CN110222658A - The acquisition methods and device of road vanishing point position - Google Patents
The acquisition methods and device of road vanishing point position Download PDFInfo
- Publication number
- CN110222658A CN110222658A CN201910502866.5A CN201910502866A CN110222658A CN 110222658 A CN110222658 A CN 110222658A CN 201910502866 A CN201910502866 A CN 201910502866A CN 110222658 A CN110222658 A CN 110222658A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- described image
- lane
- vanishing point
- vehicle arrangement
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- G06V20/584—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of vehicle lights or traffic lights
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the invention provides the acquisition methods and device of a kind of road vanishing point position, are applied to the first vehicle arrangement, and method includes: the image that acquisition includes the second vehicle arrangement；Described image based on acquisition carries out lane detection, obtains the lane line information in lane locating for the first vehicle arrangement described in described image；Described image based on acquisition carries out vehicle detection, obtains vehicle parameter of second vehicle arrangement relative to described image；Based on the lane line information and the vehicle parameter, the corresponding vanishing point position in lane locating for first vehicle arrangement is determined.
Description
Technical field
The present invention relates to the acquisition methods and device in auxiliary driving field more particularly to a kind of road vanishing point position.
Background technique
Advanced DAS (Driver Assistant System) (ADAS, Advanced the Driving Assistant of view-based access control model algorithm
System) it is widely used in safe driving, utilizes vehicle-mounted sensor, while the car is driving induction in real time
The environment and navigation instrument map datum of surrounding, carry out the operation and analysis of system, so that allowing driver to perceive in advance can
The danger that can occur, effectively increases the comfortableness and security of car steering, and road vanishing point position is the pass of wherein every technology
Key input information.
The relevant technologies obtain the position of road vanishing point by way of to image progress lane line drawing or using iteration, but
Such technology at least needs to find two lane lines, hides in the case where lane line is very difficult to acquisition, such as caused by congestion
Gear, it is desirable to which finding a plurality of lane line just becomes more difficult, so as to cause the road that can not obtain road vanishing point position or acquisition
The precision of vanishing point position is not high.
Summary of the invention
The embodiment of the present invention provides the acquisition methods and device of a kind of road vanishing point position, can be quick by information of vehicles
Accurately get road vanishing point position.
The technical solution of the embodiment of the present invention is achieved in that
The embodiment of the present invention provides a kind of acquisition methods of road vanishing point position, is applied to the first vehicle arrangement, comprising:
Acquisition includes the image of the second vehicle arrangement；
Described image based on acquisition carries out lane detection, obtains vehicle locating for the first vehicle arrangement described in described image
The lane line information in road；
Described image based on acquisition carries out vehicle detection, obtains vehicle of second vehicle arrangement relative to described image
Parameter；
Based on the lane line information and the vehicle parameter, determines locating for first vehicle arrangement that lane is corresponding and go out
Point position.
In above scheme, the described image based on acquisition carries out lane detection, obtains described in described image the
The lane line information in lane locating for one vehicle arrangement, comprising:
To described image carry out region interception, obtain include lane locating for first vehicle arrangement image-region；
Obtain the pixel for belonging to lane line in described image region；
By carrying out curve fitting to the pixel of acquisition, the first vehicle arrangement described in described image region is determined
The lane line in locating lane.
It is described by carrying out curve fitting to the pixel of acquisition in above scheme, it determines in described image region
The lane line in lane locating for first vehicle arrangement, comprising:
Left-lane line and right-lane line are included at least in response to the lane line, the pixel based on acquisition determines
The higher lane line of integrality in the left-lane line and right-lane line；
It carries out curve fitting to the pixel of the higher lane line of the integrality, to determine described in described image region
The lane line in lane locating for the first vehicle arrangement.
In above scheme, the described image based on acquisition carries out vehicle detection, obtains the second vehicle arrangement phase
For the vehicle parameter of described image, comprising:
The neural network model that the described image input training of acquisition is obtained, identification obtain second vehicle arrangement
Multiple frames；
Non-maximum restraining (NMS, Non Maximum Supp ression) is carried out to the multiple frame that identification obtains
Processing, obtains the target frame for corresponding to second vehicle arrangement；
Based on the relative positional relationship of the target frame and described image, second vehicle arrangement is obtained relative to institute
State the vehicle parameter of image.
In above scheme, the relative positional relationship based on the target frame and described image obtains described second
Vehicle parameter of the vehicle arrangement relative to described image, comprising:
The position of upper side frame and lower frame based on the target frame in described image determines that second vehicle is set
The standby height of car relative to described image；
Position of the upper side frame based on the target frame in described image, determines the upper side frame relative to the figure
The first distance in the first coordinate system direction, the upper side frame are parallel with first coordinate system direction as in.
It is described to be based on the vehicle parameter and the lane line information in above scheme, determine first vehicle arrangement
The corresponding vanishing point position in locating lane, comprising:
Determine second distance of the vanishing point position relative to the first coordinate system direction in described image；
Drafting is parallel with the first coordinate system direction in described image, and spacing is the first straight line of the second distance；
First position of the intersection point of the first straight line and the lane line drawn in described image is obtained, it will be described
First position is as the vanishing point position.
In above scheme, second distance of the determining vanishing point position relative to the first coordinate system direction in described image,
Include:
Second distance of the vanishing point position relative to the first coordinate system direction in described image is determined by following relationship:
The second distance be equal to the height of car constant times and the first distance and.
It is described to be based on the vehicle parameter and the lane line information in above scheme, determine first vehicle arrangement
The corresponding vanishing point position in locating lane, comprising:
Include height of car of second vehicle arrangement relative to described image in response to the vehicle parameter, is based on institute
State the second distance Yu the height of car of height of car, the vanishing point position relative to the first coordinate system direction in described image
The positive correlation for being in determines the second distance；
Drafting is parallel with the top edge of described image, and spacing is the second straight line of the second distance；
The second position of the intersection point of the second straight line and the lane line drawn in described image is obtained, it will be described
The second position is as the vanishing point position.
The embodiment of the present invention provides a kind of acquisition device of road vanishing point position, is applied to the first vehicle arrangement, comprising:
Acquisition unit, for acquiring the image including the second vehicle arrangement；
First detection unit carries out lane detection for the described image based on acquisition, obtains described in described image
The lane line information in lane locating for the first vehicle arrangement；
Second detection unit carries out vehicle detection for the described image based on acquisition, obtains second vehicle arrangement
Vehicle parameter relative to described image；
Determination unit determines the first vehicle arrangement institute for being based on the lane line information and the vehicle parameter
Locate the corresponding vanishing point position in lane.
In above scheme, the first detection unit obtains including described for carrying out region interception to described image
The image-region in lane locating for one vehicle arrangement；
Obtain the pixel for belonging to lane line in described image region；
By carrying out curve fitting to the pixel of acquisition, the first vehicle arrangement described in described image region is determined
The lane line in locating lane.
In above scheme, the first detection unit, for including at least left-lane line and the right side in response to the lane line
Lane line, the pixel based on acquisition determine the higher lane line of integrality in the left-lane line and right-lane line；
It carries out curve fitting to the pixel of the higher lane line of the integrality, to determine described in described image region
The lane line in lane locating for the first vehicle arrangement.
