lane using the lane.

[0009] In addition, the step of detecting a lane using the above lanes can determine the lane by extracting the centerlines of two adjacent lanes among the multiple lanes when multiple lanes are detected.

[0010] Additionally, when at least one lane is determined for a plurality of detected lanes, the step of detecting the plurality of lanes and the step of detecting a lane using the lanes can be repeated outside the area containing a plurality of points corresponding to the at least one detected lane.

[0011] Additionally, the step of detecting the plurality of lanes may include: a step of determining a first lane from a travel path estimated using the plurality of points; and a step of estimating a second lane by translating the first lane.

[0012] Additionally, the step of determining the first lane may include: a step of randomly selecting n points among the plurality of points; a step of generating a curve using the n points; and a step of determining the first lane based on the number of points adjacent to the curve.

[0013] Additionally, the step of estimating the second lane may include: a step of translating the first lane in the lateral direction of the first lane within a preset movement range; and a step of determining the second lane based on the number of points adjacent to the translated first lane.

[0014] Additionally, the step of estimating the second lane may further include: a step of moving the second lane in the lateral direction of the second lane within the movement range when the determination of the second lane within the movement range is successful with respect to the first lane; and a step of determining the third lane based on the number of points adjacent to the moved second lane.

[0015] Additionally, the step of estimating the second lane may include a step of stopping the estimation of the second lane when, with respect to the first lane, the determination of the second lane within the movement range fails.

[0016] In addition, the lane may be a line extended by a preset distance from an end point in the direction of travel of the lane among a plurality of points defined for the lane.

[0017] Additionally, each of the above multiple points may be such that at least one moving body is detected in each of the above multiple image frames, and the center point of each of the at least one detected moving body is obtained as the point.

[0018] Another aspect of the present invention may include: a memory in which a plurality of points regarding the positions of moving bodies in a plurality of image frames are acquired and said plurality of points are stored; and a processor that detects a lane using said plurality of points and detects a lane using said lane.

[0019] In addition, when a plurality of lanes are detected, the processor can determine the lane by extracting the centerlines of two adjacent lanes among the plurality of lanes.

[0020] In addition, when at least one lane is determined among the detected multiple lanes, the process of detecting multiple lanes outside the area containing multiple points corresponding to the detected at least one lane and detecting lanes using the detected lanes can be repeated.

[0021] In addition, the processor can determine a first lane from the estimated movement path using the plurality of points and estimate a second lane by moving the first lane in parallel.

[0022] In addition, the processor may randomly select n points among the plurality of points, generate a curve using the n points, and determine the first lane based on the number of points adjacent to the curve.

[0023] Additionally, the processor may translate the first lane in a lateral direction to the first lane within a preset movement range, and determine the second lane based on the number of points adjacent to the translated first lane.

[0024] Additionally, if the processor succeeds in determining the second lane within the movement range for the first lane, it may move the second lane in the lateral direction of the second lane within the movement range and determine the third lane based on the number of points adjacent to the moved second lane.

[0025] Additionally, the processor may stop estimating the second lane if it fails to determine the second lane within the movement range for the first lane.

[0026] In addition, the lane may be a line extended by a preset distance from an end point in the direction of travel of the lane among a plurality of points defined for the lane.

[0027] Additionally, each of the above multiple points may be such that at least one moving body is detected in each of the above multiple image frames, and the center point of each of the at least one detected moving body is obtained as the point.

[0028] Another aspect of the present invention is a computer-readable recording medium storing a computer program, wherein the computer program may include instructions for the processor to perform a method comprising: acquiring a plurality of points in which at least one moving object is detected in a plurality of image frames when executed by the processor; detecting a lane using the plurality of points; and detecting a lane using the lane.

