CN116259022A - Tracking method based on visual lane line, electronic equipment, medium and vehicle - Google Patents
Tracking method based on visual lane line, electronic equipment, medium and vehicle Download PDFInfo
- Publication number
- CN116259022A CN116259022A CN202211698573.7A CN202211698573A CN116259022A CN 116259022 A CN116259022 A CN 116259022A CN 202211698573 A CN202211698573 A CN 202211698573A CN 116259022 A CN116259022 A CN 116259022A
- Authority
- CN
- China
- Prior art keywords
- lane line
- matching
- track
- detection result
- lane
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
The invention relates to the technical field of vehicle perception, in particular to a tracking method based on a visual lane line, electronic equipment, a medium and a vehicle, and aims to solve the technical problem that the conventional lane line tracking method cannot effectively detect lane line abnormality for a long time. For this purpose, the visual lane line-based tracking method of the present invention comprises: obtaining a lane line detection result; judging whether the lane line detection result meets a preset condition; under the condition that the lane line detection result meets the preset condition, determining a measurement value based on the lane line detection result; creating a Kalman filter of a predicted track and initializing the Kalman filter; matching the measured value with the predicted track to obtain a matching result; updating the Kalman filter state of the predicted track according to the matching result; and outputting a tracking result based on the Kalman filter state of the updated predicted trajectory. Thus, the tracking efficiency of the lane line is improved.
Description
Technical Field
The invention relates to the technical field of vehicle perception, and particularly provides a tracking method based on visual lane lines, electronic equipment, a medium and a vehicle.
Background
After the lane line detection is completed, the lane line tracking is needed, which is also a key technology for intelligent automobile auxiliary driving.
The lane lines restrict the vehicle running area by providing road boundary information. However, almost all actual road conditions, more or less conditions such as repairing road surfaces, accumulated water, shading, blocked lane lines, fouling and the like, and more extreme working conditions such as jolt, ascending and descending of vehicles and the like, may influence the accuracy and the robustness of single-frame lane line detection. Therefore, the missing and wrong lane lines need to be detected in a complementary mode according to the tracking algorithm by combining the historical state with the road geometric relationship, and the lane lines are more stable in space position.
The existing lane line tracking method can effectively improve the stability and the anti-interference capability of road edge detection, but cannot detect the lane line abnormality for a long time, so that the tracking efficiency of the lane line is lower.
Accordingly, there is a need in the art for a new visual lane-based tracking scheme to address the above-described problems.
Disclosure of Invention
The present invention has been made to overcome the above-mentioned drawbacks, and to provide a solution or at least partially solve the above-mentioned technical problems. The invention provides a tracking method based on a visual lane line, electronic equipment, a medium and a vehicle.
In a first aspect, the present invention provides a visual lane line-based tracking method, the method comprising: obtaining a lane line detection result; judging whether the lane line detection result meets a preset condition or not; determining a measurement value based on the lane line detection result under the condition that the lane line detection result meets a preset condition; creating a Kalman filter of a predicted track and initializing the Kalman filter; matching the measured value with the predicted track to obtain a matching result; updating the Kalman filter state of the predicted track according to the matching result; and outputting a tracking result based on the updated Kalman filter state of the predicted trajectory.
In one embodiment, determining whether the lane line detection result meets a preset condition includes: determining a lane width based on the lane line detection result; judging whether the lane width is within a preset threshold range, if so, determining that the lane line detection result meets a preset condition; if not, determining that the lane line detection result does not meet the condition; and/or
Judging whether the lane line detection result meets a preset condition or not comprises the following steps: determining the area between a first lane line curve and a second lane line curve corresponding to the lane line detection result, wherein the first lane line curve is a lane line curve corresponding to a high-precision map; judging whether the area is larger than a first threshold value or not; if yes, determining that the lane line detection result does not meet a preset condition; if not, determining that the lane line detection result meets the condition.
In one embodiment, determining the measurement value based on the lane line detection result includes: acquiring the ordinate of N anchor points from the lane line detection result, wherein N is more than or equal to 4; determining the abscissa of the N anchor points based on the ordinate of the N anchor points; the measurement value is determined based on the abscissa of the N anchor points.
