CN110044333A - A kind of slope detection method and device based on monocular vision - Google Patents
A kind of slope detection method and device based on monocular vision Download PDFInfo
- Publication number
- CN110044333A CN110044333A CN201910398135.0A CN201910398135A CN110044333A CN 110044333 A CN110044333 A CN 110044333A CN 201910398135 A CN201910398135 A CN 201910398135A CN 110044333 A CN110044333 A CN 110044333A
- Authority
- CN
- China
- Prior art keywords
- lane line
- end point
- value
- line end
- vertical distance
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 30
- 238000000034 method Methods 0.000 claims abstract description 18
- 230000003287 optical effect Effects 0.000 claims description 10
- 238000002955 isolation Methods 0.000 description 8
- 238000003384 imaging method Methods 0.000 description 4
- 238000005259 measurement Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000006467 substitution reaction Methods 0.000 description 3
- 230000001154 acute effect Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C11/00—Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
- G01C11/04—Interpretation of pictures
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C11/00—Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
- G01C11/36—Videogrammetry, i.e. electronic processing of video signals from a single source or from different sources to give parallax or range information
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C9/00—Measuring inclination, e.g. by clinometers, by levels
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Signal Processing (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
Abstract
The present invention provides a kind of slope detection method and device based on monocular vision, belong to technical field of vehicle safety, method includes: the current focus for the monocular vision camera that acquisition is arranged on the downside of headstock and the picture frame including front lane line that monocular vision camera collects, and the central point in picture frame using picture frame establishes plane right-angle coordinate as origin.It determines the lane line end point in picture frame, and determines lane line end point coordinate corresponding with lane line end point in plane right-angle coordinate, using the absolute value of the ordinate of lane line end point coordinate as current vertical distance.It regard current focus and current vertical distance as input quantity, it substitutes into the pre-stored functional relation for expressing value of slope using vertical distance and focal length value, the current hill grade value exported reduces overall cost to realize high-precision slope detection based on monocular vision camera.
Description
Technical field
The present invention relates to technical field of vehicle safety, in particular to a kind of slope detection method and dress based on monocular vision
It sets.
Background technique
For the vehicle in driving status, learn that the gradient size on vehicle front road surface helps to adjust in time in time
Whole gas pedal depth and the more suitable gear of selection are climbed, and are enabled the vehicle to more smoothly by ramp, simultaneously
The spacing between correction vehicle and preceding object vehicle is also contributed to, safety when vehicle driving is improved.
It is in the prior art typically all using in such a way that front part of vehicle installs multiple radars additional, to be gone out using detections of radar
The gradient size on vehicle front road surface.
In the implementation of the present invention, the inventor finds that the existing technology has at least the following problems:
It is existing by radar come detect vehicle front gradient size in the way of, since radar cost is higher, and radar survey
It is complex to measure gradient process, causes a whole set of gradient measuring device higher cost, is difficult to popularize.
Summary of the invention
In view of this, the present invention provides a kind of slope detection method and device based on monocular vision, it can be with low cost
Realize slope detection.
Specifically, including technical solution below:
On the one hand, the present invention provides a kind of slope detection method based on monocular vision, comprising:
The current focus and monocular vision camera for obtaining the monocular vision camera being arranged on the downside of headstock acquire
The picture frame including front lane line arrived, and flat square seat is established using the central point of picture frame as origin in picture frame
Mark system.
Determine the lane line end point in picture frame, and determination is corresponding with lane line end point in plane right-angle coordinate
Lane line end point coordinate, using the absolute value of the ordinate of lane line end point coordinate as current vertical distance.
It regard current focus and current vertical distance as input quantity, substitution is pre-stored to utilize vertical distance and focal length
In functional relation of the value to express value of slope, the current hill grade value that is exported.
Selectively, the direction of travel keeping parallelism of the optical axis of monocular vision camera and vehicle.
Selectively, the lane line end point in picture frame is determined, comprising:
Two front lane lines are selected in picture frame.
The intersection point of the extended line of two front lane lines is determined as lane line end point.
Selectively, method further include:
After determining multiple lane line end points in picture frame, determining and multiple lane lines in plane right-angle coordinate
The corresponding multiple lane line end point coordinates of end point.
It is substituted into using the average value of the absolute value of the ordinate of multiple lane line end point coordinates as current vertical distance pre-
In corresponding relationship between the vertical distance first stored and value of slope.
Selectively, vertical distance is indicated with L, and the focal length value of monocular vision camera indicates that value of slope is indicated with α with f,
The functional relation of value of slope is expressed using vertical distance and focal length value are as follows:
On the other hand, the present invention provides a kind of slope detection device based on monocular vision, comprising:
Image collection module, be configured as obtaining the current focus that monocular vision camera on the downside of headstock is set and
The picture frame including front lane line that monocular vision camera collects, and made in picture frame with the central point of picture frame
Plane right-angle coordinate is established for origin.