In above scheme, the second detection unit, the nerve net for obtaining the described image input training of acquisition
Network model, identification obtain multiple frames of second vehicle arrangement；
NMS processing is carried out to the multiple frame that identification obtains, obtains the target side for corresponding to second vehicle arrangement
Frame；
Based on the relative positional relationship of the target frame and described image, second vehicle arrangement is obtained relative to institute
State the vehicle parameter of image.
In above scheme, the second detection unit, for based on the target frame upper side frame and lower frame in institute
The position in image is stated, determines height of car of second vehicle arrangement relative to described image；
Position of the upper side frame based on the target frame in described image, determines the upper side frame relative to the figure
The first distance in the first coordinate system direction, the upper side frame are parallel with first coordinate system direction as in.
In above scheme, the determination unit, for determining vanishing point position relative to the first coordinate system side in described image
To second distance；
Drafting is parallel with the first coordinate system direction in described image, and spacing is the first straight line of the second distance；
First position of the intersection point of the first straight line and the lane line drawn in described image is obtained, it will be described
First position is as the vanishing point position.
In above scheme, the determination unit, for determining the vanishing point position relative to the figure by following relationship
The second distance in the first coordinate system direction as in:
The second distance be equal to the height of car constant times and the first distance and.
In above scheme, the determination unit, for including the second vehicle arrangement phase in response to the vehicle parameter
For the height of car of described image, based on the height of car, the vanishing point position relative to the first coordinate in described image
It is the second distance in direction and the positive correlation that the height of car is in, determines the second distance；
Drafting is parallel with the top edge of described image, and spacing is the second straight line of the second distance；
The second position of the intersection point of the second straight line and the lane line drawn in described image is obtained, it will be described
The second position is as the vanishing point position.
The embodiment of the present invention provides a kind of device for obtaining road vanishing point position, comprising:
Memory, for storing executable instruction；
Processor when for executing the executable instruction stored in the memory, is realized provided in an embodiment of the present invention
The acquisition methods of road vanishing point position.
The embodiment of the present invention provides a kind of storage medium, is stored with executable instruction, real when for causing processor to execute
The acquisition methods of existing road vanishing point provided in an embodiment of the present invention position.
The embodiment of the present invention has the advantages that
Using the acquisition methods of road vanishing point provided in an embodiment of the present invention, by acquiring the figure including the second vehicle arrangement
Picture determines that the lane line information in lane locating for the first vehicle arrangement and the second vehicle arrangement are joined relative to the vehicle of image respectively
Number, the horizontal line where determining vanishing point in the picture according to obtained vehicle parameter are horizontal where based on lane line and vanishing point
The intersection point of line determines the position of vanishing point.In this way, quickly can accurately be determined using auxiliary information of vehicles and a lane line
Vanishing point position, use scope is wide, enhances user experience.
Detailed description of the invention
Fig. 1 is the configuration diagram of the acquisition system of road vanishing point provided in an embodiment of the present invention position；
Fig. 2 is the hardware structural diagram of the acquisition device of road vanishing point provided in an embodiment of the present invention position；
Fig. 3 is the flow diagram of the acquisition methods of road vanishing point provided in an embodiment of the present invention position；
Fig. 4 is lane line binary system schematic diagram provided in an embodiment of the present invention；
Fig. 5 is lane line pixel distribution schematic diagram provided in an embodiment of the present invention；
Fig. 6 is lane line Pixel fit schematic diagram provided in an embodiment of the present invention；
Fig. 7 is the schematic diagram that NMS provided in an embodiment of the present invention handles vehicle trim；
Fig. 8 is determining vehicle parameter schematic diagram provided in an embodiment of the present invention；
Fig. 9 is determining vanishing point schematic diagram provided in an embodiment of the present invention；
Figure 10 is the flow diagram of the acquisition methods of road vanishing point provided in an embodiment of the present invention position；
Figure 11 is the composed structure schematic diagram of the acquisition device of road vanishing point provided in an embodiment of the present invention.
Specific embodiment
To make the objectives, technical solutions, and advantages of the present invention clearer, below in conjunction with attached drawing to the present invention make into
It is described in detail to one step, described embodiment is not construed as limitation of the present invention, and those of ordinary skill in the art are not having
All other embodiment obtained under the premise of creative work is made, shall fall within the protection scope of the present invention.
In the following description, it is related to " some embodiments ", which depict the subsets of all possible embodiments, but can
To understand, " some embodiments " can be the same subsets or different subsets of all possible embodiments, and can not conflict
In the case where be combined with each other.
In the following description, related term " first second third " be only be the similar object of difference, no
Represent the particular sorted for being directed to object, it is possible to understand that ground, " first second third " can be interchanged specific in the case where permission
Sequence or precedence so that the embodiment of the present invention described herein can be other than illustrating herein or describing
Sequence is implemented.
Unless otherwise defined, all technical and scientific terms used herein and belong to technical field of the invention
The normally understood meaning of technical staff is identical.Term used herein is intended merely to the purpose of the description embodiment of the present invention,
It is not intended to limit the present invention.
Before the embodiment of the present invention is further elaborated, to noun involved in the embodiment of the present invention and term
It is illustrated, noun involved in the embodiment of the present invention and term are suitable for following explanation.
1) lane line, white solid line, white dotted line, yellow solid line, yellow dotted line for indicating to separate different lanes on road, and
The markings such as road boundary.
2) lane detection refers to finding lane line in picture, and with point range or straight line (curve) equation table
It shows and.
3) vanishing point, be otherwise known as end point, refers to that two or more pieces represents parallel lines lines distally Horizon on lane line
Line stretching, extension is until that point polymerizeing.
4) top view refers in the process of moving, in the picture of normal viewing angle shooting, lane line these parallel lines are all
It can intersect in vanishing point, by perspective transform, lane line can be for conversion into parallel lines, the image after this perspective transform claims
Make top view.
5) in response to the condition or state relied on for indicating performed operation, when the relied on condition of satisfaction
Or when state, performed one or more operations be can be in real time, it is possible to have the delay of setting；Do not saying especially
In the case where bright, there is no the limitations for executing sequencing for performed multiple operations.
Illustrate that the exemplary application for realizing the acquisition device of the road vanishing point position of the embodiment of the present invention, the present invention are real below
The acquisition device for applying the road vanishing point position of example offer can also be set by vehicle arrangement and server coordinated implementation by vehicle
It is standby voluntarily to implement.In the following, will be illustrated in conjunction with Fig. 1.
Fig. 1 is the configuration diagram of the acquisition system 100 of road vanishing point provided in an embodiment of the present invention position, referring to Fig. 1,
An exemplary application is supported to realize, vehicle arrangement 200 connects server 400 by network 300, and network 300 can be extensively
Domain net or local area network, or be combination, realize that data are transmitted using Radio Link.
It is provided with image collecting device on vehicle arrangement 200, for acquiring the image including the second vehicle arrangement, the second vehicle
Equipment and the first vehicle arrangement (i.e. vehicle arrangement 200) can be at same lane, can also be at different lanes, vehicle arrangement hair
The vanishing point position acquisition for carrying acquired image is sent to request to server 400, and display server returns on graphical interfaces
The corresponding vanishing point position in lane locating for the first vehicle arrangement returned.
