CN112782661A - Method and related device for detecting motor vehicle road guardrail - Google Patents
Method and related device for detecting motor vehicle road guardrail Download PDFInfo
- Publication number
- CN112782661A CN112782661A CN202110008043.4A CN202110008043A CN112782661A CN 112782661 A CN112782661 A CN 112782661A CN 202110008043 A CN202110008043 A CN 202110008043A CN 112782661 A CN112782661 A CN 112782661A
- Authority
- CN
- China
- Prior art keywords
- guardrail
- curve
- preset
- candidate detection
- initial
- 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
- 238000000034 method Methods 0.000 title claims abstract description 57
- 238000001514 detection method Methods 0.000 claims abstract description 208
- 238000012216 screening Methods 0.000 claims abstract description 24
- 239000013598 vector Substances 0.000 claims description 22
- 230000004888 barrier function Effects 0.000 claims description 4
- 230000008569 process Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000001174 ascending effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000008260 defense mechanism Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000452 restraining effect Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/48—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00
- G01S7/4802—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
Abstract
The application discloses a method for detecting a motor vehicle road guardrail and a related device, wherein the method comprises the following steps: calculating the relative distance between a detection point obtained by detecting the target guardrail and a preset reference curve, wherein the preset reference curve is obtained by translating according to a lane curve; screening detection points by adopting relative distances to obtain a candidate detection point set; and performing linear fitting on the guardrail according to the relative distance and the candidate detection point set based on a preset included angle judgment method to obtain a target guardrail curve. The guardrail detection accuracy that can alleviate current motor way is relatively poor, and the false detection easily appears, leads to self-adaptation cruise control system mistake to slow down, reduces user experience's technical problem.
Description
Technical Field
The application relates to the field of motor vehicles, in particular to a motor vehicle road guardrail detection method and a related device.
Background
The motor vehicle ownership and road traffic volume in China are increased rapidly, and the safety defense mechanism of the vehicle can ensure the safety of personnel in the vehicle after being triggered in the driving process of the motor vehicle. For example, motor vehicle cruise control systems, and in particular adaptive cruise control systems, have control units that act in concert with anti-lock braking systems and engine control systems to properly brake the wheels and reduce the output of the engine to maintain a safe distance from the vehicle to other vehicles at all times.
Because the guardrail detection scheme among the current self-adaptation cruise control system degree of accuracy is relatively poor, the misdetection easily appears, for example detect the guardrail as the vehicle, cause the mistake of system to slow down for user experience feels relatively poor.
Disclosure of Invention
The application provides a method and a related device for detecting a motor vehicle lane guardrail, which are used for solving the technical problems that the guardrail detection accuracy of the existing motor vehicle lane is poor, the false detection is easy to occur, the false deceleration of a self-adaptive cruise control system is caused, and the user experience is reduced.
In view of the above, a first aspect of the present application provides a method for detecting a guardrail of a motor vehicle roadway, including:
calculating the relative distance between a detection point obtained by detecting the target guardrail and a preset reference curve, wherein the preset reference curve is obtained by translating according to a lane curve;
screening the detection points by adopting the relative distance to obtain a candidate detection point set;
and performing linear fitting on the guardrail according to the relative distance and the candidate detection point set based on a preset included angle judgment method to obtain a target guardrail curve.
Preferably, the configuration process of the preset reference curve is as follows:
performing curve fitting on the acquired lane line by adopting a preset polynomial equation to obtain a lane curve, wherein the preset polynomial equation comprises a constant term;
and adjusting the size of the constant item according to a preset step length within a preset constant range to obtain a plurality of preset reference curves parallel to the lane curve.
Preferably, the screening the detection points by using the relative distance to obtain a candidate detection point set includes:
combining detection points corresponding to the relative distances smaller than a distance threshold value to obtain an initial candidate detection point set, wherein one preset reference curve corresponds to one initial candidate detection point set;
calculating an average distance residual error of each initial candidate detection point set according to the relative distance corresponding to the initial candidate detection points, wherein the initial candidate detection point set comprises a plurality of initial candidate detection points;
and selecting an initial candidate detection point set corresponding to the minimum average distance residual error as a candidate detection point set.