In one embodiment, the predicted track includes a primary track and a backup track; matching the measured value with the predicted track to obtain a matching result, including: matching the measurement value with the main track to obtain a first matching result, wherein the first matching result comprises any one of small-error matching, large-error matching and unmatched; and under the condition that the first matching result is large-error matching, matching the measurement value with the standby track to obtain a second matching result.
In one embodiment, the matching the measurement value with the main track to obtain a first matching result includes: calculating the IOU distance between the rectangular frame of each anchor point in the measurement value and the rectangular frame of the anchor point in the main track and the center distance of the anchor point; wherein the method comprises the steps of
In the above formula, IOU (i, j) is IOU distance, area (R) i ) To measure the Area of the rectangular frame of any anchor point in the value, area (R j ) The Area of the rectangular frame of the anchor point in the main track is the Area of intersection of the rectangular frame of any anchor point in the measurement value and the rectangular frame of the anchor point in the main track;
Dcenter＝d/c
in the above formula, dcenter is the center distance of an anchor point, d is the center point distance of two rectangular frames, and c is the diagonal length of the minimum circumscribed rectangle of the two rectangular frames;
determining an IOU distance average value and an anchor point center distance average value based on the IOU distance and the anchor point center distance;
judging whether the IOU distance average value is smaller than or equal to a second threshold value;
if the IOU distance average value is smaller than or equal to a second threshold value, small error matching is performed, and if the IOU distance average value is larger than the second threshold value, whether the anchor point center distance average value is smaller than or equal to a third threshold value is judged;
if yes, large error matching is performed, and if not, no matching is performed.
In one embodiment, updating the kalman filter state of the predicted trajectory according to the matching result includes: updating the Kalman filter state of the main track under the condition that the first matching result is small-error matching; updating the Kalman filter state of the standby track under the condition that the first matching result is large error matching and the second matching result is small error matching; and under the condition that the first matching result is unmatched, updating the Kalman filter state of the main track in a preset mode, and initializing the Kalman filter of the standby track in a measured value.
In one embodiment, outputting tracking results based on updated kalman filter states of the predicted trajectories includes: confirming the predicted track according to the Kalman filter state of the predicted track; and outputting a tracking result based on the confirmed predicted track.
In a second aspect, there is provided an electronic device comprising a processor and a storage means adapted to store a plurality of program code adapted to be loaded and executed by the processor to perform the visual lane line-based tracking method of any of the preceding claims.
In a third aspect, there is provided a computer readable storage medium having stored therein a plurality of program code adapted to be loaded and executed by a processor to perform the visual lane line based tracking method of any of the preceding claims.
In a fourth aspect, a vehicle is provided, the vehicle comprising the aforementioned electronic device.
The technical scheme provided by the invention has at least one or more of the following beneficial effects:
according to the visual lane line-based tracking method, after the lane line detection result is obtained, whether the lane line detection result meets the preset condition is judged, so that the lane line detection result is firstly subjected to abnormal detection before the measurement value is matched with the predicted track, and if the lane line detection result is abnormal, the lane line detection result is not tracked, so that the abnormal lane line is prevented from being tracked, and the lane line tracking efficiency is improved.
Drawings
The present disclosure will become more readily understood with reference to the accompanying drawings. As will be readily appreciated by those skilled in the art: the drawings are for illustrative purposes only and are not intended to limit the scope of the present invention. Moreover, like numerals in the figures are used to designate like parts, wherein:
FIG. 1 is a flow chart of the main steps of a visual lane line-based tracking method according to one embodiment of the invention;
FIG. 2 is a schematic diagram of a rectangular box of any anchor point in the metrology value and a rectangular box of an anchor point in the main track, in accordance with one embodiment of the present invention;
FIG. 3 is a schematic diagram of a minimum bounding rectangle according to one embodiment of the present invention;
FIG. 4 is a flow diagram of a lane tracking method according to one embodiment of the invention;
FIG. 5 is a complete flow diagram of a lane tracking method according to one embodiment of the invention;
fig. 6 is a schematic structural view of an electronic device according to an embodiment of the present invention.
Detailed Description
Some embodiments of the invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are merely for explaining the technical principles of the present invention, and are not intended to limit the scope of the present invention.