Server 400 carries out lane detection for receiving the request of vanishing point position acquisition, and based on received image, obtains
The lane line information in lane locating for the first vehicle arrangement in image；Image based on acquisition carries out vehicle detection, obtains the second vehicle
Vehicle parameter of the equipment relative to image；Based on lane line information and vehicle parameter, lane locating for the first vehicle arrangement is determined
Corresponding vanishing point position, and the vanishing point position is sent to vehicle arrangement 200.In this way, lane detection, vehicle detection and really
The operation for determining road vanishing point position is completed by server, reduces the data processing pressure of vehicle arrangement side, and be suitable for source figure
The larger situation of the resolution ratio of picture.
When the acquisition device of road vanishing point provided in an embodiment of the present invention position is voluntarily implemented by vehicle arrangement, vehicle is set
Image collecting device and processing chip, image collecting device are provided on standby 200 can acquire vehicle arrangement in the process of moving
The image of surrounding, as acquisition includes the second vehicle arrangement being located on same lane or on different lanes from logo equipment 200
Image；It handles chip and lane detection is carried out based on the image of acquisition, obtain the lane in lane locating for vehicle arrangement 200 in image
Line information, the image based on acquisition carry out vehicle detection, obtain vehicle parameter of second vehicle arrangement relative to image, are based on vehicle
Diatom information and vehicle parameter determine the corresponding vanishing point position in lane locating for vehicle arrangement 200.In this way, the determination of vanishing point position
Real time implementation can be realized in vehicle arrangement side, improve vanishing point position acquisition efficiency, enhance user experience.
The acquisition device of road vanishing point provided in an embodiment of the present invention position may be embodied as hardware or software and hardware combining
Mode, illustrate the various exemplary implementations of the acquisition device of road vanishing point provided in an embodiment of the present invention position below.
Fig. 2 is the hardware structural diagram of the acquisition device of road vanishing point provided in an embodiment of the present invention position, is actually answered
It can be applied to the various vehicle arrangements of operation application program above-mentioned, referring to fig. 2, road vanishing point shown in Fig. 2 position in
Acquisition device 200 include: at least one processor 201, memory 202, user interface 203 and at least one network interface
204.Various components in the acquisition device 200 of road vanishing point position are coupled by bus system 205.It is appreciated that
Bus system 205 is for realizing the connection communication between these components.Bus system 205 is also wrapped in addition to including data/address bus
Include power bus, control bus and status signal bus in addition.But for the sake of clear explanation, various buses are all marked in Fig. 2
For bus system 205.
Wherein, user interface 203 may include display, keyboard, mouse, trace ball, click wheel, key, button, sense of touch
Plate or touch screen etc..
It is appreciated that memory 202 can be volatile memory or nonvolatile memory, may also comprise volatibility and
Both nonvolatile memories.Wherein, nonvolatile memory can be read-only memory (ROM, Read Only Memory),
Programmable read only memory (PROM, Programmable Read-Only Memory), Erasable Programmable Read Only Memory EPROM
(EPROM, Erasable Programmable Read-Only Memory), flash memory (Flash Memory) etc..Volatibility is deposited
Reservoir can be random access memory (RAM, Random Access Memory), be used as External Cache.By showing
Example property but be not restricted explanation, the RAM of many forms is available, such as static random access memory (SRAM, Static
Random Access Memory), synchronous static random access memory (SSRAM, Synchronous Static Random
Access Memory).The memory 202 of description of the embodiment of the present invention is intended to include depositing for these and any other suitable type
Reservoir.
Memory 202 in the embodiment of the present invention is for storing various types of data to support obtaining for road vanishing point position
Take the operation of device 200.The example of these data includes: to appoint for what is operated on the acquisition device 200 of road vanishing point position
What computer program realizes the program of the acquisition methods of the road vanishing point position of the embodiment of the present invention such as executable program 2021
It may be embodied in executable program 2021.
The acquisition methods of road vanishing point provided in an embodiment of the present invention position can be applied in processor 201, Huo Zheyou
Processor 201 is realized.Processor 201 may be a kind of IC chip, the processing capacity with signal.In the process of realization
In, each steps of the acquisition methods of road vanishing point position can by the integrated logic circuit of the hardware in processor 201 or
The instruction of software form is completed.Above-mentioned processor 201 can be general processor, digital signal processor (DSP, Digital
Signal Processor) either other programmable logic device, discrete gate or transistor logic, discrete hardware group
Part etc..Processor 201 may be implemented or execute disclosed each method, step and logic diagram in the embodiment of the present invention.It is logical
It can be microprocessor or any conventional processor etc. with processor.The step of the method in conjunction with provided by the embodiment of the present invention
Suddenly, hardware decoding processor can be embodied directly in and execute completion, or with the hardware and software module group in decoding processor
Conjunction executes completion.Software module can be located in storage medium, which is located at memory 202, and the reading of processor 201 is deposited
Information in reservoir 202, the step of completing the acquisition methods of road vanishing point provided in an embodiment of the present invention position in conjunction with its hardware.
Next the exemplary of acquisition device of the road vanishing point position above-mentioned for realizing the embodiment of the present invention will be combined to answer
With and implement, illustrate realize the embodiment of the present invention road vanishing point position acquisition methods.
Fig. 3 is the flow diagram of the acquisition methods of road vanishing point provided in an embodiment of the present invention position, referring to Fig. 3, originally
The acquisition methods of road vanishing point position that inventive embodiments provide include:
Step 301: the acquisition of the first vehicle arrangement includes the image of the second vehicle arrangement.
Image collecting device and processing chip are provided in actual implementation, on vehicle arrangement, image collecting device can
Acquire vehicle arrangement in the process of moving around image, such as acquisition include the second vehicle arrangement image, here, the second vehicle
Equipment and the first vehicle arrangement can be on same lane, can also be on different lanes, handle image of the chip based on acquisition
Carry out subsequent processing.
In some embodiments, the first vehicle arrangement is identical as the second vehicle arrangement driving direction, it is to be understood that are as follows:
First vehicle arrangement is current vehicle, and the second vehicle arrangement is located in front of current vehicle.Certainly, in further embodiments,
The driving direction of one vehicle arrangement and the second vehicle arrangement also can not be identical, is not specifically limited herein.
Step 302: the first vehicle arrangement carries out lane detection based on the image of acquisition, obtains the first vehicle in image and sets
The lane line information in standby locating lane.
Here, the first vehicle arrangement needs to obtain lane locating for current vehicle before the determination for carrying out road vanishing point position
Lane line information, and in actual implementation, the first vehicle arrangement at least can obtain in the following way lane line information:
To image carry out region interception, obtain include lane locating for the first vehicle arrangement image-region；Obtain image district
Belong to the pixel of lane line in domain；By carrying out curve fitting to the pixel of acquisition, the first vehicle in image-region is determined
The lane line in lane locating for equipment.
In practical applications, when the distortion of the image capture device of the first vehicle arrangement is larger, to acquired image
Before being detected, needs to be corrected acquired image using modes such as gridiron patterns, the image after correction is carried out
Region interception, obtain include lane locating for current vehicle equipment area-of-interest, in this way, only to the image-region comprising lane
It is handled, treatment effeciency can be improved.