Preferably, the performing linear guardrail fitting according to the relative distance and the candidate detection point set based on a preset included angle determination method to obtain a target guardrail curve includes:
s1: selecting a first preset number of candidate detection points with the minimum relative distance from the candidate detection point set, and performing linear fitting on the first preset number of candidate detection points to obtain an initial guardrail curve;
s2: searching new candidate detection points at the extension end of the initial guardrail curve, and generating a fitting candidate vector by combining the extension end of the initial guardrail curve;
s3: and judging whether an included angle between the fitting candidate vector and the extension end tangent vector of the initial guardrail curve is smaller than a preset angle, if so, fitting the new candidate detection points into the initial guardrail curve in a preset fitting mode, repeating the steps S2-S3 until the candidate detection points on the initial guardrail curve reach a second preset number, and taking the initial guardrail curve as a target guardrail curve.
Preferably, step S3 further includes:
and judging whether the candidate detection points on the initial guardrail curve reach a third preset number, if not, the preset fitting mode is a first-order linear fitting mode, and if so, switching the preset fitting mode into a least square second-order polynomial fitting mode.
Preferably, the performing linear guardrail fitting according to the candidate detection point set and the candidate curve based on a preset included angle determination method to obtain a target guardrail curve, and then further includes:
and calculating a distance residual between a preset foreign matter point and the target guardrail curve, and if the distance residual is smaller than a preset distance value, judging that the preset foreign matter point is a non-vehicle detection point.
The application second aspect provides a motor vehicle way guardrail detection device, includes:
the calculation module is used for calculating the relative distance between a detection point obtained by detecting the target guardrail and a preset reference curve, and the preset reference curve is obtained by translating according to a lane curve;
the screening module is used for screening the detection points by adopting the relative distance to obtain a candidate detection point set;
and the curve fitting module is used for performing linear fitting on the guardrail according to the relative distance and the candidate detection point set based on a preset included angle judgment method to obtain a target guardrail curve.
Preferably, the screening module comprises:
the screening submodule is used for combining the detection points corresponding to the relative distances smaller than the distance threshold value to obtain an initial candidate detection point set, and one preset reference curve corresponds to one initial candidate detection point set;
a calculation submodule, configured to calculate an average distance residual of each initial candidate detection point set according to the relative distance corresponding to an initial candidate detection point, where the initial candidate detection point set includes a plurality of initial candidate detection points;
and the first selection submodule is used for selecting an initial candidate detection point set corresponding to the minimum average distance residual error as a candidate detection point set.
Preferably, the curve fitting module comprises:
the second selection submodule is used for selecting the first preset number of candidate detection points with the minimum relative distance from the candidate detection point set and carrying out linear fitting on the first preset number of candidate detection points to obtain an initial guardrail curve;
the searching submodule is used for searching new candidate detection points at the extension end of the initial guardrail curve and generating fitting candidate vectors by combining the extension end of the initial guardrail curve;
and the first judgment submodule is used for judging whether an included angle between the fitting candidate vector and the extension end tangent vector of the initial guardrail curve is smaller than a preset angle or not, if so, fitting the new candidate detection points into the initial guardrail curve in a preset fitting mode, and triggering the search submodule until the candidate detection points on the initial guardrail curve reach a second preset number, and then taking the initial guardrail curve as a target guardrail curve.
Preferably, the method further comprises the following steps:
and the second judgment submodule is used for judging whether the candidate detection points on the initial guardrail curve reach a third preset number, if not, the preset fitting mode is a linear fitting mode, and if so, the preset fitting mode is switched to a least square quadratic polynomial fitting mode.
From a third aspect, the present application provides a motor way barrier detection apparatus, the apparatus comprising a processor and a memory;
the memory is used for storing program codes and transmitting the program codes to the processor;
the processor is configured to execute the method of detecting a vehicle lane barrier of any of the first aspect according to instructions in the program code.
The present application provides, from a fourth aspect, a computer-readable storage medium for storing program code for executing the method for detecting a lane fence of a motor vehicle of any one of the first aspects.
According to the technical scheme, the embodiment of the application has the following advantages:
the application provides a method for detecting a motor vehicle road guardrail, which comprises the following steps: calculating the relative distance between a detection point obtained by detecting the target guardrail and a preset reference curve, wherein the preset reference curve is obtained by translating according to a lane curve; screening detection points by adopting relative distances to obtain a candidate detection point set; and performing linear fitting on the guardrail according to the relative distance and the candidate detection point set based on a preset included angle judgment method to obtain a target guardrail curve.