In the description of the present invention, a "module," "processor" may include hardware, software, or a combination of both. A module may comprise hardware circuitry, various suitable sensors, communication ports, memory, or software components, such as program code, or a combination of software and hardware. The processor may be a central processor, a microprocessor, an image processor, a digital signal processor, or any other suitable processor. The processor has data and/or signal processing functions. The processor may be implemented in software, hardware, or a combination of both. Non-transitory computer readable storage media include any suitable medium that can store program code, such as magnetic disks, hard disks, optical disks, flash memory, read-only memory, random access memory, and the like. The term "a and/or B" means all possible combinations of a and B, such as a alone, B alone or a and B. The term "at least one A or B" or "at least one of A and B" has a meaning similar to "A and/or B" and may include A alone, B alone or A and B. The singular forms "a", "an" and "the" include plural referents.
At present, the existing lane line tracking method can effectively improve the stability and the anti-interference capability of road edge detection, but cannot effectively detect the lane line abnormality for a long time, so that the lane line tracking efficiency is low.
Therefore, the application provides a tracking method, electronic equipment, medium and vehicle based on visual lane lines, which are characterized in that after a lane line detection result is obtained through a deep learning model, whether the lane line detection result meets a preset condition is judged, so that the detection result of the lane lines is firstly subjected to abnormal detection before a measured value is matched with a predicted track, and if the lane line detection result is abnormal, the lane line detection result is not tracked, so that the tracking of abnormal lane lines is avoided, and the tracking efficiency of the lane lines is improved.
Referring to fig. 1, fig. 1 is a schematic flow chart of main steps of a visual lane line-based tracking method according to an embodiment of the present invention.
As shown in fig. 1, the tracking method based on visual lane lines in the embodiment of the present invention mainly includes the following steps S101 to S106.
Step S101: and obtaining a lane line detection result, and judging whether the lane line detection result meets a preset condition.
Specifically, the forward camera is a sensor commonly used for automatic driving, and can be used for collecting images including forward lane lines during the running process of the vehicle. The image is input into a deep learning model, the detection results of the left lane line and the right lane line of the current lane can be obtained, and then the obtained detection results of the lane lines are subjected to abnormality judgment.
In a specific embodiment, determining whether the lane line detection result meets a preset condition includes: determining a lane width based on the lane line detection result; judging whether the lane width is within a preset threshold range, if so, determining that the lane line detection result meets a preset condition; if not, determining that the lane line detection result does not meet the condition; and/or
Judging whether the lane line detection result meets a preset condition or not comprises the following steps: determining the area between a first lane line curve and a second lane line curve corresponding to the lane line detection result, wherein the first lane line curve is a lane line curve corresponding to a high-precision map; judging whether the area is larger than a first threshold value or not; if yes, determining that the lane line detection result does not meet a preset condition; if not, determining that the lane line detection result meets the condition.
The preset threshold range refers to a lane width threshold range. The first threshold refers to the accuracy error threshold.
Judging whether the lane line detection result is abnormal or not through at least one of the conditions. For example, the lane width is calculated from the lane line detection result, and if the lane width is not within the lane width threshold range, for example, within the range of 1m to 5m, it is regarded that there is an abnormality in the lane line detection result, and the tracking is not participated. In addition, the area clamped between the lane line curve corresponding to the high-precision map and the lane line curve corresponding to the lane line detection result can be determined, whether the area is larger than an area threshold value is judged, if yes, it is determined that the lane line detection result does not meet the preset condition, detection abnormality exists, and tracking is not participated. In one embodiment, the area is a unit area.
Before lane tracking, whether the lane detection result is abnormal or not is judged, and the lane detection result is not tracked under the condition that the lane detection result is abnormal, so that the error tracking of the abnormal lane is avoided, the lane tracking efficiency is improved, and the safe driving of a vehicle is facilitated.
Step S102: and under the condition that the lane line detection result meets a preset condition, determining a measurement value based on the lane line detection result.
In one embodiment, determining the measurement value based on the lane line detection result includes: acquiring the ordinate of N anchor points from the lane line detection result, wherein N is more than or equal to 4; determining the abscissa of the N anchor points based on the ordinate of the N anchor points; the measurement value is determined based on the abscissa of the N anchor points.
Specifically, the vehicle forward direction is taken as the y-axis (ordinate) direction, the direction perpendicular to the ground is taken as the z-axis direction, and the direction perpendicular to the plane in which the y-axis and the z-axis lie and satisfying the right-hand rule is taken as the x-axis direction.