The environment of the road as locating for vehicle is in constantly variation, such as in different times and section, the face in lane
Color, illumination and clarity all can be different, in order to cope with changeable road conditions, need to carry out in conjunction with a variety of threshold filtering methods
Processing.Here, it is directed to the lane line of different colours, and the lane line under different illumination conditions, the lane line of different clarity, according to
Different color spaces carries out different processing using different Grads threshold and color threshold, and by each processing mode into
Row fusion, obtains the binary system figure of lane line, and Fig. 4 is lane line binary system schematic diagram provided in an embodiment of the present invention, Cong Erjin
The pixel for belonging to lane line is extracted in drawing, and statistics with histogram is carried out to the pixel of binary system picture, counts the peak of the left and right sides
Value point carries out curve fitting as the starting point coordinate of left and right lane line, so that it is determined that the lane line in image-region.
In practical applications, the binary system figure obtained is likely present certain noise pixel, in order to accurately detect lane
Boundary, it is necessary first to determine which pixel belongs to lane line, it is therefore desirable to position the basic point in lane, Fig. 5 is that the present invention is real
The lane line pixel distribution schematic diagram for applying example offer, referring to Fig. 5, basic point is the x-axis coordinate that bottom lane occurs in Fig. 5, by
In lane line pixel all concentrate in x-axis a certain range, therefore the image of thresholding is divided into two, the right and left
Pixel distribution peak value in x-axis is likely to be lane line basic point.
After the basic point positioning in left and right lane, the first vehicle arrangement can be in the following way to the pixel march of acquisition
Line fitting, determines the lane line in lane locating for the first vehicle arrangement in image-region:
It include that at least left-lane line and right-lane line, the pixel based on acquisition determines left vehicle in response to lane line
The higher lane line of integrality in diatom and right-lane line；It carries out curve fitting to the pixel of the higher lane line of integrality,
To determine the lane line in lane locating for the first vehicle arrangement in image-region.
In actual implementation, the image collecting device in vehicle arrangement can collect a plurality of lane line, preferably complete herein
Property preferable lane line carry out the acquisition of vanishing point position, the mode of sliding window fitting of a polynomial can be used to position a lane line
Pixel, Fig. 6 are lane line Pixel fit schematic diagram provided in an embodiment of the present invention, and referring to Fig. 6, left-lane line pixel is than right vehicle
The integrality of diatom pixel wants high, therefore carry out sliding window fitting of a polynomial to left-lane line, so that it is determined that image-region
The lane line in lane locating for middle current vehicle is left-lane line.
Step 303: the first vehicle arrangement based on acquisition image carry out vehicle detection, obtain the second vehicle arrangement relative to
The vehicle parameter of image.
Here, the first vehicle arrangement is before the determination for carrying out road vanishing point position, it is also necessary to obtain and current vehicle institute
Locate vehicle parameter of second vehicle arrangement in same lane relative to image, and in actual implementation, the first vehicle arrangement is at least
Vehicle parameter can be obtained in the following way:
The neural network model that the image input training of acquisition is obtained, identification obtain multiple sides of the second vehicle arrangement
Frame；NMS processing is carried out to multiple frames that identification obtains, obtains the target frame of corresponding second vehicle arrangement；Based on target side
The relative positional relationship of frame and image obtains vehicle parameter of second vehicle arrangement relative to image.
In actual implementation, the first vehicle arrangement is by training obtained neural network model (such as Recognition with Recurrent Neural Network
(RNN, Recurrent Neural Network)) identification obtain multiple frames of the second vehicle arrangement, specifically, the first vehicle
The input layer of the image input neural network of acquisition is exported to obtain the second vehicle arrangement by output layer by equipment through hidden layer
Multiple frames the training of neural network model is illustrated here, the first vehicle arrangement can train in the following way
Above-mentioned neural network model: input layer, middle layer and the output layer that initialization neural network model includes；Construct training sample set
It closes, training sample set includes vehicle image and non-vehicle image；Using the image including the second vehicle arrangement of acquisition as defeated
Enter, the conduct output of multiple frames of corresponding second vehicle arrangement, according to the loss function of neural network model, update neural network
The model parameter of model.
And in practical applications, the frame of the second vehicle arrangement exported through above-mentioned neural network model, which has, is much overlapped
Frame corresponds to the target frame of the second vehicle arrangement in order to obtain, and the multiple frames that need to obtain identification carry out NMS processing, tool
Body: the selection highest frame of confidence level calculates its degree of overlapping with other frames, if degree of overlapping is greater than certain threshold value,
The frame is rejected, is repeated the above process, until other frame ratios when front bezel confidence level is lower.Fig. 7 is the embodiment of the present invention
The schematic diagram of the NMS processing vehicle trim of offer, referring to Fig. 7, multiple frames obtain a target frame after handling through NMS,
In this way, having filtered out the frame of overlapping, it is determined that target frame.
Here, after obtaining the target frame of the second vehicle, so that it may the opposite position based on the target frame and image
Relationship is set, determines vehicle parameter of second vehicle relative to image.In some embodiments, the first vehicle arrangement can be by as follows
Mode obtains vehicle parameter of second vehicle arrangement relative to image:
The position of upper side frame and lower frame in the picture based on target frame, determines the second vehicle arrangement relative to image
Height of car；The position of upper side frame based on target frame in the picture determines upper side frame relative to the first coordinate in image
It is the first distance in direction, upper side frame is parallel with the first coordinate system direction.
In some embodiments, position of first vehicle arrangement based on target frame in the picture determines that the second vehicle is set
The standby coordinate system the distance between in parallel relative to the height of car and upper side frame of image.
Illustratively, Fig. 8 is determining vehicle parameter schematic diagram provided in an embodiment of the present invention, referring to Fig. 8, with image upper left
Angle is origin, establishes (x, y) coordinate system, and unit is pixel, and the direction x is the first coordinate direction, and y is the second coordinate direction.Assuming that
The coordinate in the direction y of the upper side frame of target frame is y1, and the coordinate in the direction y of lower frame is y2, then the second vehicle is relative to figure
The height of car h1 of picture is y2-y1, first distance y1.
Step 304: the first vehicle arrangement is based on lane line information and vehicle parameter, determines lane locating for the first vehicle arrangement
Corresponding vanishing point position.
Here, it is found according to statistical value, the horizontal line position of road vanishing point in the picture is closely related with vehicle trim
, the direction the y coordinate of road vanishing point is included between the upper side frame of vehicle trim and lower frame in Fig. 8, thus significantly
Reduce the search range of the direction road vanishing point y coordinate.
By above each step, after having obtained lane line information and vehicle parameter, in some embodiments, the first vehicle
Equipment can be determined as follows the corresponding vanishing point position in lane locating for the first vehicle arrangement:
Determine second distance of the vanishing point position relative to the first coordinate system direction in image；It draws and the first coordinate in image
It is that direction is parallel, and spacing is the first straight line of second distance；The intersection point of the first straight line and lane line drawn is obtained described
First position in image, using first position as the vanishing point position.
Here, first straight line can be drawn relative to the direction the y coordinate in image according to vanishing point position, wherein first straight line
Horizontal line where can be understood as vanishing point.First vehicle arrangement is drawn to obtain the two according to horizontal line where vanishing point and lane line
Position where intersection point is vanishing point position.Fig. 9 is determining vanishing point schematic diagram provided in an embodiment of the present invention, referring to Fig. 9,
On the basis of having obtained vanishing point y direction coordinate, it is only necessary to which a lane line can get the position of vanishing point.