According to the method for detecting the motor vehicle lane guardrail, the guardrail on the lane is detected in a linear fitting mode, the assumption is that the curve formed by the lane guardrail and the lane line is parallel, and the advantage is that the continuous curve obtained through linear fitting meets the actual characteristic of the guardrail, namely the continuity; and moreover, some error detection points can be removed through screening of the detection points, and the reliability of the obtained target guardrail curve as an actual guardrail is higher. Therefore, the guardrail detection accuracy that present motorway can be alleviated to this application is relatively poor, and the false detection easily appears, leads to the self-adaptation cruise control system mistake to slow down, reduces user experience's technical problem.
Drawings
Fig. 1 is a schematic flow chart of a method for detecting a guardrail of a motor vehicle roadway according to an embodiment of the present application;
FIG. 2 is another schematic flow chart of a method for detecting a guardrail of a motor vehicle roadway according to an embodiment of the present application;
FIG. 3 is a schematic structural diagram of a device for detecting a guardrail of a motor vehicle road according to an embodiment of the present application;
FIG. 4 is a schematic diagram of relative distances provided by an embodiment of the present application;
FIG. 5 is a schematic diagram of a target guardrail curve search fit provided by an embodiment of the present application;
fig. 6 is a schematic distance diagram of a preset foreign object point and a target guardrail curve provided by the embodiment of the application.
Detailed Description
In order to make the technical solutions of the present application better understood, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
For easy understanding, referring to fig. 1, a first embodiment of a method for detecting a guardrail of a motor vehicle road provided by the present application includes:
The millimeter wave radar is adopted for detecting the target guardrail, and the millimeter wave radar can easily obtain a plurality of detection points of the guardrail beside the lane, particularly the metal guardrail. The acquisition number of the detection points can be set according to actual conditions, and a plurality of detection points in a preset range can be acquired to ensure accuracy. Besides the millimeter wave radar is adopted to detect the guardrail information, the laser radar can also be adopted to collect the point cloud data of the guardrail.
The preset reference curve is obtained according to the translation of the lane curve, and a priori knowledge or an assumption is applied to the preset reference curve, so that the trend of most lane guardrails is consistent with the trend of the corresponding side lane lines. The specific process is as follows: the lane line can be identified through a camera on the motor vehicle, and the lane line at the near end has higher precision; then, fitting the lane line by adopting a polynomial equation to obtain a lane curve; and finally, translating and traversing the constant term of the lane curve on the axis of the set coordinate system to obtain a plurality of different preset reference curves, wherein each preset reference curve is parallel to the lane curve.
Each preset reference curve can be used for calculating the relative distance, and the relative distance between each detection point and different preset reference curves is different, so that each detection point can obtain a batch of relative distance values according to the preset reference curves.
And 102, screening the detection points by adopting the relative distance to obtain a candidate detection point set.
All detection points are not accurate and are likely to have false detection or noise points, so that screening needs to be performed through relative distance, and the basic principle is that the closer the detection points are to a preset reference curve, the higher the possibility that the detection points are points on a guardrail is, the detection points need to be reserved to form a candidate detection point set; and directly rejecting detection points which are too far away.
And 103, performing linear fitting on the guardrail according to the relative distance and the candidate detection point set based on a preset included angle judgment method to obtain a target guardrail curve.
The preset included angle judgment method is used for restraining curve trend in a linear fitting process, the curve trend is not too far away from the actual complete trend of a guardrail, the larger the included angle is, the higher the possibility of deviation of fitting is, and therefore, the curve trend in the fitting process needs to be ensured to be within a certain included angle range.
The linear fitting is actually a process of connecting the candidate detection points according to a given linear fitting mode to obtain curve expression. The obtained continuous target guardrail curve is the continuous lane guardrail.
According to the method for detecting the motor vehicle lane guardrail, the guardrail on the lane is detected in a linear fitting mode, the assumption is that the curve formed by the lane guardrail and the lane line is parallel, and the advantage is that the continuous curve obtained through linear fitting meets the actual characteristic of the guardrail, namely the continuity; and moreover, some error detection points can be removed through screening of the detection points, and the reliability of the obtained target guardrail curve as an actual guardrail is higher. Therefore, the guardrail detection accuracy that can alleviate current motor way is relatively poor, and the misdetection easily appears leads to self-adaptation cruise control system mistake to slow down, reduces user experience's technical problem.