Generally, the lane line detection result expression is obtained by taking a cubic curve as an example. The two curves of the left lane line and the right lane line need to be processed independently, and the processing methods are consistent.
Illustratively, let N =The step of determining the measurement value will be described in detail with reference to fig. 4. The detected lane lines are first sampled for an anchor point, meaning a point of fixed y-coordinate sampled on the curve. At least 4 points are needed to calculate a curve equation for a 3-time curve, so 4 different y coordinates are selected on the cubic curve, the 4 y coordinates are fixed values, and corresponding x coordinates are calculated through the curve equation and recorded as x 0 ，x 1 ，x 2 ，x 3 . The selected y-coordinate distribution is kept at 2:2:1:1 in consideration of the smoothness of the road edge detection range and the overall output. The subsequent tracking is based on anchor tracking, the measurement value is the x coordinate of the anchor, i.e. the filter observation vector is z (k) = (x) 0 ,x 1 ,x 2 ,x 3 ) T 。
Step S103: a kalman filter of the predicted trajectory is created and initialized.
Specifically, the predicted track includes a main track and a spare track. The Kalman filter of the established main track and the spare track can be expressed as
Wherein v is 0 ,v 1 ,v 2 ,v 3 Respectively represent the movement speeds of the 4 anchor points in the x-axis direction in turn.
The detection result of each lane line is converted into an observation vector z (k) in step S102, and z (0) of the first frame is selected as an initialization value of a kalman filter of the main track and the standby track.
Step S104: and matching the measured value with the predicted track to obtain a matching result.
In one embodiment, the predicted track includes a primary track and a backup track; matching the measured value with the predicted track to obtain a matching result, including: matching the measurement value with the main track to obtain a first matching result, wherein the first matching result comprises any one of small-error matching, large-error matching and unmatched; and under the condition that the first matching result is large-error matching, matching the measurement value with the standby track to obtain a second matching result.
Specifically, the main track and the spare track are constructed based on a three-dimensional coordinate system with the vehicle as an origin, wherein the three-dimensional coordinate system has a vehicle advancing direction as a y-axis (ordinate) direction, a direction perpendicular to the ground as a z-axis direction, and a direction perpendicular to a plane in which the y-axis and the z-axis lie and satisfying a right-hand rule as an x-axis direction.
In a specific embodiment, the matching the measurement value with the main track to obtain a first matching result includes:
calculating the IOU distance between the rectangular frame of any anchor point in the measurement value and the rectangular frame of the anchor point in the main track and the center distance of the anchor point; wherein the method comprises the steps of
In the above formula, IOU (i, j) is IOU distance, area (R) i ) To measure the Area of the rectangular frame of any anchor point in the value, area (R j ) The Area of the rectangular frame of the anchor point in the main track is the Area of intersection of the rectangular frame of any anchor point in the measurement value and the rectangular frame of the anchor point in the main track;
Dcenter＝d/c
in the above formula, dcenter is the center distance of an anchor point, d is the center point distance of two rectangular frames, and c is the diagonal length of the minimum circumscribed rectangle of the two rectangular frames; the rectangular frame of any anchor point in the measurement value and the rectangular frame of the anchor point in the main track are specifically shown in fig. 2, and the minimum circumscribed rectangle of the two rectangular frames is specifically shown as the maximum rectangle in fig. 3;
determining an IOU distance average value and an anchor point center distance average value based on the IOU distance and the anchor point center distance;
judging whether the IOU distance average value is smaller than or equal to a second threshold value;
if the IOU distance average value is smaller than or equal to a second threshold value, small error matching is performed, and if the IOU distance average value is larger than the second threshold value, whether the anchor point center distance average value is smaller than or equal to a third threshold value is judged;
if yes, large error matching is performed, and if not, no matching is performed.
Specifically, in the process of performing correlation matching between the measured value and the predicted track, the measured value is first matched with the main track to obtain any one of matching results of small error matching, large error matching and non-matching. In addition, when the result obtained by matching the measured value with the main track is large-error matching, the correlation matching of the measured value and the standby track is carried out, and any one matching result of small-error matching, large-error matching and non-matching is also obtained.