In some embodiments, the first vehicle arrangement can be determined as follows vanishing point position relative in image first
The second distance in coordinate system direction:
Second distance of the vanishing point position relative to the first coordinate system direction in image: second distance is determined by following relationship
Equal to height of car constant times and first distance and.
Illustratively, since the horizontal line position of road vanishing point in the picture is closely related with vehicle trim, herein
The direction vanishing point y coordinate can be defined as one using vehicle trim parameter, camera mounting height as variable function h=f (y1, h1,
Vh, a), wherein h is the direction vanishing point y coordinate, and y1 is the direction the y coordinate of vehicle trim top edge, and h1 is height of car, and vh is phase
Machine mounting height (unit: rice), a are a parameter that can learn.The function f for calculating h can have many implementations, such as one
A simple formula can be taken as by rule of thumb: h=y1+h1* (a/vh-1), wherein camera mounting height vh can be according to vehicle
Height is arranged to fixed value, such as general car, can set camera mounting height to 1.2 meters of unified value.According to warp
It tests or counts or with simple fitting algorithm, a reasonable value of our available a, corresponding to 1.2 meters of camera
Highly, a value can obtain preferable effect for 1.5 or so.In addition to this, (y1, h, vh can also a) pass through collection to h=f
A large amount of sampled data, i.e., the combination of different (h, y1, h1, the vh) actually occurred, uses the method for regression fit, such as line
Property fitting, quadratic fit, Support vector regression (SVR, Support Vector Regression) etc. with obtain it is more accurate
Function.
In some embodiments, the first vehicle arrangement can also be believed based on vehicle parameter and lane line in the following way
Breath, determines the corresponding vanishing point position in lane locating for the first vehicle arrangement:
Include height of car of second vehicle arrangement relative to image in response to vehicle parameter, is based on height of car, vanishing point
The positive correlation that position is in relative to the second distance Yu height of car in the first coordinate system direction in image, determine second away from
From；Drafting is parallel with the top edge of image, and spacing is the second straight line of second distance；Obtain the second straight line drawn and lane
The second position of the intersection point of line in the picture, using the second position as vanishing point position.
Here, according to the second vehicle height of car in the picture with vanishing point position relative to the positive at a distance from the direction x
Pass relationship determines the direction the y coordinate of vanishing point in the picture, and horizontal line where determining vanishing point in the picture according to this is being gone out
On the basis of point place is horizontal, it is only necessary to which a lane line can get the position of vanishing point.
Using the above embodiment of the present invention, image of first vehicle arrangement by acquisition including the second vehicle arrangement, difference
Determine the vehicle parameter of the lane line information and the second vehicle arrangement in lane locating for the first vehicle arrangement relative to image, according to
Obtained vehicle parameter determine vanishing point in the picture where horizontal line, the horizontal intersection point where based on lane line and vanishing point
Determine the position of vanishing point.In this way, quickly can accurately determine vanishing point place using auxiliary information of vehicles and a lane line
Position, use scope is wide, enhances user experience.
Figure 10 is the flow diagram of the acquisition methods of road vanishing point provided in an embodiment of the present invention position, the road vanishing point
The acquisition methods of position can be by vehicle arrangement and server cooperative achievement, referring to Figure 10, the road vanishing point of the embodiment of the present invention
The acquisition methods of position include:
Step 1001: the acquisition of the first vehicle arrangement includes the image of the second vehicle arrangement.
In actual implementation, the first vehicle arrangement is provided with image collecting device, can be by itself image collecting device
(such as camera) acquires the image of the second vehicle arrangement, in some embodiments, the first vehicle arrangement and the second vehicle arrangement row
It is identical to sail direction, the first vehicle arrangement and the first vehicle arrangement are on same lane, can be regarded as here: the first vehicle arrangement
For current vehicle, the second vehicle arrangement is located in front of current vehicle.
Certainly, in further embodiments, the first vehicle arrangement can also be on different lanes from the second vehicle arrangement, or
Person, the first vehicle arrangement also can not be identical with the second vehicle arrangement driving direction, are not specifically limited herein.
Step 1002: the first vehicle arrangement sends vanishing point position acquisition and requests to server, is carried in request including the
The image of two vehicle arrangements.
Here, the first vehicle arrangement sends vanishing point position acquisition and requests to server, carries the first vehicle mark in request
Knowledge and the image including the second vehicle arrangement make server carry out the determination of vanishing point position based on received image, and will be true
Fixed vanishing point position is based on the first vehicles identifications and returns to the first vehicle arrangement.
Step 1003: server is requested based on vanishing point position acquisition, is carried out lane detection to received image, is obtained figure
The lane line information in lane locating for the first vehicle arrangement as in.
Here, server carries out lane detection to the received image of institute using lane detection technology, in actual implementation
When, the image collecting device in the first vehicle arrangement at least collects a lane line, and it is polynomial that sliding window can be used herein
Fit approach is fitted the preferable lane line of integrality to position a lane line, and the lane line having good positioning is determined
For the lane line information in lane locating for the first vehicle arrangement.
Step 1004: server is based on vehicle testing techniques and carries out vehicle detection to received image, obtains the second vehicle
Vehicle parameter of the equipment relative to image.
In actual implementation, the neural network model that server obtains the input training of received image obtains the second vehicle
Multiple frames of equipment, and NMS processing is carried out to obtained multiple frames, obtain the target side of corresponding second vehicle arrangement
Frame, based on the relative positional relationship of obtained target frame and image, the vehicle for obtaining the second vehicle arrangement relative to image is joined
Number, referring to Fig. 8, here, vehicle parameter includes the coordinate and height of car in the direction y of the upper side frame of vehicle trim, vehicle in Fig. 8
The coordinate in the direction y of the upper side frame of frame is y1, and the coordinate in the direction y of lower frame is y2, then the second vehicle is relative to image
Height of car h1 is y2-y1.
Step 1005: vehicle parameter of the server based on the second vehicle arrangement relative to image obtains vanishing point institute in image
In horizontal line
Here, since the horizontal line position of road vanishing point in the picture is closely related with vehicle trim, is being obtained
Two vehicle arrangements relative to image height of car h1 and vehicle trim upper side frame away from the distance y1 of x-axis after, server according to
Distance horizon of the vanishing point position relative to image is calculated in formula h=y1+h1* (a/vh-1), wherein camera installation is high
Degree vh can be arranged to fixed value according to the height of vehicle, and a is the parameter that can learn, such as general car, can incite somebody to action
Camera mounting height is set as 1.2 meters of unified value, then correspondingly, rule of thumb or counting or being calculated with simple fitting
Method, a value are 1.5.Referring to Fig. 9, after h has been determined, an and spacing parallel with x-axis can be drawn as straight line (such as Fig. 9 of h
Middle number 1), which is horizontal line where the vanishing point in image.
Step 1006: server determines vanishing point position based on horizontal line where obtained lane line information and vanishing point.
Here, it after the horizontal line where obtaining vanishing point, in conjunction with the lane line that above-mentioned detection obtains, can draw
The intersection point (number 2 in such as Fig. 9) of horizontal line and this lane line where to vanishing point, the position of the intersection point is determined as where vanishing point
Position.
Step 1007: server returns to vanishing point position to the first vehicle arrangement.