Coordinate determining module, the lane line end point being configured to determine that in picture frame, and in plane right-angle coordinate
It determines lane line end point coordinate corresponding with lane line end point, the absolute value of the ordinate of lane line end point coordinate is made
For current vertical distance.
Gradient computing module is configured as current focus and current vertical distance as input quantity, and substitution is deposited in advance
In the functional relation for expressing value of slope using vertical distance and focal length value of storage, the current hill grade value that is exported.
Selectively, the direction of travel keeping parallelism of the optical axis of monocular vision camera and vehicle.
Selectively, coordinate determining module is also configured to
Two front lane lines are selected in picture frame.
The intersection point of the extended line of two front lane lines is determined as lane line end point.
Selectively, coordinate determining module is also configured to
After determining multiple lane line end points in picture frame, determining and multiple lane lines in plane right-angle coordinate
The corresponding multiple lane line end point coordinates of end point.
Gradient computing module is also configured to
It is substituted into using the average value of the absolute value of the ordinate of multiple lane line end point coordinates as current vertical distance pre-
In corresponding relationship between the vertical distance first stored and value of slope.
Selectively, vertical distance is indicated with L, and the focal length of monocular vision camera indicates that value of slope is indicated with α with f, benefit
The functional relation of value of slope is expressed with vertical distance and focal length value are as follows:
The beneficial effect of technical solution provided in an embodiment of the present invention includes at least:
Slope detection method and device provided by the invention based on monocular vision, method include: to obtain to be arranged in headstock
The figure including front lane line that the current focus and monocular vision camera of the monocular vision camera of downside collect
As frame, and the central point in picture frame using picture frame establishes plane right-angle coordinate as origin, to analyze road surface ahead
Situation.Determine the lane line end point in picture frame, and determination is corresponding with lane line end point in plane right-angle coordinate
Lane line end point coordinate, using the absolute value of the ordinate of lane line end point coordinate as current vertical distance, to be vehicle
Road end point assigns numerical value meaning.It regard current focus and current vertical distance as input quantity, substitutes into pre-stored utilization
In the functional relation of vertical distance and focal length value to express value of slope, the current hill grade value exported, to be regarded based on monocular
Feel that camera realizes high-precision slope detection, reduces overall cost.
Detailed description of the invention
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment
Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for
For those of ordinary skill in the art, without creative efforts, it can also be obtained according to these attached drawings other
Attached drawing.
Attached drawing 1 is the flow chart for the slope detection method based on monocular vision that the embodiment of the present invention one provides；
Attached drawing 2 is the geometrical relationship schematic diagram of the slope detection method provided in an embodiment of the present invention based on monocular vision；
Attached drawing 3-4 is the lane in the picture frame of the slope detection method provided in an embodiment of the present invention based on monocular vision
End point schematic diagram；
The block diagram of the slope detection device provided by Embodiment 2 of the present invention based on monocular vision of attached drawing 5.
Specific embodiment
To keep technical solution of the present invention and advantage clearer, below in conjunction with attached drawing to embodiment of the present invention make into
One step it is described in detail.
Embodiment one
Present embodiments provide a kind of slope detection method based on monocular vision, as shown in Figure 1, include step S101,
S102 and S103 below will be specifically introduced each step.
Technical solution provided by the present embodiment needs to pre-establish between focal length value, vertical distance and value of slope three
Functional relation.Focal length value refers to the focal length of monocular vision camera, and vertical distance refers to that monocular vision camera collects
Picture frame in certain point to picture frame central point fore-and-aft distance.
Vertical distance indicates that the focal length value of monocular vision camera indicates that value of slope is indicated with α with f with L, using vertical
Distance and focal length value express the functional relation of value of slope are as follows:
The process for obtaining above formula is as follows:
Firstly, first analyzing the geometrical relationship of slope detection:
In the present embodiment, the direction of travel keeping parallelism of the optical axis of monocular vision camera and vehicle.As shown in Fig. 2,
According to image-forming principle it is found that on slope pavement infinite point P point after the center C of monocular vision camera as in plane at
Seem P ' point.Here picture plane can be photosensitive screen, and the P " point with ground intersection, figure are extended to after incident ray is extended
T is the point of interface of level road and slope pavement in 2.O point is the central point for the picture frame that monocular vision camera collects,
F is the focal length of monocular vision camera, i.e. camera center to the distance as plane, and α is the value of slope of slope pavement, i.e. slope
Folded acute angle between road surface and level road.