[0029] Another aspect of the present invention is a computer program stored on a computer-readable recording medium, wherein the computer program may include instructions for the processor to perform a method comprising: acquiring a plurality of points in which at least one moving object is detected in a plurality of image frames when executed by a processor; detecting a lane using the plurality of points; and detecting a lane using the lane. Effects of the invention

[0030] According to one aspect of the present invention described above, by providing a lane detection method and apparatus using the movement path of a moving body, a plurality of moving bodies are detected in an image frame to estimate the movement path, and a lane can be detected using this. Brief explanation of the drawing

[0031] FIG. 1 is a block diagram of a lane detection device according to one embodiment of the present invention. FIG. 2 is a block diagram conceptually illustrating an embodiment of the lane detection program of FIG. 1. FIG. 3 is a diagram showing an example of a plurality of points acquired by the processor of FIG. 1. FIG. 4 is a diagram showing an example of a first step generated by the processor of FIG. 1. FIG. 5 is a diagram showing an example of a second stage generated by the processor of FIG. 1. FIGS. 6 to 8 are drawings showing an example of a lane generated by the processor of FIG. 1. FIG. 9 is a flowchart of a lane detection method according to one embodiment of the present invention. Figure 10 is a detailed flowchart of the step of detecting a lane using multiple points of Figure 9. FIG. 11 is a detailed flowchart of the step of determining the first lane from the estimated movement path using multiple points of FIG. 10. Figure 12 is a detailed flowchart of the step of estimating the second lane by shifting the first lane of Figure 10 in parallel. Figure 13 is a detailed flowchart of the step of detecting a lane using multiple points of Figure 9. Specific details for implementing the invention

[0032] The advantages and features of the present invention and the methods for achieving them will become clear by referring to the embodiments described below in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below but can be implemented in various different forms. These embodiments are provided merely to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the invention, and the present invention is defined only by the scope of the claims.

[0033] In describing the embodiments of the present invention, specific descriptions of known functions or configurations will be omitted if it is determined that such detailed descriptions could unnecessarily obscure the essence of the invention. Furthermore, the terms described below are defined in consideration of their functions in the embodiments of the present invention, and these definitions may vary depending on the intentions or practices of the user or operator. Therefore, such definitions should be based on the content throughout this specification.

[0034] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings.

[0035] FIG. 1 is a block diagram of a lane detection device according to one embodiment of the present invention.

[0036] Referring to FIG. 1, the lane detection device (100) may include an input / output module (110), a processor (120), and a memory (130).

[0037] The processor (120) can acquire a plurality of video frames or a plurality of points in which at least one moving body is detected in a plurality of video frames through the input / output module (110).

[0038] Here, the video frames may be generated by a separate shooting device. In this case, the shooting device may be installed to photograph the road. That is, each of the multiple video frames may include the road and at least one moving object traveling on the road.

[0039] In addition, multiple points may be the center points of each of at least one moving object detected in the video frame.

[0040] To this end, the processor (120) detects at least one moving object in each of a plurality of video frames and obtains the center point of each of the at least one detected moving object as a point, or obtains a point detected by a separate moving object detection device through the input / output module (110).

[0041] In this regard, the processor (120) can detect moving objects in an image frame using a known technique. For example, the processor (120) can detect moving objects using a neural network model that has been pre-trained to detect moving objects in an image frame.

[0042] Meanwhile, the processor (120) can detect a lane using multiple points and detect a lane using the detected lane.

[0043] Here, a roadway is a line set for a moving object to move, and a lane can be a line indicating the boundary between lanes.

[0044] Meanwhile, the memory (130) may store information necessary for the execution of the lane detection program (200) and the lane detection program (200).

[0045] In the present specification, the lane detection program (200) may refer to software that includes instructions programmed to detect a lane using a plurality of points and to detect a lane using the detected lane.

[0046] Accordingly, the processor (120) can load the lane detection program (200) and information necessary for the execution of the lane detection program (200) from memory (130) in order to execute the lane detection program (200).

[0047] Meanwhile, the function and / or operation of the lane detection program (200) will be examined in detail through FIG. 2.

[0049] FIG. 2 is a block diagram conceptually illustrating an embodiment of the lane detection program of FIG. 1.

[0050] Referring to FIG. 2, the lane detection program (200) may include a lane detection unit (210) and a lane detection unit (220).

[0051] The lane detection unit (210) and lane detection unit (220) illustrated in FIG. 2 are conceptually divided to easily explain the functions of the lane detection program (200), but are not limited thereto. Depending on the embodiments, the functions of the lane detection unit (210) and lane detection unit (220) may be merged / separated and may be implemented as a series of commands included in a single program.