The above is an embodiment of a method for detecting a guardrail of a motor vehicle lane provided by the present application, and the following is another embodiment of the method for detecting a guardrail of a motor vehicle lane provided by the present application.
For easy understanding, please refer to fig. 2, the present application provides a second embodiment of a method for detecting a guardrail of a motor vehicle, comprising:
And 202, adjusting the size of the constant item according to a preset step length in a preset constant range to obtain a plurality of preset reference curves parallel to the lane curve.
In the embodiment of the application, the preset polynomial equation is set to be a cubic polynomial equation, and coefficients of a first order term, a second order term, a third order term and a constant term of the equation can respectively adopt C1、C2、C3And C0And (4) showing. The lane line is acquired by a camera.
In order to facilitate the selection of the preset constant range, a coordinate system can be set, the lane curve moves in parallel in the coordinate system, and the size of a constant item is adjusted, so that a plurality of preset reference lines are obtained; the operation in the coordinate system is simpler and more convenient, and the preset constant range is not easy to exceed, and the preset constant range is configured according to the actual situation, and is not limited herein.
And 203, calculating the relative distance between a detection point obtained by detecting the target guardrail and a preset reference curve, wherein the preset reference curve is obtained by translating according to a lane curve.
And 204, combining the detection points corresponding to the relative distances smaller than the distance threshold to obtain an initial candidate detection point set, wherein one preset reference curve corresponds to one initial candidate detection point set.
Referring to fig. 4, in the coordinate system, the relative distance between each detection point SGU and the preset reference curve is recorded as r, the detection points smaller than the distance threshold are retained and combined to obtain an initial candidate detection point set, and the distance threshold is set according to actual needs. The screening of the part is to directly screen from relative distance and eliminate points with larger phase difference. After the initial screening, the number of the detection points in the initial candidate detection point set corresponding to each preset reference curve is less than or equal to the number of the original detection points.
And step 206, selecting an initial candidate detection point set corresponding to the minimum average distance residual as a candidate detection point set.
Due to the fact that a plurality of initial candidate detection point sets exist, the candidate detection point set with the highest matching degree with the guardrail can be selected through screening of average distance residual errors. And each initial candidate detection point set can obtain an average distance residual error, the average distance residual errors corresponding to the initial candidate detection point sets corresponding to different preset reference curves are arranged in an ascending order, and the initial candidate detection point set corresponding to the first minimum average distance residual error is the point set with the highest guardrail matching degree, namely the candidate detection point set.
And step 207, selecting a first preset number of candidate detection points with the minimum relative distance from the candidate detection point set, and performing linear fitting on the first preset number of candidate detection points to obtain an initial guardrail curve.
Referring to fig. 5, in the embodiment of the present application, the first preset number is set to 4; the relative distances of all the candidate detection points in the candidate detection point set are arranged in an ascending order, the front 4 candidate detection points with the minimum relative distance can be obtained, linear fitting can be performed once according to the four points, an initial guardrail curve l is obtained, the initial guardrail curve obtained through preliminary fitting is a straight line, and the curve can be understood as a special curve.
And step 208, searching new candidate detection points at the extension end of the initial guardrail curve, and generating a fitting candidate vector by combining the extension end of the initial guardrail curve.
And 209, judging whether an included angle between the fitted candidate vector and the tangential vector of the extending end of the initial guardrail curve is smaller than a preset angle, if so, fitting new candidate detection points into the initial guardrail curve in a preset fitting mode, and repeating the step 208 and the step 209 until the candidate detection points on the initial guardrail curve reach a second preset number, and taking the initial guardrail curve as a target guardrail curve.
Point a in fig. 5 is an extension end of the initial guardrail curve and is also a starting point for starting to search for candidate detection points, point B and point C are new candidate detection points, point C is searched for first, and fitting candidate vectors can be generated by combining point a of the extension end of the initial guardrail curve
The tangent vector of the extending end of the initial guardrail curve is
(Vector)
And vector
The included angle between the two is represented by theta, if the included angle is smaller than the preset angle theta0And adding the point C into the curve, namely fitting the new candidate detection points into the guardrail curve in a preset fitting mode. If the included angle is not smaller than the preset angle, the point is discarded, and the search of a new candidate detection point is continued. The candidate detection points and the fitting curve are repeatedly searched, so that all points in the candidate detection points can be completely traversed, but in the actual operation process, complete traversal is not needed, and when the candidate detection points on the initial guardrail curve reach a second preset number, the initial guardrail curve obtained by fitting can be determined to be the target guardrail curve, namely the guardrail is detected.