In triangle PTP ', it can be obtained according to triangle edges angular dependence:
It is available after above formula is converted:
Because P point is the point of infinite point on slope pavement, lane line end point hereinafter can regard slope as
As imaging in plane after monocular vision camera, this point hereinafter will do it infinite point P point on road surface
Specific explanations, so, the expression formula of sin (∠ P) can be approximated to be:
Again because the value of SIN function sin (∠ PP " T) is between -1 to 1, TP " limited length can be in the hope of: sin (∠
P)≈0
And then obtain: ∠ P=0
The size of the value of slope α of slope pavement known to ∠ P=0 and triangle exterior angle theorem are as follows: α=∠ P+ ∠ PP " T
=∠ PP " T
It is since the optical axis of monocular vision camera is parallel with vehicle forward direction, i.e., parallel to the ground, therefore: ∠ PP " T=
∠oCP'
In triangle OCP, ∠ oCP'=arctan (oP'/f)
So that α=arctan (oP'/f)
Wherein, oP ' is exactly P ' vertical distance L, therefore available:
After establishing and storing the functional relation between focal length value, vertical distance and value of slope three, so that it may utilize
Monocular vision camera carrys out acquired image frames.
Analyzed by above-mentioned geometrical relationship it is found that when monocular vision camera is closer to slope pavement, line PC just closer to
For the angle value of line PT, ∠ PP " T just closer to α, measurement accuracy is higher.Therefore in the present embodiment, monocular vision camera
The precision of slope detection can be improved in ground proximity as far as possible, is arranged as a kind of alternative embodiment, monocular vision camera
Height from the ground can be between 15cm-30cm.As a kind of alternative embodiment, monocular vision camera be can be set in vehicle
The center downside of front bumper.
Vehicle is during the entire process of driving to slope pavement from level road, when vehicle front-wheel is just pressed onto point of interface T
When point, the headstock of vehicle is nearest from slope pavement, and the monocular vision camera in vehicle front bumper is arranged at this time from slope
Also recently, measurement accuracy highest.When vehicle front-wheel is just pressed onto point of interface T point, pressure that vehicle front-wheel is applied by road surface
Value has one and increases suddenly, and specific pressure value increase can show as tire pressure and increase suddenly, therefore pass through vehicle front-wheel
Tire pressure variation may determine that whether the current front-wheel of vehicle is just pressed onto point of interface T point, and using this moment as detection moment.
As a kind of alternative embodiment, before executing step S101, this method can also include: to obtain tyre pressure sensor to acquire in real time
Obtained tire pressure value executes step S101 when tire pressure value increases default threshold pressure within a preset time interval.
In step s101, the current focus and monocular vision of the monocular vision camera being arranged on the downside of headstock are obtained
The picture frame including front lane line that camera collects, and the central point in picture frame using picture frame is built as origin
Vertical plane right-angle coordinate.
Picture frame is obtained by the photosensitive screen in monocular cam.
In step s 102, the lane line end point in picture frame is determined, and the determining and vehicle in plane right-angle coordinate
The corresponding lane line end point coordinate of diatom end point, using the absolute value of the ordinate of lane line end point coordinate as current perpendicular
Straight distance.
As a kind of alternative embodiment, the lane line end point in picture frame is determined, comprising:
Two front lane lines are selected in picture frame.It is understood that since the lane on road surface may be more than
Two, therefore the front lane line in picture frame can be more than two.
The intersection point of the extended line of two front lane lines is determined as lane line end point.It is understood that picture frame
In include front lane line extended line, can by picture frame front lane line carry out feature point extraction, connection
It ties multiple characteristic points and extends, to obtain the extended line of front lane line, the image for including front lane line extended line can be with
The virtual image new as one is stored.
Due to foring certain angle between road surface and the optical axis and road surface of monocular vision camera, monocular vision is taken the photograph
The two front lane lines collected as head are not parallel to each other completely, but are mutually formed with certain angle before
, and the extended line of last two fronts lane line can intersect at a point, which is known as lane line end point.Lane heading line off
Point is not necessary being in picture frame, received by photosensitive screen a bit, but one obtained after image procossing
A virtual point.Lane line end point can be regarded as the infinite point on slope pavement after monocular vision camera as flat
Imaging on face.
As shown in figure 3, the vertical distance of lane line end point is smaller, illustrate that the value of slope of slope pavement is smaller.Such as Fig. 4 institute
Show, the vertical distance of lane line end point is bigger, illustrates that the value of slope of slope pavement is bigger.
Simultaneously as lane line is decorated in the position on some road surfaces close to roadside, but center is to replace vehicle using isolation strip
Diatom, therefore, as a kind of alternative embodiment, this method can also include: the selection isolation strip lower edge line in picture frame
And the front lane line adjacent with isolation strip lower edge line.By the extended line of isolation strip lower edge line and adjacent front vehicle
The intersection point of the extended line of diatom is determined as lane line end point.
Ideally, multiple in picture frame if all lane lines on road surface are all parallel to each other
The extended line of front lane line finally can all intersect at same point, therefore only one lane in lower picture frame of ideal situation
Heading line off point.