In practical applications, vanishing point position is the skills such as leading vehicle distance calculating, pedestrian's distance estimations, lane detection in ADAS
The key input information of art, the first vehicle arrangement can be carried out corresponding drive based on determining vanishing point position and assisted.
Illustratively, vanishing point and lane line are interactional, in the case where known lane line, can pass through two lanes
The point of line intersection determines that the position of vanishing point then can determine lane line according to vanishing point position in the case where known vanishing point position.This
In, the first vehicle arrangement carries out the detection of lane line based on determining vanishing point position, reduces the difficulty of lane detection.
By above-mentioned steps, the image of acquisition is sent to server and carries out subsequent processing by vehicle arrangement, i.e. lane line is examined
The operation of survey, vehicle detection and determining road vanishing point position is completed by server, in this way, reducing the data of vehicle arrangement side
Processing pressure, and it is suitable for the larger situation of resolution ratio of source images.
In the following, will illustrate exemplary application of the embodiment of the present invention in an actual application scenarios.
In DAS (Driver Assistant System), vehicle arrangement is equipped with image collecting device and processing chip, image collecting device energy
Enough acquisition vehicle arrangement in the process of moving or stop when ambient condition information, such as acquisition include front vehicles image,
Here, front vehicles and current vehicle equipment can be located on same lane, may also be on different lanes.Processing chip is based on adopting
The image of collection carries out lane detection, obtains the lane line information in lane locating for current vehicle in image, the image based on acquisition
Vehicle detection is carried out, vehicle parameter of the front vehicles relative to image is obtained, based on obtained lane line information and vehicle parameter,
Determine the vanishing point position in image.
Specifically, according to statistical value it can be found that the direction the y coordinate, that is, horizontal position of vanishing point in the picture is and vehicle
Frame has close correlation.As shown in figure 8, the direction the y coordinate of vanishing point is included in the lower edges of vehicle trim
Between, the search range of the direction vanishing point y coordinate has been greatly reduced in this way, it is only necessary in this small range, find frame
With the specific location relationship of the direction vanishing point y coordinate.And in DAS (Driver Assistant System), vehicle detection and front truck early warning are one
Indispensable function, wants not needing additional calculation amount using the frame information of vehicle, that is, vehicle testing techniques can be used to adopting
The image of collection carries out vehicle detection, obtains the vehicle trim parameter of vehicle in the picture.
Here it is possible to which the direction vanishing point y coordinate is defined as one using vehicle trim parameter, camera mounting height as variable
(yt, h1, vh, a), wherein h is the direction vanishing point y coordinate, and yt is the direction the y coordinate of vehicle trim top edge, h1 for function, i.e. h=f
It is height of car (unit: pixel), vh is camera mounting height (unit: rice), and a is a parameter that can learn.Calculate the letter of h
Number f can have many design methods, such as a simple formula can be taken as by rule of thumb: h=yt+h1* (a/vh-1).Camera shooting
Head mounting height vh, can be arranged to fixed value, such as general car, camera can be installed according to the height of vehicle
High unity is set as 1.2 meters, and can rule of thumb or statistics or with simple fitting algorithm obtain one of a it is reasonable
Value, for example, corresponding to 1.2 meters of camera shooting grease head highness, a value can obtain preferable effect for 1.5 or so.Except before this, h=
(yt, h1, vh, a) formula can also be there are many kinds of design method, for example, that is, reality occurs by collecting a large amount of sampled data by f
Different (h, yt, h1, vh) combination, the method that various regression fits can be used, for example, linear fit, quadratic fit with
SVR etc., to obtain more accurately formula.
On the basis of horizontal where vanishing point in image has been determined, then by lane detection technology to the image of acquisition
Carry out lane detection, it is only necessary to detect a lane line, it can obtain with a lane line according to horizontal line where vanishing point
The position of vanishing point is taken, the position of the vanishing point is the intersection point position of horizontal line and lane line, as shown in Figure 9.
In this way, can fast and accurately obtain vanishing point position by the above method, use scope is wide, especially in existing side
Case can not obtain under the partial picture of vanishing point position, because of the severe feelings such as lane line blocks caused by traffic congestion or lane line is fuzzy
Condition determines to obtain two lane lines progress vanishing points, and can be also accurately obtained and be gone out using method provided by the present invention
Point position.
It is continuous that the software realization of the acquisition device of road vanishing point provided in an embodiment of the present invention is illustrated.Figure 11 is this hair
The composed structure schematic diagram of the acquisition device for the road vanishing point that bright embodiment provides, referring to Figure 11, road of the embodiment of the present invention goes out
The acquisition device 11 of point is applied to the first vehicle arrangement, comprising:
Acquisition unit 1101, for acquiring the image including the second vehicle arrangement；
First detection unit 1102 carries out lane detection for the image based on acquisition, obtains the first vehicle in image
The lane line information in lane locating for equipment；
Second detection unit 1103 carries out vehicle detection for the image based on acquisition, it is opposite to obtain the second vehicle arrangement
In the vehicle parameter of image；
Determination unit 1104 determines lane pair locating for the first vehicle arrangement for being based on lane line information and vehicle parameter
The vanishing point position answered.
In some embodiments, first detection unit 1102 obtain including the first vehicle for carrying out region interception to image
The image-region in lane locating for equipment；
Obtain the pixel for belonging to lane line in image-region；
By carrying out curve fitting to the pixel of acquisition, lane locating for the first vehicle arrangement in image-region is determined
Lane line.
In some embodiments, first detection unit 1102, for including at least left-lane line and the right side in response to lane line
Lane line, the pixel based on acquisition determine the higher lane line of integrality in left-lane line and right-lane line；
It carries out curve fitting to the pixel of the higher lane line of integrality, to determine the first vehicle arrangement in image-region
The lane line in locating lane.
In some embodiments, second detection unit 1103, the nerve net for obtaining the image input training of acquisition
Network model, identification obtain multiple frames of the second vehicle arrangement；
NMS processing is carried out to multiple frames that identification obtains, obtains the target frame of corresponding second vehicle arrangement；
Relative positional relationship based on target frame and image obtains vehicle of second vehicle arrangement relative to described image
Parameter.
In some embodiments, second detection unit 1103, for based on target frame upper side frame and lower frame scheming
Position as in, determines height of car of second vehicle arrangement relative to image；
The position of upper side frame based on target frame in the picture determines upper side frame relative to the first coordinate system side in image
To first distance, upper side frame is parallel with the first coordinate system direction.
In some embodiments, determination unit 1104, for determining vanishing point position relative to the first coordinate system side in image
To second distance；
Drafting is parallel with the first coordinate system direction in image, and spacing is the first straight line of second distance；
Obtain and draw the first position of the intersection point for stating first straight line and lane line in the picture, first position is used as described in
Vanishing point position.
In some embodiments, determination unit 1104, for determining vanishing point position relative in image by following relationship
The second distance in the first coordinate system direction:
Second distance be equal to height of car constant times and first distance and.
In some embodiments, determination unit 1104, in response to vehicle parameter include the second vehicle arrangement relative to
The height of car of image, second distance and vehicle based on height of car, vanishing point position relative to the first coordinate system direction in image
The positive correlation that height is in, determines second distance；
Drafting is parallel with the top edge of image, and spacing is the second straight line of second distance；
The intersection point second position in the picture for obtaining the second straight line and the lane line drawn, using the second position as
Vanishing point position.