[0052] The lane detection unit (210) can detect a lane using multiple points. To this end, the lane detection unit (210) can determine a first lane from an estimated travel path using multiple points.

[0053] For example, the lane detection unit (210) can randomly select n points from a plurality of points (n is a natural number greater than or equal to 3), generate a curve using the selected n points, and determine the first lane based on the number of points adjacent to the generated curve.

[0054] That is, the lane detection unit (210) can generate an m-degree function curve using a selected portion of a plurality of points (m is a natural number greater than or equal to 2), and determine a first lane based on the number of points adjacent to the generated m-degree function curve.

[0055] According to an embodiment, the m-th degree function curve can be determined based on the number of points selected by the lane detection unit (210). Here, the relationship m = n - 1 may hold. For example, if three points are selected from a plurality of points, the generated curve equation may be a quadratic function curve.

[0056] In one embodiment, the lane detection unit (210) may determine the curve as the first lane when the number of points adjacent to the generated curve exceeds a preset threshold value, or generate a preset number of different curves and determine the one with the largest number of adjacent points among the generated curves as the first lane.

[0057] In this regard, the lane detection unit (210) may determine a first lane for multiple points using a RANSAC (Random Sample Consensus) algorithm.

[0058] Meanwhile, the lane detection unit (210) can estimate the second lane by moving the first lane in parallel. To do this, the lane detection unit (210) can move the first lane in parallel in the lateral direction of the first lane (e.g., in the case of a straight lane, in a direction perpendicular to the lane) within a preset movement range.

[0059] Here, the preset travel range may be set based on the distance between adjacent lanes set to proceed in the same direction, or based on the distance between adjacent lanes set for the same direction.

[0060] Therefore, for any one lane, another lane detected within a preset movement range may be an adjacent lane set to proceed in the same direction.

[0061] The lane detection unit (210) can determine the second lane based on the number of points adjacent to the first lane that has been paralleled.

[0062] In one embodiment, the lane detection unit (210) may determine the curve as the second lane when the number of points adjacent to the first lane that has been paralleled exceeds a preset threshold value, or, within the movement range, generate a plurality of curves that have been paralleled to the first lane at preset distance intervals, and determine the one among the generated plurality of curves that has the largest number of adjacent points as the second lane.

[0063] Accordingly, the lane detection unit (210) can detect another lane adjacent to the other lane that was successfully determined within a preset movement range for the detected lane.

[0064] For example, if the lane detection unit (210) succeeds in determining the second lane within a preset movement range for the first lane, it can move the second lane in a direction perpendicular to the second lane within the preset movement range and determine the third lane based on the number of points adjacent to the moved second lane.

[0065] In this way, the lane detection unit (210) can repeatedly perform the process of detecting other lanes within a preset movement range for any one lane. Accordingly, the lane detection unit (210) can detect at least one lane having a gap within a preset movement range.

[0066] At this time, the lane detection unit (210) may stop the lane detection process if it fails to determine another lane within a preset movement range for the detected lane.

[0067] Here, stopping the lane detection process may mean stopping the process of determining another curve by translating the previously detected curve.

[0068] For example, the lane detection unit (210) may stop estimating the second lane if it fails to determine the second lane within a preset movement range for the first lane.

[0069] Meanwhile, a lane may be a line extended by a preset distance from an end point in the direction of travel of the lane among a number of points defined for the lane.

[0070] To this end, the lane detection unit (210) can extend for each of the at least one detected lane by a predetermined distance from the end point in the direction of travel of the lane among a plurality of points defined for the lane.

[0071] In one embodiment, the preset distance may be 30 meters.

[0072] Accordingly, the lane detection unit (220) can detect a lane using the lanes. To this end, when multiple lanes are detected, the lane detection unit (220) can determine the lane by extracting the centerlines of two adjacent lanes among the multiple lanes.

[0073] At this time, the lane detection unit (220) can determine the lanes for both outer sides based on the distance interval between each lane and the inner lane determined by each of the two outer lanes for the two lanes detected on both outer sides.