But due to that may have common runway, fast, Emergency Vehicle Lane, ring road, turning guided vehicle road on road surface
All lane lines etc. a variety of positions, the lane of a variety of curvature, therefore on road surface may not be to be parallel to each other completely, at this time root
It is possible to can determine whether out more than one lane line end point according to multiple front lane lines in picture frame.
As a kind of alternative embodiment, step S102 further include:
After determining multiple lane line end points in picture frame, determining and multiple lane lines in plane right-angle coordinate
The corresponding multiple lane line end point coordinates of end point.
In step s 103, current focus and current vertical distance are regard as input quantity, substitutes into pre-stored utilization
In the functional relation of vertical distance and focal length value to express value of slope, the current hill grade value that is exported.
When there are multiple lane end point coordinates, as a kind of alternative embodiment, step S103 further include:
It is substituted into using the average value of the absolute value of the ordinate of multiple lane line end point coordinates as current vertical distance pre-
In corresponding relationship between the vertical distance first stored and value of slope.
The beneficial effect of technical solution provided in an embodiment of the present invention includes at least:
Slope detection method provided by the invention based on monocular vision is regarded by obtaining the monocular being arranged on the downside of headstock
Feel the current focus of camera and the picture frame including front lane line that monocular vision camera collects, and in image
Central point in frame using picture frame establishes plane right-angle coordinate as origin, to analyze road surface ahead situation.Determine image
Lane line end point in frame, and determine that lane line end point corresponding with lane line end point is sat in plane right-angle coordinate
Mark, using the absolute value of the ordinate of lane line end point coordinate as current vertical distance, to assign number for lane end point
It is worth meaning.It regard current focus and current vertical distance as input quantity, substitution is pre-stored to utilize vertical distance and focal length
In functional relation of the value to express value of slope, the current hill grade value exported, to be realized based on monocular vision camera high
Precision slope detection, reduces overall cost.
All the above alternatives can form alternative embodiment of the invention using any combination, herein no longer
It repeats one by one.
Embodiment two
A kind of slope detection device based on monocular vision is present embodiments provided, as shown in figure 5, including that image obtains mould
Block 501, coordinate determining module 502 and gradient computing module 503 below will be specifically introduced each module.
Technical solution provided by the present embodiment needs to pre-establish three between focal length value, vertical distance and value of slope
Functional relation.Focal length value refers to the focal length of monocular vision camera, and vertical distance refers to that monocular vision camera collects
Picture frame in certain point to picture frame central point fore-and-aft distance.
Vertical distance indicates that the focal length value of monocular vision camera indicates that value of slope is indicated with α with f with L, using vertical
Distance and focal length value express the functional relation of value of slope are as follows:
The process for obtaining above formula is as follows:
Firstly, first analyzing the geometrical relationship of slope detection:
In the present embodiment, the direction of travel keeping parallelism of the optical axis of monocular vision camera and vehicle.As shown in Fig. 2,
According to image-forming principle it is found that on slope pavement infinite point P point after the center C of monocular vision camera as in plane at
Seem P ' point.Here picture plane can be photosensitive screen, and the P " point with ground intersection, figure are extended to after incident ray is extended
T is the point of interface of level road and slope pavement in 2.O point is the central point for the picture frame that monocular vision camera collects,
F is the focal length of monocular vision camera, i.e. camera center to the distance as plane, and α is the value of slope of slope pavement, i.e. slope
Folded acute angle between road surface and level road.
In triangle PTP ', it can be obtained according to triangle edges angular dependence:
It is available after above formula is converted:
Because P point is the point of infinite point on slope pavement, lane line end point hereinafter can regard slope as
As imaging in plane after monocular vision camera, this point hereinafter will do it infinite point P point on road surface
Specific explanations, so, the expression formula of sin (∠ P) can be approximated to be:
Again because the value of SIN function sin (∠ PP " T) is between -1 to 1, TP " limited length can be in the hope of: sin (∠
P)≈0
And then obtain: ∠ P=0
The size of the value of slope α of slope pavement known to ∠ P=0 and triangle exterior angle theorem are as follows: α=∠ P+ ∠ PP " T
=∠ PP " T
It is since the optical axis of monocular vision camera is parallel with vehicle forward direction, i.e., parallel to the ground, therefore: ∠ PP " T=
∠oCP'
In triangle OC, ∠ oCP'=arctan (oP'/f)
So that α=arctan (oP'/f)
Wherein, oP ' is exactly P ' vertical distance L, therefore available:
It is establishing and is storing between focal length value, vertical distance and value of slope after the functional relation of three, so that it may utilize
Monocular vision camera carrys out acquired image frames.