The preset angle, although set according to the actual situation, is generally smaller, for example 3 °. The preset fitting mode comprises a simpler linear fitting mode and a more accurate quadratic polynomial fitting mode of a least square method, and can be selected according to the judgment requirement. The second preset number is set to be larger than the first preset number, and the number of the second preset number is set to be 10 in the embodiment of the application, that is, if 10 candidate detection points exist on the initial guardrail curve, the initial guardrail curve is considered as the target guardrail curve.
It will be appreciated that the target guardrail curve shape and course are both the same as the lane curve, but the constant terms describing the equations are different.
Further, step 209 further comprises:
and judging whether the candidate detection points on the initial guardrail curve reach a third preset number, if not, the preset fitting mode is a linear fitting mode, and if so, switching the preset fitting mode into a least square quadratic polynomial fitting mode.
The two listed preset fitting modes can be selected according to the number of the candidate detection points on the initial guardrail curve, the third preset number is set to be 7 according to the number of the candidate detection points, namely, linear fitting operation is carried out by adopting a linear fitting mode from the initial 4 candidate detection points to the 7 th candidate detection point; and when the 8 th candidate detection point is detected, if the angle condition is judged to be met, switching the preset fitting mode into a least square method quadratic polynomial fitting mode, and fitting the 8 th candidate detection point into the initial guardrail curve. The curve obtained by the least square quadratic polynomial fitting mode can be matched with the detection points as much as possible, so that the curve description is more accurate.
And step 210, calculating a distance residual between the preset foreign matter point and the target guardrail curve, and if the distance residual is smaller than a preset distance value, judging that the preset foreign matter point is a non-vehicle detection point.
The method is used for eliminating suspected vehicle points of the railing, namely the guardrail is mistakenly identified as a motor vehicle or a self-propelled vehicle in the adaptive cruise control system, so that the vehicle is mistakenly decelerated in the adaptive cruise process. The preset foreign object point is a target point which is detected by the camera and the millimeter wave radar together, the camera cannot acquire guardrail information under normal conditions, only vehicle and lane line information are acquired, but under the condition of false detection, guardrail information corresponding to the millimeter wave radar detection is detected, at the moment, the guardrail information is considered as vehicle information through fusion of the two kinds of information, and therefore trigger information with small speed is output to cause false deceleration. Referring to FIG. 6, the preset foreign object point is denoted as TGU, and the distance residual is denoted as dgt。
The setting of the preset distance value is typically small and may be 0.3 m. And if the preset foreign object point is correctly judged to be the non-vehicle detection point, the false deceleration caused by the false detection in the self-adaptive cruise control system can be avoided.
The judgment is carried out through the preset distance value, if the distance residual error value between the preset foreign matter point and the target guardrail curve is smaller than the preset distance value, the preset foreign matter point can be judged as a non-vehicle detection point,
the above is an embodiment of a method for detecting a guardrail of a motor vehicle road provided by the present application, and the following is an embodiment of a device for detecting a guardrail of a motor vehicle road provided by the present application.
For ease of understanding, referring to fig. 3, the present application also provides an embodiment of a railroad barrier detection device comprising:
the calculation module 301 is configured to calculate a relative distance between a detection point obtained by detecting the target guardrail and a preset reference curve, where the preset reference curve is obtained by translating according to a lane curve;
a screening module 302, configured to screen the detection points by using the relative distances to obtain a candidate detection point set;
and the curve fitting module 303 is configured to perform linear fitting on the guardrail according to the relative distance and the candidate detection point set based on a preset included angle determination method to obtain a target guardrail curve.
Further, the screening module 302 includes:
a screening submodule 3021, configured to combine detection points corresponding to relative distances smaller than a distance threshold to obtain an initial candidate detection point set, where one preset reference curve corresponds to one initial candidate detection point set;
a calculating submodule 3022, configured to calculate an average distance residual of each initial candidate detection point set according to a relative distance corresponding to the initial candidate detection point, where the initial candidate detection point set includes a plurality of initial candidate detection points;
the first selecting submodule 3023 is configured to select an initial candidate detection point set corresponding to the minimum average distance residual as a candidate detection point set.