A kind of storage medium for being stored with executable instruction of the offer of the embodiment of the present invention, wherein being stored with executable finger
It enables, when executable instruction is executed by processor, processor will be caused to execute road vanishing point provided in an embodiment of the present invention position
Acquisition methods, for example, the acquisition methods of road vanishing point position as shown in Figure 3.
In some embodiments, storage medium can be FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface and deposit
The memories such as reservoir, CD or CD-ROM；Be also possible to include one of above-mentioned memory or any combination various equipment.
In some embodiments, executable instruction can use program, software, software module, the form of script or code,
By any form of programming language (including compiling or interpretative code, or declaratively or process programming language) write, and its
It can be disposed by arbitrary form, including be deployed as independent program or be deployed as module, component, subroutine or be suitble to
Calculate other units used in environment.
As an example, executable instruction can with but not necessarily correspond to the file in file system, can be stored in
A part of the file of other programs or data is saved, for example, being stored in hypertext markup language (HTML, Hyper Text
Markup Language) in one or more scripts in document, it is stored in the single file for being exclusively used in discussed program
In, alternatively, being stored in multiple coordinated files (for example, the file for storing one or more modules, subprogram or code section).
As an example, executable instruction can be deployed as executing in a calculating equipment, or it is being located at one place
Multiple calculating equipment on execute, or, be distributed in multiple places and by multiple calculating equipment of interconnection of telecommunication network
Upper execution.
It include the by acquisition in conclusion the acquisition methods of the road vanishing point position provided through the embodiment of the present invention
The image of two vehicle arrangements determines that the lane line information in lane locating for the first vehicle arrangement and the second vehicle arrangement are opposite respectively
In the vehicle parameter of image, horizontal line where determining vanishing point in the picture according to obtained vehicle parameter is being based on lane line
The position of vanishing point is determined with intersection point horizontal where vanishing point.In this way, utilizing auxiliary information of vehicles and a lane line
Quickly accurate to determine vanishing point position, use scope is wide, enhances user experience.
The above, only the embodiment of the present invention, are not intended to limit the scope of the present invention.It is all in this hair
Made any modifications, equivalent replacements, and improvements etc. within bright spirit and scope, be all contained in protection scope of the present invention it
It is interior.
Claims (10)
1. a kind of acquisition methods of road vanishing point position, which is characterized in that be applied to the first vehicle arrangement, which comprises
Acquisition includes the image of the second vehicle arrangement；
Described image based on acquisition carries out lane detection, obtains lane locating for the first vehicle arrangement described in described image
Lane line information；
Described image based on acquisition carries out vehicle detection, and the vehicle for obtaining second vehicle arrangement relative to described image is joined
Number；
Based on the lane line information and the vehicle parameter, the corresponding vanishing point position in lane locating for first vehicle arrangement is determined
It sets.
2. the method as described in claim 1, which is characterized in that the described image based on acquisition carries out lane detection,
Obtain the lane line information in lane locating for the first vehicle arrangement described in described image, comprising:
To described image carry out region interception, obtain include lane locating for first vehicle arrangement image-region；
Obtain the pixel for belonging to lane line in described image region；
By carrying out curve fitting to the pixel of acquisition, determine locating for the first vehicle arrangement described in described image region
The lane line in lane.
3. method according to claim 2, which is characterized in that described quasi- by carrying out curve to the pixel of acquisition
It closes, determines the lane line in lane locating for the first vehicle arrangement described in described image region, comprising:
In response to the lane line include at least left-lane line and right-lane line, the pixel based on acquisition, determine described in
The higher lane line of integrality in left-lane line and right-lane line；
It carries out curve fitting to the pixel of the higher lane line of the integrality, to determine first described in described image region
The lane line in lane locating for vehicle arrangement.
4. the method as described in claim 1, which is characterized in that the described image based on acquisition carries out vehicle detection, obtains
Vehicle parameter to second vehicle arrangement relative to described image, comprising:
The neural network model that the described image input training of acquisition is obtained, identification obtain the multiple of second vehicle arrangement
Frame；
Non-maximum restraining NMS processing is carried out to the multiple frame that identification obtains, obtains corresponding to second vehicle arrangement
Target frame；
Based on the relative positional relationship of the target frame and described image, second vehicle arrangement is obtained relative to the figure
The vehicle parameter of picture.
5. method as claimed in claim 4, which is characterized in that the opposite position based on the target frame and described image
Relationship is set, vehicle parameter of second vehicle arrangement relative to described image is obtained, comprising:
The position of upper side frame and lower frame based on the target frame in described image, determines the second vehicle arrangement phase
For the height of car of described image；
Position of the upper side frame based on the target frame in described image, determines the upper side frame relative in described image
The first distance in the first coordinate system direction, the upper side frame are parallel with first coordinate system direction.
6. method described in claim 5, which is characterized in that it is described to be based on the vehicle parameter and the lane line information, really
The corresponding vanishing point position in lane locating for fixed first vehicle arrangement, comprising:
Determine second distance of the vanishing point position relative to the first coordinate system direction in described image；
Drafting is parallel with the first coordinate system direction in described image, and spacing is the first straight line of the second distance；
First position of the intersection point of the first straight line and the lane line drawn in described image is obtained, by described first
Position is as the vanishing point position.
7. method of claim 6, which is characterized in that the determining vanishing point position is relative to the first coordinate in described image
It is the second distance in direction, comprising:
Second distance of the vanishing point position relative to the first coordinate system direction in described image is determined by following relationship:
The second distance be equal to the height of car constant times and the first distance and.
8. the method as described in claim 1, which is characterized in that it is described to be based on the vehicle parameter and the lane line information,
Determine the corresponding vanishing point position in lane locating for first vehicle arrangement, comprising:
Include height of car of second vehicle arrangement relative to described image in response to the vehicle parameter, is based on the vehicle
Height, the vanishing point position are in relative to the second distance with the height of car in the first coordinate system direction in described image
Positive correlation, determine the second distance；
Drafting is parallel with the top edge of described image, and spacing is the second straight line of the second distance；
The second position of the intersection point of the second straight line and the lane line drawn in described image is obtained, by described second
Position is as the vanishing point position.
9. a kind of acquisition device of road vanishing point position, which is characterized in that be applied to the first vehicle arrangement, described device includes:
Acquisition unit, for acquiring the image including the second vehicle arrangement；
First detection unit carries out lane detection for the described image based on acquisition, obtains first described in described image
The lane line information in lane locating for vehicle arrangement；
Second detection unit carries out vehicle detection for the described image based on acquisition, it is opposite to obtain second vehicle arrangement
In the vehicle parameter of described image；
Determination unit determines vehicle locating for first vehicle arrangement for being based on the lane line information and the vehicle parameter
The corresponding vanishing point position in road.