Analyzed by above-mentioned geometrical relationship it is found that when monocular vision camera is closer to slope pavement, line PC just closer to
For the angle value of line PT, ∠ PP " T just closer to α, measurement accuracy is higher.Therefore in the present embodiment, monocular vision camera
The precision of slope detection can be improved in ground proximity as far as possible, is arranged as a kind of alternative embodiment, monocular vision camera
Height from the ground can be between 15cm-30cm.As a kind of alternative embodiment, monocular vision camera be can be set in vehicle
The center downside of front bumper.
Image collection module 501 is configured as obtaining the current focus for the monocular vision camera being arranged on the downside of headstock
And the picture frame including front lane line that monocular vision camera collects, and with the center of picture frame in picture frame
Point establishes plane right-angle coordinate as origin.
Picture frame is obtained by the photosensitive screen in monocular cam.
Vehicle is during the entire process of driving to slope pavement from level road, when vehicle front-wheel is just pressed onto point of interface T
When point, the headstock of vehicle is nearest from slope pavement, and the monocular vision camera in vehicle front bumper is arranged at this time from slope
Also recently., measurement accuracy highest.When vehicle front-wheel is just pressed onto point of interface T point, pressure that vehicle front-wheel is applied by road surface
Value has one and increases suddenly, and specific pressure value increase can show as tire pressure and increase suddenly, therefore pass through vehicle front-wheel
Tire pressure variation may determine that whether the current front-wheel of vehicle is just pressed onto point of interface T point, and using this moment as detection moment.
As a kind of alternative embodiment, image collection module 501 is configured as obtaining the tire pressure value that tyre pressure sensor collects in real time,
When tire pressure value increases default threshold pressure within a preset time interval, the monocular vision camera being arranged on the downside of headstock is obtained
Current focus and the picture frame including front lane line that collects of monocular vision camera.
Coordinate determining module 502, the lane line end point being configured to determine that in picture frame, and in plane right-angle coordinate
Middle determination and the corresponding lane line end point coordinate of lane line end point, by the absolute value of the ordinate of lane line end point coordinate
As current vertical distance.
As a kind of alternative embodiment, the lane line end point in picture frame is determined, comprising:
Two front lane lines are selected in picture frame.It is understood that since the lane on road surface may be more than
Two, therefore the front lane line in picture frame can be more than two.
The intersection point of the extended line of two front lane lines is determined as lane line end point.It is understood that picture frame
In include front lane line extended line, can by picture frame front lane line carry out feature point extraction, connection
It ties multiple characteristic points and extends, to obtain the extended line of front lane line, the image for including front lane line extended line can be with
The virtual image new as one is stored.
Due to foring certain angle between road surface and the optical axis and road surface of monocular vision camera, monocular vision is taken the photograph
The two front lane lines collected as head are not parallel to each other completely, but are mutually formed with certain angle before
, and the extended line of last two fronts lane line can intersect at a point, which is known as lane line end point.Lane heading line off
Point is not necessary being in picture frame, received by photosensitive screen a bit, but one obtained after image procossing
A virtual point.Lane line end point can be regarded as the infinite point on slope pavement after monocular vision camera as flat
Imaging on face.
As shown in figure 3, the vertical distance of lane line end point is smaller, illustrate that the value of slope of slope pavement is smaller.Such as Fig. 4 institute
Show, the vertical distance of lane line end point is bigger, illustrates that the value of slope of slope pavement is bigger.
Simultaneously as lane line is decorated in the position on some road surfaces close to roadside, but center is to replace vehicle using isolation strip
Diatom, therefore, as a kind of alternative embodiment, this method can also be chosen in picture frame isolation strip lower edge line and with
The adjacent front lane line of isolation strip lower edge line.By the extended line of isolation strip lower edge line and adjacent front lane line
The intersection point of extended line is determined as lane line end point.
Ideally, multiple in picture frame if all lane lines on road surface are all parallel to each other
The extended line of front lane line finally can all intersect at same point, therefore only one lane line end point in a picture frame.
But due to that may have common runway, fast, Emergency Vehicle Lane, ring road, turning guided vehicle road on road surface
All lane lines etc. a variety of positions, the lane of a variety of curvature, therefore on road surface may not be to be parallel to each other completely, at this time root
It is possible to can determine whether out more than one lane line end point according to multiple front lane lines in picture frame.
As a kind of alternative embodiment, coordinate determining module 502 is also configured to
After determining multiple lane line end points in picture frame, determining and multiple lane lines in plane right-angle coordinate
The corresponding multiple lane line end point coordinates of end point.
Gradient computing module 503 is configured as input quantity substituting into current focus and current vertical distance preparatory
In the functional relation for expressing value of slope using vertical distance and focal length value of storage, the current hill grade value that is exported.
When there are multiple lane end point coordinates, as a kind of alternative embodiment, gradient computing module 503 is also configured
Are as follows:
It is substituted into using the average value of the absolute value of the ordinate of multiple lane line end point coordinates as current vertical distance pre-
In corresponding relationship between the vertical distance first stored and value of slope.