Whether the measurement value is matched with the main track or the standby track is determined whether the measurement value is successfully matched with the main track or the standby track by judging whether the IOU distance average value and the anchor point center distance average value meet preset conditions or not. Illustratively, the correlation matching between the measurement value and the main track is described in detail.
Taking 4 anchor points as an example, in the process of matching the measurement value with the main track, respectively calculating the IOU distance and the anchor point center distance of the rectangular frame of each anchor point in the measurement value and the rectangular frame of the corresponding anchor point in the main track through the two formulas, then determining an IOU distance average value and an anchor point center distance average value based on the IOU distance and the anchor point center distance corresponding to each anchor point, and if the IOU distance average value is smaller than or equal to a second threshold value, then performing small-error matching, wherein the second threshold value is a small-error matching distance threshold value. If the IOU distance average value is larger than the small error matching distance threshold value and the anchor point center distance average value is smaller than or equal to a third threshold value, the large error matching is performed, otherwise, the large error matching is performed, and the third threshold value is the large error matching threshold value. And when the matching result of the measured value and the main track is large error matching, continuing to perform correlation matching on the measured value and the standby track, wherein the correlation matching principle of the measured value and the standby track is similar to that of the measured value and the main track, and the specific process of performing correlation matching on the measured value and the standby track can be referred to the process of performing correlation matching on the measured value and the main track, which is not repeated herein.
The matching correlation between the measured value and the main track or the standby track is judged by combining the IOU distance and the anchor point center distance, so that the accuracy of a matching result is further improved, and a basic support is provided for safe driving of the vehicle.
Step S105: and updating the Kalman filter state of the predicted track according to the matching result.
In one embodiment, updating the kalman filter state of the predicted trajectory according to the matching result includes: updating the Kalman filter state of the main track under the condition that the first matching result is small-error matching; updating the Kalman filter state of the standby track under the condition that the first matching result is large error matching and the second matching result is small error matching; and under the condition that the first matching result is unmatched, updating the Kalman filter state of the main track in a preset mode, and initializing the Kalman filter of the standby track in a measured value.
The preset mode is a non-measuring mode, which means that the predicted track is updated by using the Kalman filtering value at the previous moment. Specifically, when the matching result of the measurement value and the main track is small error matching, updating the state of the Kalman filter of the main track, and exiting the matching. If the matching state of the measured value and the main track is large error matching, the matching of the measured value and the standby track is entered, if the matching result of the measured value and the standby track is small error matching, the Kalman filter state of the standby track is updated, and meanwhile, the main track state is updated in a non-measured mode, and the matching is exited. Otherwise, the state of the main track is updated in a non-measuring mode, a Kalman filter of the standby track is initialized according to the measuring value, and the matching is exited.
If the measurement value is not matched with the main track, updating the main track state in a non-measurement mode, initializing a Kalman filter of the standby track with the measurement value, and exiting the matching.
Specifically, tracking filtering is carried out on the track according to different matching results, and the main body of the filtering algorithm is Kalman filtering. The target track state vector is:
wherein v is 0 ,v 1 ,v 2 ,v 3 Respectively represent the movement speeds of the 4 anchor points in the x-axis direction in turn.
The update status is divided into two types: there are measured updated trajectories and there are no measured updated trajectories.
The tracking state quantity includes a continuous unassociated count counter (time_sine_update) and an associated count counter (hits).
When the measurement update track exists, the unassociated times counter is cleared, the associated times counter is increased by one, the Kalman filtering is updated by using the measurement value, and the tracking state is updated.
The measurement-free update track is updated by adding one to the unassociated times counter, the Kalman filtering is updated by using the Kalman filtering value at the previous moment, and the state update is tracked.
Assuming that the measurement noise is gaussian white noise, the metrology model can be expressed as:
z(k)＝x(k)+v(k)
where z (k) represents an observation vector and v (k) represents observation noise.
Prediction state and update error covariance matrix:
wherein the method comprises the steps of
Is a state vector predicted at the next time, x (k) is a state estimate at the current time,
for the error covariance predicted at the next time, P (k) is the error covariance estimate at the current time, and ω (k) is the process noise.
Calculating Kalman filtering gain:
updating the state and the error covariance matrix:
where x (k+1) is the anchor x coordinate vector that tracks the final output.
Step S106: and outputting a tracking result based on the Kalman filter state of the updated predicted trajectory.