10. a kind of acquisition device of road vanishing point position characterized by comprising
Memory, for storing executable instruction；
Processor when for executing the executable instruction stored in the memory, is realized described in any one of claim 1 to 9
The acquisition methods of road vanishing point position.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910502866.5A CN110222658A (en) | 2019-06-11 | 2019-06-11 | The acquisition methods and device of road vanishing point position |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910502866.5A CN110222658A (en) | 2019-06-11 | 2019-06-11 | The acquisition methods and device of road vanishing point position |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110222658A true CN110222658A (en) | 2019-09-10 |
Family
ID=67816606
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910502866.5A Pending CN110222658A (en) | 2019-06-11 | 2019-06-11 | The acquisition methods and device of road vanishing point position |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110222658A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111126306A (en) * | 2019-12-26 | 2020-05-08 | 江苏罗思韦尔电气有限公司 | Lane line detection method based on edge features and sliding window |
| CN111174796A (en) * | 2019-12-31 | 2020-05-19 | 驭势科技(南京)有限公司 | Navigation method based on single vanishing point, electronic equipment and storage medium |
| CN112101321A (en) * | 2020-11-18 | 2020-12-18 | 蘑菇车联信息科技有限公司 | Vanishing point extraction method and device, electronic equipment and storage medium |
| CN112150559A (en) * | 2020-09-24 | 2020-12-29 | 深圳佑驾创新科技有限公司 | Calibration method of image acquisition device, computer equipment and storage medium |
| CN112199999A (en) * | 2020-09-09 | 2021-01-08 | 浙江大华技术股份有限公司 | Road detection method, road detection device, storage medium and electronic equipment |
| CN117036505A (en) * | 2023-08-23 | 2023-11-10 | 长和有盈电子科技(深圳)有限公司 | On-line calibration method and system for vehicle-mounted camera |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20110001427A (en) * | 2009-06-30 | 2011-01-06 | 태성전장주식회사 | High speed road lane detection method based on extraction of roi-lb |
| WO2012011713A2 (en) * | 2010-07-19 | 2012-01-26 | 주식회사 이미지넥스트 | System and method for traffic lane recognition |
| US20140241579A1 (en) * | 2013-02-28 | 2014-08-28 | Fujifilm Corporation | Vehicle-to-vehicle distance calculation apparatus and method |
| CN106682563A (en) * | 2015-11-05 | 2017-05-17 | 腾讯科技（深圳）有限公司 | Lane line detection self-adaptive adjusting method and device |
| CN108230254A (en) * | 2017-08-31 | 2018-06-29 | 北京同方软件股份有限公司 | A kind of full lane line automatic testing method of the high-speed transit of adaptive scene switching |
| CN109685858A (en) * | 2018-12-29 | 2019-04-26 | 北京茵沃汽车科技有限公司 | A kind of monocular cam online calibration method |
-
2019
- 2019-06-11 CN CN201910502866.5A patent/CN110222658A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20110001427A (en) * | 2009-06-30 | 2011-01-06 | 태성전장주식회사 | High speed road lane detection method based on extraction of roi-lb |
| WO2012011713A2 (en) * | 2010-07-19 | 2012-01-26 | 주식회사 이미지넥스트 | System and method for traffic lane recognition |
| US20140241579A1 (en) * | 2013-02-28 | 2014-08-28 | Fujifilm Corporation | Vehicle-to-vehicle distance calculation apparatus and method |
| CN106682563A (en) * | 2015-11-05 | 2017-05-17 | 腾讯科技（深圳）有限公司 | Lane line detection self-adaptive adjusting method and device |
| CN108230254A (en) * | 2017-08-31 | 2018-06-29 | 北京同方软件股份有限公司 | A kind of full lane line automatic testing method of the high-speed transit of adaptive scene switching |
| CN109685858A (en) * | 2018-12-29 | 2019-04-26 | 北京茵沃汽车科技有限公司 | A kind of monocular cam online calibration method |
Non-Patent Citations (1)
| Title |
|---|
| 郭子逸;: "一种基于单目视觉的车道偏离检测与预警方法", 机械制造, no. 05 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111126306A (en) * | 2019-12-26 | 2020-05-08 | 江苏罗思韦尔电气有限公司 | Lane line detection method based on edge features and sliding window |
| CN111174796A (en) * | 2019-12-31 | 2020-05-19 | 驭势科技(南京)有限公司 | Navigation method based on single vanishing point, electronic equipment and storage medium |
| CN111174796B (en) * | 2019-12-31 | 2022-04-29 | 驭势科技（浙江）有限公司 | Navigation method based on single vanishing point, electronic equipment and storage medium |
| CN112199999A (en) * | 2020-09-09 | 2021-01-08 | 浙江大华技术股份有限公司 | Road detection method, road detection device, storage medium and electronic equipment |
| CN112150559A (en) * | 2020-09-24 | 2020-12-29 | 深圳佑驾创新科技有限公司 | Calibration method of image acquisition device, computer equipment and storage medium |
| CN112101321A (en) * | 2020-11-18 | 2020-12-18 | 蘑菇车联信息科技有限公司 | Vanishing point extraction method and device, electronic equipment and storage medium |
| CN112101321B (en) * | 2020-11-18 | 2021-02-02 | 蘑菇车联信息科技有限公司 | Vanishing point extraction method and device, electronic equipment and storage medium |
| CN117036505A (en) * | 2023-08-23 | 2023-11-10 | 长和有盈电子科技(深圳)有限公司 | On-line calibration method and system for vehicle-mounted camera |
| CN117036505B (en) * | 2023-08-23 | 2024-03-29 | 长和有盈电子科技(深圳)有限公司 | On-line calibration method and system for vehicle-mounted camera |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110222658A (en) | The acquisition methods and device of road vanishing point position | |
| CN107341453B (en) | Lane line extraction method and device | |
| JP2019109245A (en) | Visual reinforcement navigation | |
| CN110386142A (en) | Pitch angle calibration method for automatic driving vehicle | |
| CN110753892A (en) | Method and system for instant object tagging via cross-modality verification in autonomous vehicles | |
| CN108875600A (en) | A kind of information of vehicles detection and tracking method, apparatus and computer storage medium based on YOLO | |
| CN102609682B (en) | Feedback pedestrian detection method for region of interest | |
| US9607220B1 (en) | Image-based vehicle speed estimation | |
| CN110869936A (en) | Method and system for distributed learning and adaptation in autonomous vehicles | |
| CN110799982A (en) | Method and system for object-centric stereo vision in an autonomous vehicle | |
| CN105513038A (en) | Image matching method and mobile phone application test platform | |
| CN106558051A (en) | A kind of improved method for detecting road from single image | |
| CN109584300A (en) | A kind of method and device of determining headstock towards angle | |
| CN105608417A (en) | Traffic signal lamp detection method and device | |
| DE102014204320A1 (en) | Information query by pointing | |
| CN105956632A (en) | Target detection method and device | |
| CN110967018B (en) | Parking lot positioning method and device, electronic equipment and computer readable medium | |
| EP3710780A2 (en) | Method and system for enhanced sensing capabilities for vehicles | |
| CN111400423B (en) | Smart city CIM three-dimensional vehicle pose modeling system based on multi-view geometry | |
| CN112525147A (en) | Distance measurement method for automatic driving equipment and related device | |
| CN113743163A (en) | Traffic target recognition model training method, traffic target positioning method and device | |
| CN109740502B (en) | Road quality detection method and device | |
| CN114897987B (en) | Method, device, equipment and medium for determining vehicle ground projection | |
| Shen et al. | A real-time siamese tracker deployed on UAVs | |
| JP5773334B2 (en) | Optical flow processing apparatus and display radius map generation apparatus |
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 |