The beneficial effect of technical solution provided in an embodiment of the present invention includes at least:
Slope detection device provided by the invention based on monocular vision, including image collection module 501, coordinate determine mould
Block 502 and gradient computing module 503, image collection module 501 are configured as obtaining the monocular vision being arranged on the downside of headstock
The picture frame including front lane line that the current focus and monocular vision camera of camera collect, and in picture frame
In plane right-angle coordinate established as origin using the central point of picture frame, to analyze road surface ahead situation.Coordinate determines mould
Block 502 is configured to determine that the lane line end point in picture frame, and the determining and lane heading line off in plane right-angle coordinate
The corresponding lane line end point coordinate of point, using the absolute value of the ordinate of lane line end point coordinate as current vertical distance,
To assign numerical value meaning for lane end point.Gradient computing module 503 be configured as by current focus and current vertical away from
From as input quantity, substitutes into the pre-stored functional relation for expressing value of slope using vertical distance and focal length value, obtain
The current hill grade value of output reduces overall cost to realize high-precision slope detection based on monocular vision camera.
All the above alternatives can form alternative embodiment of the invention using any combination, herein no longer
It repeats one by one.
The present embodiment and embodiment one are based on identical inventive concept, are that device corresponding with embodiment of the method one is implemented
Example, therefore it should be appreciated by those skilled in the art that being similarly adapted to the present embodiment to the explanation of embodiment one, some technologies are thin
Section is no longer described in detail in the present embodiment.
In this application, it should be understood that term " first ", " second " etc. are used for description purposes only, and should not be understood as
Indication or suggestion relative importance or the quantity for implicitly indicating indicated technical characteristic.
The above is merely for convenience of it will be understood by those skilled in the art that technical solution of the present invention, not to limit
The present invention.All within the spirits and principles of the present invention, any modification, equivalent replacement, improvement and so on should be included in this
The protection scope of invention.
Claims (10)
1. a kind of slope detection method based on monocular vision characterized by comprising
The current focus and the monocular vision camera for obtaining the monocular vision camera being arranged on the downside of headstock acquire
The picture frame including front lane line arrived, and established and put down using the central point of described image frame as origin in described image frame
Face rectangular coordinate system；
Determine the lane line end point in described image frame, and the determining and lane line disappears in the plane right-angle coordinate
The corresponding lane line end point coordinate of point is lost, using the absolute value of the ordinate of the lane line end point coordinate as current vertical
Distance；
Regard the current focus and current vertical distance as input quantity, substitute into it is pre-stored using vertical distance and
In functional relation of the focal length value to express value of slope, the current hill grade value that is exported.
2. the method according to claim 1, wherein the traveling of the optical axis and vehicle of the monocular vision camera
Direction keeping parallelism.
3. the method according to claim 1, wherein the lane line end point in the determining described image frame,
Include:
Two front lane lines are selected in described image frame；
The intersection point of the extended line of two fronts lane line is determined as the lane line end point.
4. the method according to claim 1, wherein the method also includes:
It is determining and described more in the plane right-angle coordinate after determining multiple lane line end points in described image frame
The corresponding multiple lane line end point coordinates of a lane line end point；
It is substituted into using the average value of the absolute value of the ordinate of the multiple lane line end point coordinate as current vertical distance pre-
In corresponding relationship between the vertical distance first stored and value of slope.
5. the monocular vision images the method according to claim 1, wherein the vertical distance is indicated with L
The focal length value of head indicates that the value of slope is indicated with α with f, the letter that value of slope is expressed using vertical distance and focal length value
Number relationship are as follows:
6. a kind of slope detection device based on monocular vision characterized by comprising
Image collection module is configured as obtaining the current focus that monocular vision camera on the downside of headstock is arranged in and described
The picture frame including front lane line that monocular vision camera collects, and with described image frame in described image frame
Central point establishes plane right-angle coordinate as origin；
Coordinate determining module, the lane line end point being configured to determine that in described image frame, and in the plane rectangular coordinates
Lane line end point coordinate corresponding with the lane line end point is determined in system, by the vertical seat of the lane line end point coordinate
Target absolute value is as current vertical distance；
Gradient computing module is configured as input quantity substituting into the current focus and current vertical distance pre-
In the functional relation for expressing value of slope using vertical distance and focal length value first stored, the current hill grade value that is exported.
7. device according to claim 6, which is characterized in that the traveling of the optical axis and vehicle of the monocular vision camera
Direction keeping parallelism.
8. device according to claim 6, which is characterized in that the coordinate determining module is also configured to
Two front lane lines are selected in described image frame；
The intersection point of the extended line of two fronts lane line is determined as the lane line end point.