[0074] Specifically, the lane detection unit (220) can determine the outer lane in the outer direction relative to the outer lane based on the distance between the outer lane and the inner lane determined in the inner direction where another lane is located relative to the outer lane, in the outer lane detected at the outermost side in the lateral direction of any one of the multiple lanes.

[0075] Meanwhile, when at least one lane is determined for the detected multiple lanes, the lane detection unit (210) can repeat the process of detecting multiple lanes outside the area containing multiple points corresponding to the detected at least one lane.

[0076] Accordingly, the lane detection unit (220) can repeat the process of detecting lanes using the detected lanes according to the operation of the lane detection unit (210).

[0077] At this time, the area containing multiple points corresponding to at least one detected lane may be an inner area according to the outer lane determined by the lane detection unit (220).

[0079] FIG. 3 is a diagram showing an example of a plurality of points acquired by the processor of FIG. 1.

[0080] Referring to FIG. 3, a plurality of moving bodies (10) detected in an image frame and a point determined for each of the plurality of moving bodies (10) can be identified.

[0081] Accordingly, it can be understood that multiple points are generated to represent movement paths for multiple moving bodies (10).

[0083] FIG. 4 is a diagram showing an example of a first stage generated by the processor of FIG. 1, and FIG. 5 is a diagram showing an example of a second stage generated by the processor of FIG. 1.

[0084] Referring to FIG. 4, a first lane (20) generated for a plurality of points can be seen, and referring to FIG. 5, a second lane (30) generated for a plurality of points based on the first lane (20) can be seen.

[0085] In this way, the processor (120) can determine a first lane (20) from an estimated movement path using multiple points, and estimate a second lane (30) by moving the first lane (20) in parallel.

[0087] FIGS. 6 to 8 are drawings showing an example of a lane generated by the processor of FIG. 1.

[0088] Referring to FIG. 6, a lane (40) detected as the centerline of two different lanes (30) can be seen, and referring to FIG. 7, a number of lanes (40) determined for a number of points can be seen.

[0089] In this way, when a number of lanes is detected, the processor (120) can determine the lane (40) by extracting the centerlines of two adjacent lanes among the number of lanes.

[0090] At this time, the processor (120) can determine an outer lane (40b) in the outer direction for the outer lane based on the distance between the outer lane and the inner lane (40a) determined in the inner direction where another lane is located relative to the outer lane, in the outer lane detected at the outermost side in the lateral direction of one of the multiple lanes.

[0091] Meanwhile, referring to FIG. 8, a number of lanes (40) determined in the entire area of ​​the video frame can be seen.

[0092] In this way, when at least one lane (40) is determined for a plurality of detected lanes, the processor (120) can repeat the process of detecting a plurality of lanes and the process of detecting a lane (40) using the detected lanes outside the area containing a plurality of points corresponding to the at least one detected lane (40).

[0094] FIG. 9 is a flowchart of a lane detection method according to one embodiment of the present invention.

[0095] Referring to FIG. 9, the processor (120) can acquire multiple points in which at least one moving body is detected in multiple video frames (S100).

[0096] Accordingly, the processor (120) can detect a lane using multiple points (S200) and detect a lane using the detected lane (S300).

[0098] Figure 10 is a detailed flowchart of the step of detecting a lane using multiple points of Figure 9.

[0099] Referring to FIG. 10, the processor (120) can determine a first lane from an estimated movement path using a plurality of points (S210).

[0100] Accordingly, the processor (120) can estimate the second lane by shifting the first lane in parallel (S220).

[0102] FIG. 11 is a detailed flowchart of the step of determining the first lane from the estimated movement path using multiple points of FIG. 10.

[0103] Referring to FIG. 11, the processor (120) can randomly select n points from a plurality of points (S211) and generate a curve using the selected n points (S212).

[0104] Accordingly, the processor (120) can determine the first lane based on the number of points adjacent to the curve (S213).

[0106] Figure 12 is a detailed flowchart of the step of estimating the second lane by shifting the first lane of Figure 10 in parallel.