Further, the curve fitting module 303 includes:
the second selecting submodule 3031 is configured to select a first preset number of candidate detection points with a smallest relative distance from the candidate detection point set, and perform linear fitting on the first preset number of candidate detection points to obtain an initial guardrail curve;
the searching submodule 3032 is used for searching new candidate detection points at the extension end of the initial guardrail curve and generating a fitting candidate vector by combining the extension end of the initial guardrail curve;
and the first judging submodule 3033 is configured to judge whether an included angle between the fitted candidate vector and the tangential vector of the extending end of the initial guardrail curve is smaller than a preset angle, if so, fit new candidate detection points into the initial guardrail curve in a preset fitting manner, and trigger the searching submodule until the candidate detection points on the initial guardrail curve reach a second preset number, so that the initial guardrail curve is used as a target guardrail curve.
Further, still include:
and the second judging submodule 304 is configured to judge whether the number of candidate detection points on the initial guardrail curve reaches a third preset number, if not, the preset fitting mode is a linear fitting mode, and if so, the preset fitting mode is switched to a least square quadratic polynomial fitting mode.
The application also provides a device for detecting the motor way guardrail, which comprises a processor and a memory;
the memory is used for storing program codes and transmitting the program codes to the processor;
the processor is configured to execute the method for detecting a lane guard of a vehicle in the above-described method embodiment according to instructions in the program code.
The present application also provides a computer-readable storage medium for storing program code for executing the method of detecting a lane fence in the above-described method embodiments.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be substantially implemented or contributed to by the prior art, or all or part of the technical solution may be embodied in a software product, which is stored in a storage medium and includes instructions for executing all or part of the steps of the method described in the embodiments of the present application through a computer device (which may be a personal computer, a server, or a network device). And the aforementioned storage medium includes: a U disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.
Claims (12)
1. A method for detecting a motor vehicle guardrail is characterized by comprising the following steps:
calculating the relative distance between a detection point obtained by detecting the target guardrail and a preset reference curve, wherein the preset reference curve is obtained by translating according to a lane curve;
screening the detection points by adopting the relative distance to obtain a candidate detection point set;
and performing linear fitting on the guardrail according to the relative distance and the candidate detection point set based on a preset included angle judgment method to obtain a target guardrail curve.
2. The method of detecting a lane fence according to claim 1, wherein the preset reference curve is configured by:
performing curve fitting on the acquired lane line by adopting a preset polynomial equation to obtain a lane curve, wherein the preset polynomial equation comprises a constant term;
and adjusting the size of the constant item according to a preset step length within a preset constant range to obtain a plurality of preset reference curves parallel to the lane curve.
3. The method for detecting a motor vehicle lane guardrail according to claim 1, wherein the screening the detection points by using the relative distance to obtain a candidate detection point set comprises:
combining detection points corresponding to the relative distances smaller than a distance threshold value to obtain an initial candidate detection point set, wherein one preset reference curve corresponds to one initial candidate detection point set;
calculating an average distance residual error of each initial candidate detection point set according to the relative distance corresponding to the initial candidate detection points, wherein the initial candidate detection point set comprises a plurality of initial candidate detection points;
and selecting an initial candidate detection point set corresponding to the minimum average distance residual error as a candidate detection point set.
4. The method for detecting the guardrail of the motor way according to claim 1, wherein the step of performing linear guardrail fitting according to the relative distance and the candidate detection point set based on a preset included angle judgment method to obtain a target guardrail curve comprises the following steps:
s1: selecting a first preset number of candidate detection points with the minimum relative distance from the candidate detection point set, and performing linear fitting on the first preset number of candidate detection points to obtain an initial guardrail curve;
s2: searching new candidate detection points at the extension end of the initial guardrail curve, and generating a fitting candidate vector by combining the extension end of the initial guardrail curve;
s3: and judging whether an included angle between the fitting candidate vector and the extension end tangent vector of the initial guardrail curve is smaller than a preset angle, if so, fitting the new candidate detection points into the initial guardrail curve in a preset fitting mode, repeating the steps S2-S3 until the candidate detection points on the initial guardrail curve reach a second preset number, and taking the initial guardrail curve as a target guardrail curve.
5. The method for detecting a lane fence according to claim 4, wherein step S3 further comprises:
and judging whether the candidate detection points on the initial guardrail curve reach a third preset number, if not, the preset fitting mode is a first-order linear fitting mode, and if so, switching the preset fitting mode into a least square second-order polynomial fitting mode.