9. device according to claim 6, which is characterized in that the coordinate determining module is also configured to
It is determining and described more in the plane right-angle coordinate after determining multiple lane line end points in described image frame
The corresponding multiple lane line end point coordinates of a lane line end point；
The gradient computing module is also configured to
It is substituted into using the average value of the absolute value of the ordinate of the multiple lane line end point coordinate as current vertical distance pre-
In corresponding relationship between the vertical distance first stored and value of slope.
10. device according to claim 7, which is characterized in that the vertical distance indicates that the monocular vision is taken the photograph with L
As the focal length value of head is indicated with f, the value of slope is indicated with α, described to express value of slope using vertical distance and focal length value
Functional relation are as follows:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910398135.0A CN110044333A (en) | 2019-05-14 | 2019-05-14 | A kind of slope detection method and device based on monocular vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910398135.0A CN110044333A (en) | 2019-05-14 | 2019-05-14 | A kind of slope detection method and device based on monocular vision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110044333A true CN110044333A (en) | 2019-07-23 |
Family
ID=67281963
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910398135.0A Pending CN110044333A (en) | 2019-05-14 | 2019-05-14 | A kind of slope detection method and device based on monocular vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110044333A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110751127A (en) * | 2019-10-30 | 2020-02-04 | 芜湖汽车前瞻技术研究院有限公司 | Distance determination method, device and storage medium |
| CN111077893A (en) * | 2019-12-31 | 2020-04-28 | 驭势科技(南京)有限公司 | Navigation method based on multiple vanishing points, electronic equipment and storage medium |
| CN111595253A (en) * | 2020-05-13 | 2020-08-28 | 北京三快在线科技有限公司 | Method, device and equipment for determining distance between vehicle and lane line and storage medium |
| CN112525147A (en) * | 2020-12-08 | 2021-03-19 | 北京嘀嘀无限科技发展有限公司 | Distance measurement method for automatic driving equipment and related device |
| CN112639814A (en) * | 2020-05-29 | 2021-04-09 | 华为技术有限公司 | Road condition detection method and device |
| CN112862890A (en) * | 2021-02-07 | 2021-05-28 | 黑芝麻智能科技(重庆)有限公司 | Road gradient prediction method, road gradient prediction device and storage medium |
| CN113433939A (en) * | 2021-03-11 | 2021-09-24 | 上海擎朗智能科技有限公司 | Control method and device for mobile robot, mobile robot and storage medium |
| CN113514036A (en) * | 2021-04-24 | 2021-10-19 | 中国建筑第五工程局有限公司 | Super high-rise building perpendicularity measurement and control method |
| CN113516712A (en) * | 2021-04-27 | 2021-10-19 | 北京车和家信息技术有限公司 | Target position determining method and device |
| CN114141011A (en) * | 2021-11-23 | 2022-03-04 | 北京安融畅信息技术有限公司 | Steep slope and continuous downhill road section identification method and potential safety hazard troubleshooting method |
| CN115597550A (en) * | 2022-10-09 | 2023-01-13 | 深圳市欧冶半导体有限公司（Cn） | Ramp monocular distance measurement method and device based on vanishing point and target ground point |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6024473A (en) * | 1997-02-21 | 2000-02-15 | Valeo Vision | Motor vehicle headlight reflector having laterally juxtaposed zones, a headlight constructed therefrom and a method of making the reflector |
| CN101922929A (en) * | 2009-06-09 | 2010-12-22 | 财团法人车辆研究测试中心 | Vehicle inclination sensing method and head lamp automatic leveling system applying same |
| CN103679119A (en) * | 2012-09-10 | 2014-03-26 | 株式会社理光 | Self-adapting slope detection method and device |
-
2019
- 2019-05-14 CN CN201910398135.0A patent/CN110044333A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6024473A (en) * | 1997-02-21 | 2000-02-15 | Valeo Vision | Motor vehicle headlight reflector having laterally juxtaposed zones, a headlight constructed therefrom and a method of making the reflector |
| CN101922929A (en) * | 2009-06-09 | 2010-12-22 | 财团法人车辆研究测试中心 | Vehicle inclination sensing method and head lamp automatic leveling system applying same |
| CN103679119A (en) * | 2012-09-10 | 2014-03-26 | 株式会社理光 | Self-adapting slope detection method and device |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110751127B (en) * | 2019-10-30 | 2022-07-12 | 芜湖汽车前瞻技术研究院有限公司 | Distance determination method, device and storage medium |
| CN110751127A (en) * | 2019-10-30 | 2020-02-04 | 芜湖汽车前瞻技术研究院有限公司 | Distance determination method, device and storage medium |
| CN111077893A (en) * | 2019-12-31 | 2020-04-28 | 驭势科技(南京)有限公司 | Navigation method based on multiple vanishing points, electronic equipment and storage medium |