In one embodiment, outputting the tracking result based on the updated kalman filter state of the predicted trajectory includes: confirming the predicted track according to the Kalman filter state of the predicted track; and outputting a tracking result based on the confirmed predicted track.
Specifically, when there is a measurement value, the track state is confirmed if the association count counter is equal to or greater than the association counter threshold, and is not confirmed if the association count counter is less than the association counter threshold. And when the measurement value is not measured, if the number of times of unassociated counter is smaller than or equal to the threshold value of the continuous unassociated times, continuing the track state at the last moment, otherwise, deleting the track state, and resetting the associated number of times counter.
Further, when the main track state is confirmed, outputting a main track result; and if the main track state is deleted, replacing the standby track with the main track and outputting the main track.
Based on the above steps S101 to S106, specifically, after the lane line detection result is obtained through the deep learning model, it is determined whether the lane line detection result meets the preset condition, so that before the measurement value is matched with the predicted track, the anomaly detection is performed on the lane line detection result, if the lane line detection result is determined to be anomaly, the lane line detection result is not tracked, so that error tracking on the anomaly lane line is avoided, and the tracking efficiency of the lane line is improved.
In a specific embodiment, as shown in the flow chart of the lane line tracking method shown in fig. 4, firstly, the detection results of the left lane line and the right lane line are obtained, the lane line is detected abnormally, then the measured value is matched with the main track or the standby track to obtain a matching result, secondly, the track is updated according to different matching results, and after the track is confirmed, the tracking result is output.
In a specific embodiment, as shown in a complete flow chart of the lane tracking method shown in fig. 5, after the lane is acquired, firstly, abnormality detection is performed, if there is an abnormality, tracking is exited, and if there is no abnormality, whether the main track and the standby track are initialized is determined. If so, performing correlation matching on the main track and the measurement value, updating the Kalman filter state corresponding to the main track under the condition that the matching result of the main track and the measurement value belongs to small error matching, and further performing correlation matching on the measurement value and the standby track under the condition that the matching result of the main track and the measurement value belongs to large error matching. And under the condition that the measured value and the standby track are matched with small errors, updating the Kalman filter state of the standby track, and replacing the standby track with the current main track and outputting the current main track. Therefore, the tracking efficiency and the tracking precision of the lane line are improved, and the safety and the stability of automatic driving are improved.
It should be noted that, although the foregoing embodiments describe the steps in a specific order, it will be understood by those skilled in the art that, in order to achieve the effects of the present invention, the steps are not necessarily performed in such an order, and may be performed simultaneously (in parallel) or in other orders, and these variations are within the scope of the present invention.
It will be appreciated by those skilled in the art that the present invention may implement all or part of the above-described methods according to the above-described embodiments, or may be implemented by means of a computer program for instructing relevant hardware, where the computer program may be stored in a computer readable storage medium, and where the computer program may implement the steps of the above-described embodiments of the method when executed by a processor. Wherein the computer program comprises computer program code which may be in source code form, object code form, executable file or some intermediate form etc. The computer readable storage medium may include: any entity or device, medium, usb disk, removable hard disk, magnetic disk, optical disk, computer memory, read-only memory, random access memory, electrical carrier wave signals, telecommunications signals, software distribution media, and the like capable of carrying the computer program code. It should be noted that the computer readable storage medium may include content that is subject to appropriate increases and decreases as required by jurisdictions and by jurisdictions in which such computer readable storage medium does not include electrical carrier signals and telecommunications signals.
The invention further provides electronic equipment. In one electronic device embodiment according to the present invention, as shown in fig. 6, the electronic device includes a processor 601 and a storage 602, the storage may be configured to store a program for performing the visual lane line-based tracking method of the above-described method embodiment, and the processor may be configured to execute the program in the storage, including, but not limited to, the program for performing the visual lane line-based tracking method of the above-described method embodiment. For convenience of explanation, only those portions of the embodiments of the present invention that are relevant to the embodiments of the present invention are shown, and specific technical details are not disclosed, please refer to the method portions of the embodiments of the present invention.