6. The method for detecting the guardrail of the motor vehicle lane according to claim 1, wherein the linear fitting of the guardrail is performed according to the candidate detection point set and the candidate curve based on a preset included angle determination method to obtain a target guardrail curve, and then further comprising:
and calculating a distance residual between a preset foreign matter point and the target guardrail curve, and if the distance residual is smaller than a preset distance value, judging that the preset foreign matter point is a non-vehicle detection point.
7. A motor vehicle guardrail detection device, comprising:
the calculation module is used for calculating the relative distance between a detection point obtained by detecting the target guardrail and a preset reference curve, and the preset reference curve is obtained by translating according to a lane curve;
the screening module is used for screening the detection points by adopting the relative distance to obtain a candidate detection point set;
and the curve fitting module is used for performing linear fitting on the guardrail according to the relative distance and the candidate detection point set based on a preset included angle judgment method to obtain a target guardrail curve.
8. The lane guard detection apparatus of claim 7, wherein the screening module comprises:
the screening submodule is used for combining the detection points corresponding to the relative distances smaller than the distance threshold value to obtain an initial candidate detection point set, and one preset reference curve corresponds to one initial candidate detection point set;
a calculation submodule, configured to calculate an average distance residual of each initial candidate detection point set according to the relative distance corresponding to an initial candidate detection point, where the initial candidate detection point set includes a plurality of initial candidate detection points;
and the first selection submodule is used for selecting an initial candidate detection point set corresponding to the minimum average distance residual error as a candidate detection point set.
9. The lane guard detection apparatus of claim 7, wherein the curve fitting module comprises:
the second selection submodule is used for selecting the first preset number of candidate detection points with the minimum relative distance from the candidate detection point set and carrying out linear fitting on the first preset number of candidate detection points to obtain an initial guardrail curve;
the searching submodule is used for searching new candidate detection points at the extension end of the initial guardrail curve and generating fitting candidate vectors by combining the extension end of the initial guardrail curve;
and the first judgment submodule is used for judging whether an included angle between the fitting candidate vector and the extension end tangent vector of the initial guardrail curve is smaller than a preset angle or not, if so, fitting the new candidate detection points into the initial guardrail curve in a preset fitting mode, and triggering the search submodule until the candidate detection points on the initial guardrail curve reach a second preset number, and then taking the initial guardrail curve as a target guardrail curve.
10. The lane guard detection device of claim 9, further comprising:
and the second judgment submodule is used for judging whether the candidate detection points on the initial guardrail curve reach a third preset number, if not, the preset fitting mode is a linear fitting mode, and if so, the preset fitting mode is switched to a least square quadratic polynomial fitting mode.
11. An apparatus for detecting a motor vehicle guardrail, the apparatus comprising a processor and a memory;
the memory is used for storing program codes and transmitting the program codes to the processor;
the processor is configured to execute the method of detecting a vehicle lane barrier of any one of claims 1-6 according to instructions in the program code.
12. A computer-readable storage medium characterized in that the computer-readable storage medium stores program code for executing the method of detecting a lane fence according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110008043.4A CN112782661A (en) | 2021-01-05 | 2021-01-05 | Method and related device for detecting motor vehicle road guardrail |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110008043.4A CN112782661A (en) | 2021-01-05 | 2021-01-05 | Method and related device for detecting motor vehicle road guardrail |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112782661A true CN112782661A (en) | 2021-05-11 |
Family
ID=75755452
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110008043.4A Pending CN112782661A (en) | 2021-01-05 | 2021-01-05 | Method and related device for detecting motor vehicle road guardrail |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112782661A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050228580A1 (en) * | 2002-04-27 | 2005-10-13 | Hermann Winner | Method and device for predicting the course of motor vehicles |