| CN111595253A (en) * | 2020-05-13 | 2020-08-28 | 北京三快在线科技有限公司 | Method, device and equipment for determining distance between vehicle and lane line and storage medium |
| CN112639814A (en) * | 2020-05-29 | 2021-04-09 | 华为技术有限公司 | Road condition detection method and device |
| CN112525147B (en) * | 2020-12-08 | 2022-11-08 | 北京嘀嘀无限科技发展有限公司 | Distance measurement method for automatic driving equipment and related device |
| CN112525147A (en) * | 2020-12-08 | 2021-03-19 | 北京嘀嘀无限科技发展有限公司 | Distance measurement method for automatic driving equipment and related device |
| CN112862890A (en) * | 2021-02-07 | 2021-05-28 | 黑芝麻智能科技(重庆)有限公司 | Road gradient prediction method, road gradient prediction device and storage medium |
| CN112862890B (en) * | 2021-02-07 | 2023-07-04 | 黑芝麻智能科技(重庆)有限公司 | Road gradient prediction method, device and storage medium |
| CN113433939A (en) * | 2021-03-11 | 2021-09-24 | 上海擎朗智能科技有限公司 | Control method and device for mobile robot, mobile robot and storage medium |
| CN113433939B (en) * | 2021-03-11 | 2024-04-12 | 上海擎朗智能科技有限公司 | Control method and device for mobile robot, mobile robot and storage medium |
| CN113514036A (en) * | 2021-04-24 | 2021-10-19 | 中国建筑第五工程局有限公司 | Super high-rise building perpendicularity measurement and control method |
| CN113514036B (en) * | 2021-04-24 | 2022-05-17 | 中国建筑第五工程局有限公司 | Super high-rise building verticality measurement and control method |
| CN113516712A (en) * | 2021-04-27 | 2021-10-19 | 北京车和家信息技术有限公司 | Target position determining method and device |
| CN114141011A (en) * | 2021-11-23 | 2022-03-04 | 北京安融畅信息技术有限公司 | Steep slope and continuous downhill road section identification method and potential safety hazard troubleshooting method |
| CN115597550A (en) * | 2022-10-09 | 2023-01-13 | 深圳市欧冶半导体有限公司（Cn） | Ramp monocular distance measurement method and device based on vanishing point and target ground point |
| CN115597550B (en) * | 2022-10-09 | 2023-09-26 | 深圳市欧冶半导体有限公司 | Ramp monocular ranging method and device based on vanishing point and target grounding point |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110044333A (en) | A kind of slope detection method and device based on monocular vision | |
| CN103885573B (en) | The auto-correction method of automobile-used display system and its system | |
| US8369577B2 (en) | Vehicle position recognition system | |
| JP5157067B2 (en) | Automatic travel map creation device and automatic travel device. | |
| US20190026918A1 (en) | Information Processing Apparatus | |
| TWI534764B (en) | Apparatus and method for vehicle positioning | |
| US20180134289A1 (en) | Lane division line recognition apparatus, lane division line recognition method, driving assist apparatus including lane division line recognition apparatus, and driving assist method including lane division line recognition method | |
| US11533455B2 (en) | Imaging apparatus and distance measuring apparatus using the same | |
| CN109410264B (en) | Front vehicle distance measuring method based on laser point cloud and image fusion | |
| US20150161457A1 (en) | Three-dimensional object detection device, and three-dimensional object detection method | |
| EP2372607A2 (en) | Scene matching reference data generation system and position measurement system | |
| EP2372611A2 (en) | Scene matching reference data generation system and position measurement system | |
| US20090122136A1 (en) | Object detection device | |
| CN106289159B (en) | Vehicle distance measurement method and device based on distance measurement compensation | |
| EP2928182B1 (en) | Onboard image processing system | |
| CA2975139A1 (en) | Stereo camera system for collision avoidance during aircraft surface operations | |
| US20170313253A1 (en) | Method for operating a driver assistance system of a motor vehicle, driver assistance system and motor vehicle | |
| WO2014104183A1 (en) | Boundary recognition device and branch assessment device | |
| CN105716567A (en) | Method for determining the distance between an object and a motor vehicle by means of a monocular imaging device | |
| CN111507130A (en) | Lane level positioning method and system, computer equipment, vehicle and storage medium | |
| CN111595253B (en) | Method, device and equipment for determining distance between vehicle and lane line and storage medium | |
| WO2018179281A1 (en) | Object detection device and vehicle | |
| CN108280401A (en) | A kind of pavement detection method, apparatus, cloud server and computer program product | |
| JP2018077162A (en) | Vehicle position detection device, vehicle position detection method and computer program for vehicle position detection | |
| JP2015194397A (en) | Vehicle location detection device, vehicle location detection method, vehicle location detection computer program and vehicle location detection system |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190723 |