Further, the invention also provides a computer readable storage medium. In one computer-readable storage medium embodiment according to the present invention, the computer-readable storage medium may be configured to store a program for performing the visual lane-based tracking method of the above-described method embodiment, which may be loaded and executed by a processor to implement the visual lane-based tracking method described above. For convenience of explanation, only those portions of the embodiments of the present invention that are relevant to the embodiments of the present invention are shown, and specific technical details are not disclosed, please refer to the method portions of the embodiments of the present invention. The computer readable storage medium may be a storage device including various electronic devices, and optionally, the computer readable storage medium in the embodiments of the present invention is a non-transitory computer readable storage medium.
Further, the invention also provides a vehicle, which comprises the electronic equipment.
Thus far, the technical solution of the present invention has been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of protection of the present invention is not limited to these specific embodiments. Equivalent modifications and substitutions for related technical features may be made by those skilled in the art without departing from the principles of the present invention, and such modifications and substitutions will fall within the scope of the present invention.
Claims (10)
1. A method of tracking based on visual lane lines, the method comprising:
obtaining a lane line detection result;
judging whether the lane line detection result meets a preset condition or not;
determining a measurement value based on the lane line detection result under the condition that the lane line detection result meets a preset condition;
creating a Kalman filter of a predicted track and initializing the Kalman filter;
matching the measured value with the predicted track to obtain a matching result;
updating the Kalman filter state of the predicted track according to the matching result;
and outputting a tracking result based on the updated Kalman filter state of the predicted trajectory.
2. The visual lane-based tracking method according to claim 1, wherein determining whether the lane-line detection result satisfies a preset condition comprises:
determining a lane width based on the lane line detection result;
judging whether the lane width is within a preset threshold range,
if yes, determining that the lane line detection result meets a preset condition; if not, determining that the lane line detection result does not meet the condition; and/or
Judging whether the lane line detection result meets a preset condition or not comprises the following steps:
determining the area between a first lane line curve and a second lane line curve corresponding to the lane line detection result, wherein the first lane line curve is a lane line curve corresponding to a high-precision map;
judging whether the area is larger than a first threshold value or not;
if yes, determining that the lane line detection result does not meet a preset condition; if not, determining that the lane line detection result meets the condition.
3. The visual lane-based tracking method of claim 1, wherein determining a measurement value based on the lane-line detection result comprises:
acquiring the ordinate of N anchor points from the lane line detection result, wherein N is more than or equal to 4;
determining the abscissa of the N anchor points based on the ordinate of the N anchor points;
the measurement value is determined based on the abscissa of the N anchor points.
4. The visual lane-based tracking method of claim 3, wherein the predicted track comprises a primary track and a backup track; matching the measured value with the predicted track to obtain a matching result, including:
matching the measurement value with the main track to obtain a first matching result, wherein the first matching result comprises any one of small-error matching, large-error matching and unmatched;
and under the condition that the first matching result is large-error matching, matching the measurement value with the standby track to obtain a second matching result.
5. The method of claim 4, wherein matching the measured value with the main track to obtain a first matching result comprises:
calculating the IOU distance between the rectangular frame of each anchor point in the measurement value and the rectangular frame of the anchor point in the main track and the center distance of the anchor point; wherein the method comprises the steps of
In the above formula, IOU (i, j) is IOU distance, area (R) i ) To measure the Area of the rectangular frame of any anchor point in the value, area (R j ) The Area of the rectangular frame of the anchor point in the main track is the Area of intersection of the rectangular frame of any anchor point in the measurement value and the rectangular frame of the anchor point in the main track;
Dcenter＝d/c
in the above formula, dcenter is the center distance of an anchor point, d is the center point distance of two rectangular frames, and c is the diagonal length of the minimum circumscribed rectangle of the two rectangular frames;
determining an IOU distance average value and an anchor point center distance average value based on the IOU distance and the anchor point center distance;
judging whether the IOU distance average value is smaller than or equal to a second threshold value;
if the IOU distance average value is smaller than or equal to a second threshold value, small error matching is performed, and if the IOU distance average value is larger than the second threshold value, whether the anchor point center distance average value is smaller than or equal to a third threshold value is judged;
if yes, large error matching is performed, and if not, no matching is performed.
6. The visual lane-based tracking method of claim 4, wherein updating the kalman filter state of the predicted trajectory according to the matching result comprises:
updating the Kalman filter state of the main track under the condition that the first matching result is small-error matching;
updating the Kalman filter state of the standby track under the condition that the first matching result is large error matching and the second matching result is small error matching;
and under the condition that the first matching result is unmatched, updating the Kalman filter state of the main track in a preset mode, and initializing the Kalman filter of the standby track in a measured value.