[0107] Referring to FIG. 12, the processor (120) can move the first lane in the lateral direction of the first lane within a preset movement range (S221).

[0108] Accordingly, the processor (120) can determine the second lane based on the number of points adjacent to the first lane that has been translated (S222).

[0110] Figure 13 is a detailed flowchart of the step of detecting a lane using multiple points of Figure 9.

[0111] Referring to FIG. 13, the processor (120) can determine whether the decision for the second lane within the movement range was successful for the first lane (S230).

[0112] At this time, if the processor (120) succeeds in determining the second lane within the movement range for the first lane, it can move the second lane in parallel in the lateral direction of the second lane within the movement range (S240).

[0113] In this case, the processor (120) can determine the third lane based on the number of points adjacent to the second lane that has been translated (S250).

[0114] Meanwhile, the processor (120) may stop estimating the second lane if it fails to determine the second lane within the movement range for the first lane (S260).

[0115] Various embodiments of this document may be implemented as software (e.g., a program) comprising instructions stored in a machine-readable storage medium (e.g., memory (internal or external memory)) of a machine (e.g., a computer). The machine may include an electronic device (e.g., a device) according to the disclosed embodiments, which is a device capable of calling instructions stored from the storage medium and operating according to the called instructions. When said instructions are executed by a processor (e.g., a processor), the processor may perform a function corresponding to said instructions directly or using other components under the control of said processor. The instructions may include code generated or executed by a compiler or an interpreter. The machine-readable storage medium may be provided in the form of a non-transitory storage medium. Here, "non-transitory" means only that the storage medium does not contain a signal and is tangible, and does not distinguish whether data is stored semi-permanently or temporarily in the storage medium.

[0116] According to one embodiment, the method according to the various embodiments disclosed in this document may be provided by being included in a computer program product.

[0117] The above description is merely an illustrative explanation of the technical concept of the present invention, and those skilled in the art to which the present invention pertains will be able to make various modifications and variations within the scope of the essential quality of the present invention. Accordingly, the embodiments disclosed in the present invention are intended to explain, not limit, the technical concept of the present invention, and the scope of the technical concept of the present invention is not limited by such embodiments. The scope of protection of the present invention shall be interpreted by the claims below, and all technical concepts within the equivalent scope shall be interpreted as being included within the scope of rights of the present invention. Explanation of the symbols

[0118] 10: Moving body 20: By the first car 30: 2nd lane 40: Lane 40a: Inner lane 40b: outer lane 100: Lane detection device

Claims

Claim 1 A lane detection method performed by a lane detection device, comprising: a step of acquiring a plurality of points for the position of at least one moving body in a plurality of image frames; a step of detecting a lane using the plurality of points; and a step of detecting a lane using the lane; wherein the step of detecting a lane includes: a step of determining a first lane from a movement path estimated using the plurality of points; a step of estimating a second lane by moving the first lane in parallel to the lateral direction of the first lane within a preset movement range; and a step of determining whether to determine the estimated second lane based on whether the number of points adjacent to the estimated second lane exceeds a preset threshold value; wherein in the step of detecting a lane using the lane, if a plurality of lanes are detected, the centerlines of two adjacent lanes among the plurality of lanes are extracted to determine the lane, and if at least one lane is determined for the plurality of detected lanes, the step of detecting the lane and the step of detecting the lane are repeated outside a predetermined area containing a point corresponding to the at least one detected lane. Claim 2 delete Claim 3 delete Claim 4 delete Claim 5 A lane detection method according to claim 1, wherein the step of determining the first lane comprises: a step of randomly selecting n points (where n is a natural number greater than or equal to 3) among the plurality of points; a step of generating a curve using the n points; and a step of determining the first lane based on the number of points adjacent to the curve. Claim 6 delete Claim 7 A lane detection method according to claim 1, wherein the step of detecting the lane further comprises: a step of paralleling the second lane in the lateral direction of the second lane within the movement range when the determination of the second lane within the movement range is successful with respect to the first lane; and a step of determining a third lane based on the number of points adjacent to the parallelized second lane. Claim 8 A lane detection method according to claim 1, wherein the step of detecting the lane comprises: a step of stopping the estimation of the second lane when, with respect to the first lane, the determination of the second lane within the movement range fails. Claim 9 A lane detection method according to claim 1, wherein the lane is a line extended by a preset distance from an end point in the direction of travel of the lane among a plurality of points defined for the lane. Claim 10 A lane detection method according to claim 1, wherein each of the plurality of points is obtained by detecting at least one moving body in each of the plurality of image frames, and the center point of each of the at least one detected moving body is obtained as the point. Claim 11 A lane detection device comprising: a memory in which a plurality of points regarding the positions of moving bodies in a plurality of video frames are obtained and said plurality of points are stored; and a processor that detects a lane using the plurality of points and detects a lane using said lane, wherein the processor determines a first lane from a movement path estimated using the plurality of points, estimates a second lane by moving the first lane in parallel to the lateral direction of the first lane within a preset movement range, determines whether to determine the estimated second lane based on whether the number of points adjacent to the estimated second lane exceeds a preset threshold value, and when a plurality of lanes are detected, extracts the centerlines of two adjacent lanes among the plurality of lanes to determine said lane, and when at least one lane is determined for the plurality of detected lanes, repeats the process of detecting the lane and the process of detecting said lane outside a predetermined area containing a point corresponding to the at least one detected lane. Claim 12 delete Claim 13 delete Claim 14 delete Claim 15 A lane detection device according to claim 11, wherein the processor randomly selects n points (where n is a natural number greater than or equal to 3) among the plurality of points, generates a curve using the n points, and determines the first lane based on the number of points adjacent to the curve. Claim 16 delete Claim 17 A lane detection device according to claim 11, wherein, when the processor succeeds in determining the second lane within the movement range for the first lane, the processor moves the second lane in the lateral direction of the second lane within the movement range and determines a third lane based on the number of points adjacent to the moved second lane. Claim 18 In claim 11, the lane detection device, wherein the processor stops estimating the second lane when it fails to determine the second lane within the movement range for the first lane. Claim 19 In claim 11, the lane is a lane detection device in which the lane is a line extended by a preset distance from an end point in the direction of travel of the lane among a plurality of points defined for the lane. Claim 20 A lane detection device according to claim 11, wherein each of the plurality of points is obtained by detecting at least one moving body in each of the plurality of image frames, and the center point of each of the detected at least one moving body is obtained as the point. Claim 21 A computer-readable recording medium storing a computer program, wherein the computer program, when executed by a processor, comprises: a step of acquiring a plurality of points regarding the position of at least one moving body in a plurality of image frames; a step of detecting a lane using the plurality of points; and a step of detecting a lane using the lane; wherein the step of detecting a lane comprises: a step of determining a first lane from a movement path estimated using the plurality of points; and a step of estimating a second lane by translating the first lane in the lateral direction of the first lane within a preset movement range. A computer-readable recording medium comprising instructions for a processor to perform a method of detecting a lane using the lane, wherein, when a plurality of lanes are detected, the centerlines of two adjacent lanes among the plurality of lanes are extracted to determine the lane, and when at least one lane is determined for the plurality of detected lanes, the step of detecting the lane and the step of detecting the lane are repeated outside a predetermined area containing a point corresponding to the at least one detected lane. Claim 22 A computer program stored on a computer-readable recording medium, wherein, when executed by a processor, the computer program comprises: a step of acquiring a plurality of points regarding the position of at least one moving body in a plurality of image frames; a step of detecting a lane using the plurality of points; and a step of detecting a lane using the lane; wherein the step of detecting a lane comprises: a step of determining a first lane from a movement path estimated using the plurality of points; and a step of estimating a second lane by translating the first lane in the lateral direction of the first lane within a preset movement range. A computer program comprising instructions for the processor to perform a method of detecting a lane using the lane, wherein, when a plurality of lanes are detected, the centerlines of two adjacent lanes among the plurality of lanes are extracted to determine the lane, and when at least one lane is determined for the plurality of detected lanes, the step of detecting the lane and the step of detecting the lane are repeated outside a predetermined area containing a point corresponding to the at least one detected lane.

Images