| CN106096525A (en) * | 2016-06-06 | 2016-11-09 | 重庆邮电大学 | A kind of compound lane recognition system and method |
| CN106407893A (en) * | 2016-08-29 | 2017-02-15 | 东软集团股份有限公司 | Method, device and equipment for detecting lane line |
| CN109254289A (en) * | 2018-11-01 | 2019-01-22 | 百度在线网络技术（北京）有限公司 | The detection method and detection device of road guard |
| CN110174113A (en) * | 2019-04-28 | 2019-08-27 | 福瑞泰克智能系统有限公司 | A kind of localization method, device and the terminal in vehicle driving lane |
| US20200307560A1 (en) * | 2019-03-27 | 2020-10-01 | Mando Corporation | Apparatus and method for controlling vehicle |
| KR20200141871A (en) * | 2019-06-11 | 2020-12-21 | 에스케이텔레콤 주식회사 | Apparatus and method for obtaining lane information |
| CN112161628A (en) * | 2020-09-27 | 2021-01-01 | 广州小鹏自动驾驶科技有限公司 | Path data processing method and device, vehicle and readable medium |
-
2021
- 2021-01-05 CN CN202110008043.4A patent/CN112782661A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050228580A1 (en) * | 2002-04-27 | 2005-10-13 | Hermann Winner | Method and device for predicting the course of motor vehicles |
| CN106096525A (en) * | 2016-06-06 | 2016-11-09 | 重庆邮电大学 | A kind of compound lane recognition system and method |
| CN106407893A (en) * | 2016-08-29 | 2017-02-15 | 东软集团股份有限公司 | Method, device and equipment for detecting lane line |
| CN109254289A (en) * | 2018-11-01 | 2019-01-22 | 百度在线网络技术（北京）有限公司 | The detection method and detection device of road guard |
| US20200307560A1 (en) * | 2019-03-27 | 2020-10-01 | Mando Corporation | Apparatus and method for controlling vehicle |
| CN110174113A (en) * | 2019-04-28 | 2019-08-27 | 福瑞泰克智能系统有限公司 | A kind of localization method, device and the terminal in vehicle driving lane |
| KR20200141871A (en) * | 2019-06-11 | 2020-12-21 | 에스케이텔레콤 주식회사 | Apparatus and method for obtaining lane information |
| CN112161628A (en) * | 2020-09-27 | 2021-01-01 | 广州小鹏自动驾驶科技有限公司 | Path data processing method and device, vehicle and readable medium |
Non-Patent Citations (2)
| Title |
|---|
| 沈海栋;孔明;丁超;王英军;赵军;: "汽车防护栏刚度检测系统", 测控技术, no. 10, 18 October 2016 (2016-10-18) * |
| 王永忠;王晓云;文成林;: "梯度点对约束的结构化车道检测", 中国图象图形学报, no. 06, 16 June 2012 (2012-06-16) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10650271B2 (en) | Image processing apparatus, imaging device, moving object device control system, and image processing method | |
| US20170358102A1 (en) | Object recognition processing apparatus, object recognition processing method, and autonomous driving system | |
| WO2016129403A1 (en) | Object detection device | |
| CN110632921A (en) | Robot path planning method and device, electronic equipment and storage medium | |
| CN110632617B (en) | Laser radar point cloud data processing method and device | |
| WO2016189112A1 (en) | Collision avoidance method, computer program product for said collision avoidance method and collision avoidance system | |
| JP5178276B2 (en) | Image recognition device | |
| CN109720348B (en) | In-vehicle device, information processing system, and information processing method | |
| KR101857554B1 (en) | External data intrusion detection apparatus and method for vehicles | |
| CN111016919A (en) | Device and method for identifying a close cut into a vehicle and vehicle comprising such a device | |
| CN108859952B (en) | Vehicle lane change early warning method and device and radar | |
| CN104773177A (en) | Aided driving method and aided driving device | |
| KR20200113915A (en) | Apparatus and method for controlling vehicle | |
| KR20200040559A (en) | Apparatus and method for predicting concurrent lane change vehicle and vehicle including the same | |
| KR102323692B1 (en) | Method and apparatus for evaluating driver using adas | |
| KR20180109497A (en) | Ecu, autonomous vehicle including the ecu, and method of recognizing near vehicle for the same | |
| CN109195849B (en) | Image pickup apparatus | |
| CN112782661A (en) | Method and related device for detecting motor vehicle road guardrail | |
| KR102337034B1 (en) | Autonomous driving situation recognition program performance test method and apparatus for porceeding the same | |
| CN111301407B (en) | Dangerous vehicle determination method and device, electronic equipment and readable storage medium | |
| CN110618420A (en) | Ultrasonic data processing method and system, vehicle and storage medium | |
| KR102085910B1 (en) | Apparatus and method for discrimination of stationary target and moving target | |
| JP2001082954A (en) | Image processing device and image processing distance- measuring method | |
| JP3740531B2 (en) | Parked vehicle detection method, detection system, and parked vehicle detection device | |
| CN110239533B (en) | Method, device, apparatus and readable storage medium for determining driving risk of vehicle |
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 |