7. The visual lane-based tracking method according to claim 1, wherein outputting a tracking result based on the updated kalman filter state of the predicted trajectory comprises:
confirming the predicted track according to the Kalman filter state of the predicted track;
and outputting a tracking result based on the confirmed predicted track.
8. An electronic device comprising a processor and a storage means adapted to store a plurality of program code, characterized in that the program code is adapted to be loaded and executed by the processor to perform the visual lane line based tracking method of any one of claims 1 to 7.
9. A computer readable storage medium, in which a plurality of program codes are stored, characterized in that the program codes are adapted to be loaded and run by a processor to perform the visual lane line based tracking method of any one of claims 1 to 7.
10. A vehicle, characterized in that it comprises the electronic device of claim 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211698573.7A CN116259022A (en) | 2022-12-28 | 2022-12-28 | Tracking method based on visual lane line, electronic equipment, medium and vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211698573.7A CN116259022A (en) | 2022-12-28 | 2022-12-28 | Tracking method based on visual lane line, electronic equipment, medium and vehicle |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116259022A true CN116259022A (en) | 2023-06-13 |
Family
ID=86687174
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211698573.7A Pending CN116259022A (en) | 2022-12-28 | 2022-12-28 | Tracking method based on visual lane line, electronic equipment, medium and vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116259022A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117392634A (en) * | 2023-12-13 | 2024-01-12 | 上海闪马智能科技有限公司 | Lane line acquisition method and device, storage medium and electronic device |
-
2022
- 2022-12-28 CN CN202211698573.7A patent/CN116259022A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117392634A (en) * | 2023-12-13 | 2024-01-12 | 上海闪马智能科技有限公司 | Lane line acquisition method and device, storage medium and electronic device |
| CN117392634B (en) * | 2023-12-13 | 2024-02-27 | 上海闪马智能科技有限公司 | Lane line acquisition method and device, storage medium and electronic device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109522804B (en) | Road edge identification method and system | |
| CN110390306B (en) | Method for detecting right-angle parking space, vehicle and computer readable storage medium | |
| CN112880694B (en) | Method for determining the position of a vehicle | |
| US20200363809A1 (en) | Method and system for fusing occupancy maps | |
| CN114842450B (en) | Method, device and equipment for detecting drivable area | |
| CN116259022A (en) | Tracking method based on visual lane line, electronic equipment, medium and vehicle | |
| CN114943952A (en) | Method, system, device and medium for obstacle fusion under multi-camera overlapped view field | |
| CN113379776A (en) | Road boundary detection method | |
| CN112884771A (en) | Automatic detection method for power line foreign matter hanging based on laser LIDAR point cloud | |
| JPWO2007142267A1 (en) | Travel amount calculation system and obstacle detection system | |
| CN117289296A (en) | Mining area counter slope detection method and system based on laser radar | |
| CN116543271A (en) | Method, device, electronic equipment and medium for determining target detection evaluation index | |
| CN115902839A (en) | Port laser radar calibration method and device, storage medium and electronic equipment | |
| CN115359089A (en) | Point cloud target tracking method, electronic device, medium and vehicle | |
| CN116129378A (en) | Lane line detection method, device, equipment, vehicle and medium | |
| CN111857113B (en) | Positioning method and positioning device for movable equipment | |
| Wang et al. | Road edge detection based on improved RANSAC and 2D LIDAR Data | |
| CN117075135B (en) | Vehicle feature detection method, system, storage medium and electronic equipment | |
| WO2023097873A1 (en) | Method and apparatus for checking vehicle positioning, and storage medium and device | |
| CN118031967A (en) | Nuclear power station path navigation method, device, computer equipment and storage medium | |
| CN117611959A (en) | Multi-sensor perception fusion method, equipment and storage medium for driving system | |
| CN117657219A (en) | Automatic driving vehicle control method and device at crossroad | |
| Schexnaydre et al. | Using automated vehicle positioning to improve efficiency in vehicle platooning | |
| CN116107302A (en) | Method, device, robot and storage medium for getting rid of poverty | |
| CN115963822A (en) | Particle filter positioning-based slip processing method and device, medium and robot |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |