CN109960959A - Method and apparatus for handling image - Google Patents
Method and apparatus for handling image Download PDFInfo
- Publication number
- CN109960959A CN109960959A CN201711337034.XA CN201711337034A CN109960959A CN 109960959 A CN109960959 A CN 109960959A CN 201711337034 A CN201711337034 A CN 201711337034A CN 109960959 A CN109960959 A CN 109960959A
- Authority
- CN
- China
- Prior art keywords
- lane line
- line
- location information
- target image
- image
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
-
- 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)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the present application discloses the method and apparatus for handling image.One specific embodiment of this method includes: the color of each pixel in multiple pixels based on target image, in multiple pixel in the location information of the target image, determines the location information of multiple pixels where at least one lane line;Linear fit is carried out to the location information of multiple pixels where at least one lane line, to determine the location information of at least two positions on the line segment where the target image, at least one lane line each lane line;Based on the location information of at least two positions on the line segment where each lane line, the region at each lane line place in the target image is determined, and determine the line style of lane line in each region.The embodiment of the present application improves the accuracy of identification lane line.
Description
Technical field
The invention relates to field of computer technology, and in particular to Internet technical field is more particularly, to located
The method and apparatus for managing image.
Background technique
Lane line is the index line being most commonly encountered during vehicle is travelled on highway, can all be drawn on nearly all road
There is lane line.The information for obtaining lane line can indicate driver, to preferably guarantee the traffic safety of vehicle.
Summary of the invention
The embodiment of the present application proposes the method and apparatus for handling image.
In a first aspect, the embodiment of the present application provides a kind of method for handling image, comprising: based on target image
The color of each pixel in multiple pixels determines at least one in location information of multiple pixels in target image
The location information of multiple pixels where lane line;To the location informations of multiple pixels where at least one lane line into
Row linear fit, to determine on the line segment where each lane line in target image, at least one lane line at least
The location information of two positions;Based on the location information of at least two positions on the line segment where each lane line, determine each
Lane line in the target image where region, and determine the line style of lane line in each region.
In some embodiments, Linear Quasi is carried out to the location information of multiple pixels where at least one lane line
It closes, to determine at least two positions on the line segment where each lane line in target image, at least one lane line
Location information, comprising: linear fit is carried out to the location informations of multiple pixels where at least one lane line, obtain to
The function of the line segment where each lane line in a few lane line;For every lane line, determine target image, meet
At least two location informations of the function of the line segment where the lane line.
In some embodiments, in multiple pixels based on target image each pixel color, in multiple pixels
Point determines the location information of multiple pixels where at least one lane line in the location information of target image, comprising: will
The location information input of the multiple pixels and multiple pixels of target image in the target image color classification trained in advance
Model, wherein face of the color classification model to belong to lane line to the pixel progress color classification and output that include color
The pixel and location information of color classification;At least one colour type that obtain the output of color classification model, belonging to lane line
In multiple pixels of each colour type and multiple pixels location information in the target image of the colour type.
In some embodiments, the location information of at least two positions is the position of two endpoints of the line segment where lane line
Confidence breath and the location information at the midpoint for two broadsides in region, the region where each lane line are rectangle;And it is based on
The location information of at least two positions on line segment where each lane line, where determining each lane line in the target image
Region, comprising: for every lane line, the midpoint of two broadsides based on the region where the line segment where the lane line
Location information, in the target image determine width be predetermined width and length be two endpoint lines length rectangular area.
In some embodiments, the line style of lane line in each region is determined, comprising: by the region where each lane line
Input line style disaggregated model trained in advance, obtains the line for the lane line for including in each region of line style disaggregated model output
Type, wherein line style disaggregated model is classified to the line style to lane line.
In some embodiments, method further include: straight-line detection and/or frequency domain detection are carried out to the image formerly obtained,
It determines the subgraph where the lane line of acquired image, and subgraph is determined as target image.
Second aspect, the embodiment of the present application provide a kind of for handling the device of image, comprising: determination unit, configuration
For the color of each pixel in multiple pixels based on target image, believe in multiple pixels in the position of target image
In breath, the location information of multiple pixels where at least one lane line is determined;Fitting unit is configured to at least one
The location information of multiple pixels where lane line carries out linear fit, to determine in target image, at least one lane
The location information of at least two positions on the line segment where each lane line in line;Line style determination unit, is configured to base
Location information at least two positions on the line segment where each lane line determines each lane line institute in the target image
Region, and determine each region in lane line line style.
In some embodiments, fitting unit, comprising: fitting module is configured to where at least one lane line
The location information of multiple pixels carries out linear fit, obtains the line segment where each lane line at least one lane line
Function;Determining module, is configured to for every lane line, determine target image, meet line segment where the lane line
At least two location informations of function.
In some embodiments, determination unit, comprising: input module is configured to multiple pixels of target image
The color classification model trained in advance with the location information input of multiple pixels in the target image, wherein color classification mould
The pixel of colour type of the type to belong to lane line to the pixel progress color classification and output that include color and position
Information;Output module is configured to every at least one colour type that obtain the output of color classification model, belonging to lane line
The location information of multiple pixels of multiple pixels of a colour type and the colour type in the target image.
In some embodiments, the location information of at least two positions is the position of two endpoints of the line segment where lane line
Confidence breath and the location information at the midpoint for two broadsides in region, the region where each lane line are rectangle;And line style
Determination unit is further configured to: for every lane line, two based on the region where the line segment where the lane line
The location information at the midpoint of broadside determines that width is predetermined width and length is the length of two endpoint lines in the target image
Rectangular area.
In some embodiments, line style determination unit is further configured to: the region where each lane line is inputted
Trained line style disaggregated model in advance obtains the line style for the lane line for including in each region of line style disaggregated model output,
In, line style disaggregated model is classified to the line style to lane line.
In some embodiments, device further include: pretreatment unit is configured to carry out the image formerly obtained straight
Line detection and/or frequency domain detection, determine the subgraph where the lane line of acquired image, and subgraph is determined as target figure
Picture.
The third aspect, the embodiment of the present application provide a kind of electronic equipment, comprising: one or more processors;Storage dress
It sets, for storing one or more programs, when one or more programs are executed by one or more processors, so that one or more
A processor realizes the method such as any embodiment in the method for handling image.
Fourth aspect, the embodiment of the present application provide a kind of computer readable storage medium, are stored thereon with computer journey
Sequence realizes the method such as any embodiment in the method for handling image when the program is executed by processor.
Method and apparatus provided by the embodiments of the present application for handling image, firstly, multiple pictures based on target image
The color of each pixel in vegetarian refreshments determines at least one lane line in location information of multiple pixels in target image
The location information of multiple pixels at place.Later, to the location information of multiple pixels where at least one lane line into
Row linear fit, to determine on the line segment where each lane line in target image, at least one lane line at least
The location information of two positions.Finally, the location information based at least two positions on the line segment where each lane line, really
Region where determining each lane line in the target image, and determine the line style of lane line in each region.The embodiment of the present application
Improve the accuracy of identification lane line.
Detailed description of the invention
By reading a detailed description of non-restrictive embodiments in the light of the attached drawings below, the application's is other
Feature, objects and advantages will become more apparent upon:
Fig. 1 is that this application can be applied to exemplary system architecture figures therein;
Fig. 2 is the flow chart according to one embodiment of the method for handling image of the application;
Fig. 3 is the schematic diagram according to an application scenarios of the method for handling image of the application;
Fig. 4 is the flow chart according to another embodiment of the method for handling image of the application;
Fig. 5 is the structural schematic diagram according to one embodiment of the device for handling image of the application;
Fig. 6 is adapted for the structural schematic diagram for the computer system for realizing the electronic equipment of the embodiment of the present application.
Specific embodiment
The application is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining related invention, rather than the restriction to the invention.It also should be noted that in order to
Convenient for description, part relevant to related invention is illustrated only in attached drawing.
It should be noted that in the absence of conflict, the features in the embodiments and the embodiments of the present application can phase
Mutually combination.The application is described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
Fig. 1 is shown can be using the method for handling image of the application or the implementation of the device for handling image
The exemplary system architecture 100 of example.
As shown in Figure 1, system architecture 100 may include terminal device 101,102,103, network 104 and server 105.
Network 104 between terminal device 101,102,103 and server 105 to provide the medium of communication link.Network 104 can be with
Including various connection types, such as wired, wireless communication link or fiber optic cables etc..
User can be used terminal device 101,102,103 and be interacted by network 104 with server 105, to receive or send out
Send message etc..Various telecommunication customer end applications can be installed on terminal device 101,102,103, such as automobile navigation application,
Web browser applications, shopping class application, searching class application, instant messaging tools, mailbox client, social platform software etc..
Terminal device 101,102,103 can be with display screen and communicate to connect various electronic equipments, including but
It is not limited to vehicle, Vehicular navigation system, smart phone, tablet computer, pocket computer on knee and desktop computer etc..
Server 105 can be to provide the server of various services, such as to showing on terminal device 101,102,103
The line style of lane line provides the background server supported.Background server can divide the data such as the target image received
The processing such as analysis, and the processing result line style of lane line (such as in target image) is fed back into terminal device.
It should be noted that the method provided by the embodiment of the present application for handling image is generally held by server 105
Row, correspondingly, the device for handling image is generally positioned in server 105.
It should be understood that the number of terminal device, network and server in Fig. 1 is only schematical.According to realization need
It wants, can have any number of terminal device, network and server.
With continued reference to Fig. 2, the process of one embodiment of the method for handling image according to the application is shown
200.The method for being used to handle image, comprising the following steps:
Step 201, in multiple pixels based on target image each pixel color, in multiple pixels in target
In the location information of image, the location information of multiple pixels where at least one lane line is determined.
In the present embodiment, the method for handling image runs electronic equipment (such as service shown in FIG. 1 thereon
Device) color of each pixel in multiple pixels based on target image determines extremely in the location information of multiple pixels
The location information of multiple pixels where a few lane line.Target image can be to be obtained from local or other electronic equipments
The image taken.Specifically, target image can be the top view and front view that road pavement is shot.Here more preferable in order to obtain
Image analysis result, the target image of use can be high-resolution image.Here the location information of pixel be in
In target image.There are multiple pixels in target image, each pixel has its location information in the target image.
Specifically, location information can be indicated in the form of coordinate.Lane line is the line drawn on vehicle driving road
Item indicates the traveling-position of vehicle.May exist one or more than one lane line in the picture.In practice, Ke Yicong
The pixel where lane line is determined in multiple pixels of target image, and then determines the position letter of lane line place pixel
Breath.It, can be by carrying out face to each pixel because the color of lane line and other colors (such as road surface background color) have difference
Color identification, and then determine the pixel where lane line.By the color of preset lane line, then it can determine and belong to vehicle
The pixel of the color of diatom.And these pixels for belonging to lane line color are then the pixels where lane line, Jin Erke
To determine the location information of these pixels as the location information of the pixel where lane line.
Step 202, linear fit is carried out to the location information of multiple pixels where at least one lane line, with determination
Believe the position of at least two positions on the line segment where target image, each lane line at least one lane line
Breath.
In the present embodiment, above-mentioned electronic equipment believes the position of multiple pixels where above-mentioned at least one lane line
Breath carries out linear fit, at least two on the line segment where each lane line in above-mentioned at least one lane line of determination
The location information set.Here location information is the location information in target image, that is, indicated by the two location informations
Position is fallen in the target image.By linear fit, at least one line segment where available lane line, while also just obtaining
Line segment quantity present at least one line segment.The quantity of line segment can be determined as to the quantity of lane line.
In practice, the line segment being fitted can be using the location information of at least two positions fallen on line segment come table
Show, such as coordinate (x1, y1) and (x2, y2), it can also be indicated using the function of line segment.
Step 203, the location information based at least two positions on the line segment where each lane line, determines each vehicle
Diatom in the target image where region, and determine each region in lane line line style.
In the present embodiment, position of the above-mentioned electronic equipment based at least two positions on the line segment where each lane line
Confidence breath, the region where determining each lane line in above-mentioned target image.Later in identified each region, determine
The line style of lane line.Here the location information of at least two positions plays positioning action to lane line, can by location information
To navigate to lane line in the target image, in order to the line style of subsequent determining lane line.The line style of lane line refers to lane line institute
The pattern of presentation.For example, solid line, dotted line etc..
Specifically, it is determined that each lane line in the target image where region can use various ways.For example, can be with
Using the line of the location information of this at least two position as the center line in region, the area of specified width, which width is taken in the two sides of center line
Domain.Two parallel lines of the line segment where the location information of at least two positions can also be done, this two parallel lines distances are at least
The distance of line segment where the location information of two positions is distance to a declared goal.It later can be by the region between this two parallel lines
As the region where lane line.
After the region where each lane line has been determined, then the line style of lane line in each region can be determined.It can
Lane line in each region to be compared with the Standard linear figure of lane line.The standard that will be matched with lane line in region
Line style of the line style as the region lane line in line illustration.It can determine that the two matches when similarity is higher than threshold value.This
Outside, can also be sorted out using line style of the disaggregated model to lane line, to determine the line style of lane line in region.
With continued reference to the signal that Fig. 3, Fig. 3 are according to the application scenarios of the method for handling image of the present embodiment
Figure.In the application scenarios of Fig. 3, color of the electronic equipment 301 based on each pixel in multiple pixels of image 302,
Multiple pixels determine the location information of multiple pixels where at least one lane line in the location information of image 302
303.Linear fit is carried out to the location information of multiple pixels where at least one lane line, to determine in target image
, the location informations 304 of 2 positions on line segment where each lane line in 3 lane lines.Based on each lane line institute
Line segment on 2 positions location information 304, determine each lane line in the target image where region 305, and really
Determine the line style 306 of lane line in each region.
The method provided by the above embodiment of the application improves the accuracy of identification lane line, can determine lane line
Number and line style.
With further reference to Fig. 4, it illustrates the processes 400 of another embodiment of the method for handling image.The use
In the process 400 of the method for processing image, comprising the following steps:
Step 401, straight-line detection and/or frequency domain detection are carried out to the image formerly obtained, determines the vehicle of acquired image
Subgraph where diatom, and subgraph is determined as target image.
In the present embodiment, electronic equipment carries out straight-line detection and/or frequency domain detection to the image formerly obtained, with determination
Subgraph where the lane line of acquired image, and subgraph is determined as target image.Subgraph is the figure formerly obtained
A part as in.Here image is usually front view or top view.Straight line inspection can be carried out to front view and top view
It surveys to detect straight line in the picture, then determines the position where straight line, image is split by the position.Such as it can
To be detected using Hough transformation.Frequency domain detection can also be used, i.e., high-frequency signal and low frequency signal in detection image are low
Lane line (such as sky in front view) is not present in the corresponding topography of frequency signal, then the topography is not being generated
Subgraph in.In addition it is also possible to carry out above two detection to image, to obtain better detection effect, more accurately really
Sub-image.
Step 402, the location information of the multiple pixels and multiple pixels of target image in the target image is inputted
Trained color classification model in advance.
In the present embodiment, above-mentioned electronic equipment is by multiple pixels of target image and above-mentioned multiple pixels in target
The color classification model that location information input in image is trained in advance.Above-mentioned color classification model is to the picture including color
Vegetarian refreshments carries out color classification and exports the pixel and location information of each colour type.Specifically, color classification model can be with
It is two disaggregated models, for example, the color of pixel can be divided into road surface background color and non-road surface background color.Color classification model can also
To be three disaggregated models etc..For example, the color of pixel can be divided into road surface background color, white and yellow, white therein
It is the color of lane line with yellow.
Specifically, specifically, it can use the sample of various colors that is a large amount of, being labelled with colour type to color
Disaggregated model is trained.Color classification model can be by support vector machines (Support Vector Machine, SVM),
The training of the classifiers (Classifier) such as model-naive Bayesian (Naive Bayesian Model, NBM) model obtains.This
Outside, color classification model is also possible to based on made of certain classification functions (such as softmax function etc.) in advance training.
Step 403, each color at least one colour type that obtain the output of color classification model, belonging to lane line
The location information of multiple pixels of multiple pixels of classification and the colour type in the target image.
In the present embodiment, above-mentioned electronic equipment obtains belonging to lane line at least by what above-mentioned color classification model exported
Multiple pixels of each colour type and multiple pixels of each colour type are in target image in one colour type
In location information.Above-mentioned color classification model outputs the pixel of each colour type, and outputs each color class
The location information of other pixel.There are the colour types of lane line in each classification of output.It can be to the difference of output
The pixel of colour type and the location information of pixel be distinguish using different labels, in order to which determination belongs to lane
The pixel and location information of the colour type of line.
Step 404, linear fit is carried out to the location informations of multiple pixels where at least one lane line, obtain to
The function of the line segment where each lane line in a few lane line.
In the present embodiment, above-mentioned electronic equipment believes the position of multiple pixels where above-mentioned at least one lane line
Breath carries out linear fit.By the function for being fitted the line segment where obtaining each lane line at least one above-mentioned lane line.
Here the function obtained is linear linear function, indicates the line segment where each lane line.Variable in function can be deposited
In value range, so that the coordinate points in function are fallen in a line segment, while also falling in the target image.
Step 405, for every lane line, function determining target image, meeting line segment where the lane line
At least two location informations.
In the present embodiment, for every lane line, above-mentioned electronic equipment is determined for compliance with the line segment where the lane line
At least two location informations of function, while the location information of at least two positions here is also the point in target image.
Step 406, for every lane line, in two broadsides based on the region where the line segment where the lane line
Point location information, in the target image determine width be predetermined width and length be two endpoint lines length rectangle region
Domain.
In the present embodiment, the location information of at least two positions is the position of two endpoints of the line segment where lane line
The location information at information and the midpoint for two broadsides in region, the region where each lane line are rectangle.For every vehicle
Diatom, the location information at midpoint of the above-mentioned electronic equipment based on two broadsides in region determine rectangular area in the target image.This
In region be region where line segment where the region that will be determined and this lane line.The width of rectangular area is
Predetermined width, length are the length of two endpoint lines of the line segment where the lane line.Obtain the broadside of rectangular area two
Midpoint location information, it can rectangular area is positioned.By the length and width of rectangular area, then can determine
The size of rectangle.Each rectangle has long and width, two broadsides here to refer to two opposite side at the wide place of rectangle.
Step 407, the line style disaggregated model that the region input where each lane line is trained in advance, obtains line style classification
The line style for the lane line for including in each region of model output.
In the present embodiment, above-mentioned electronic equipment classifies the line style trained in advance of the region input where each lane line
Model obtains the line style for the lane line for including in each region of above-mentioned line style disaggregated model output.Wherein, line style disaggregated model
Classify to the line style to lane line.Above-mentioned electronic equipment can be subject to area to different lane line line styles by the model
Point, so that it is determined that lane line line style belonging to the lane line in region out.
Specifically, it can use the sample of various lane line line styles that are a large amount of, being labelled with lane line line style classification
Line style disaggregated model is trained.Line style disaggregated model can be by support vector machines (Support Vector Machine,
SVM), the classifiers (Classifier) such as model-naive Bayesian (Naive Bayesian Model, NBM) model training obtains
's.In addition, line style disaggregated model is also possible to, based on certain classification functions (such as softmax function etc.), training is formed in advance
's.
The present embodiment utilizes color classification model, can determine the color of lane line, improves the accurate of Lane detection
Degree.Meanwhile the present embodiment reduces the time-consuming of Lane detection, together by the subgraph where obtaining lane line in image
When further improve the accuracy of Lane detection.
With further reference to Fig. 5, as the realization to method shown in above-mentioned each figure, this application provides one kind for handling figure
One embodiment of the device of picture, the Installation practice is corresponding with embodiment of the method shown in Fig. 2, which can specifically answer
For in various electronic equipments.
As shown in figure 5, the device 500 for handling image of the present embodiment comprises determining that unit 501, fitting unit 502
With line style determination unit 503.Wherein it is determined that unit 501, is configured to each pixel in multiple pixels based on target image
The color of point determines multiple pixels where at least one lane line in location information of multiple pixels in target image
The location information of point.Fitting unit 502, be configured to the location informations of multiple pixels where at least one lane line into
Row linear fit, to determine on the line segment where each lane line in target image, at least one lane line at least
The location information of two positions.Line style determination unit 503 is configured to based at least two on the line segment where each lane line
The location information of a position, the region where determining each lane line in the target image, and determine lane line in each region
Line style.
It in the present embodiment, can be based on the multiple of target image for handling the determination unit 501 of the device 500 of image
The color of each pixel in pixel determines more where at least one lane line in the location information of multiple pixels
The location information of a pixel.Target image can be the image obtained from local or other electronic equipments.Specifically, target
Image can be the top view and front view that road pavement is shot.Here in order to obtain better image analysis as a result, using
Target image can be high-resolution image.Here the location information of pixel is in target image.Target figure
There are multiple pixels as in, each pixel has its location information in the target image.
In the present embodiment, fitting unit 502 believes the position of multiple pixels where above-mentioned at least one lane line
Breath carries out linear fit, at least two on the line segment where each lane line in above-mentioned at least one lane line of determination
The location information set.Here location information is the location information in target image, that is, indicated by the two location informations
Position is fallen in the target image.By linear fit, at least one line segment where available lane line, while also just obtaining
Line segment quantity present at least one line segment.The quantity of line segment can be determined as to the quantity of lane line.
In the present embodiment, line style determination unit 503 is based at least two positions on the line segment where each lane line
Location information, determine each lane line in above-mentioned target image where region.Later in identified each region,
Determine the line style of lane line.Here the location information of at least two positions plays positioning action to lane line, is believed by position
Breath, can navigate to lane line, in the target image in order to the line style of subsequent determining lane line.The line style of lane line refers to lane
The pattern that line is presented.For example, solid line, dotted line etc..
In some optional implementations of the present embodiment, fitting unit, comprising: fitting module is configured to extremely
The location information of multiple pixels where a few lane line carries out linear fit, obtains each item at least one lane line
The function of line segment where lane line;Determining module, is configured to for every lane line, determine target image, meet this
At least two location informations of the function of the line segment where lane line.
In some optional implementations of the present embodiment, determination unit, comprising: input module is configured to mesh
The location information input of the multiple pixels and multiple pixels of logo image in the target image color classification mould trained in advance
Type, wherein color of the color classification model to belong to lane line to the pixel progress color classification and output that include color
The pixel and location information of classification;Output module, be configured to obtain the output of color classification model, belong to lane line extremely
Multiple pixels of multiple pixels of each colour type and the colour type are in the target image in a few colour type
Location information.
In some optional implementations of the present embodiment, the location information of at least two positions is where lane line
The location information of two endpoints of line segment and the location information at the midpoint for two broadsides in region, the area where each lane line
Domain is rectangle;And line style determination unit is further configured to: for every lane line, based on the line segment where the lane line
The location information at the midpoint of two broadsides in the region at place determines that width is predetermined width and length is two in the target image
The rectangular area of the length of a endpoint line.
In some optional implementations of the present embodiment, line style determination unit is further configured to: by each vehicle
Region input where diatom line style disaggregated model trained in advance, obtain include in each region of line style disaggregated model output
Lane line line style, wherein line style disaggregated model is classified to the line style to lane line.
In some optional implementations of the present embodiment, device further include: pretreatment unit is configured to first
The image of acquisition carries out straight-line detection and/or frequency domain detection, determines the subgraph where the lane line of acquired image, and will be sub
Image is determined as target image.
In some optional implementations of the present embodiment, fitting unit, comprising: fitting module is configured to extremely
The location information of multiple pixels where a few lane line carries out linear fit, obtains each item at least one lane line
The function of line segment where lane line;Determining module, is configured to for every lane line, determine target image, meet this
At least two location informations of the function of the line segment where lane line.
In some optional implementations of the present embodiment, determination unit, comprising: input module is configured to mesh
The location information input of the multiple pixels and multiple pixels of logo image in the target image color classification mould trained in advance
Type, wherein color of the color classification model to belong to lane line to the pixel progress color classification and output that include color
The pixel and location information of classification;Output module, be configured to obtain the output of color classification model, belong to lane line extremely
Multiple pixels of multiple pixels of each colour type and the colour type are in the target image in a few colour type
Location information.
In some optional implementations of the present embodiment, the location information of at least two positions is where lane line
The location information of two endpoints of line segment and the location information at the midpoint for two broadsides in region, the area where each lane line
Domain is rectangle;And line style determination unit is further configured to: for every lane line, based on the line segment where the lane line
The location information at the midpoint of two broadsides in the region at place determines that width is predetermined width and length is two in the target image
The rectangular area of the length of a endpoint line.
In some optional implementations of the present embodiment, line style determination unit is further configured to: by each vehicle
Region input where diatom line style disaggregated model trained in advance, obtain include in each region of line style disaggregated model output
Lane line line style, wherein line style disaggregated model is classified to the line style to lane line.
In some optional implementations of the present embodiment, the device further include: pretreatment unit, be configured to
The image first obtained carries out straight-line detection and/or frequency domain detection, determines the subgraph where the lane line of acquired image, and will
Subgraph is determined as target image.
Below with reference to Fig. 6, it illustrates the computer systems 600 for the electronic equipment for being suitable for being used to realize the embodiment of the present application
Structural schematic diagram.Electronic equipment shown in Fig. 6 is only an example, function to the embodiment of the present application and should not use model
Shroud carrys out any restrictions.
As shown in fig. 6, computer system 600 includes central processing unit (CPU) 601, it can be read-only according to being stored in
Program in memory (ROM) 602 or be loaded into the program in random access storage device (RAM) 603 from storage section 608 and
Execute various movements appropriate and processing.In RAM 603, also it is stored with system 600 and operates required various programs and data.
CPU 601, ROM 602 and RAM 603 are connected with each other by bus 604.Input/output (I/O) interface 605 is also connected to always
Line 604.
I/O interface 605 is connected to lower component: the importation 606 including keyboard, mouse etc.;Including such as liquid crystal
Show the output par, c 607 of device (LCD) etc. and loudspeaker etc.;Storage section 608 including hard disk etc.;And including such as LAN
The communications portion 609 of the network interface card of card, modem etc..Communications portion 609 is executed via the network of such as internet
Communication process.Driver 610 is also connected to I/O interface 605 as needed.Detachable media 611, such as disk, CD, magneto-optic
Disk, semiconductor memory etc. are mounted on as needed on driver 610, in order to from the computer program root read thereon
According to needing to be mounted into storage section 608.
Particularly, in accordance with an embodiment of the present disclosure, it may be implemented as computer above with reference to the process of flow chart description
Software program.For example, embodiment of the disclosure includes a kind of computer program product comprising be carried on computer-readable medium
On computer program, which includes the program code for method shown in execution flow chart.In such reality
It applies in example, which can be downloaded and installed from network by communications portion 609, and/or from detachable media
611 are mounted.When the computer program is executed by central processing unit (CPU) 601, limited in execution the present processes
Above-mentioned function.It should be noted that the computer-readable medium of the application can be computer-readable signal media or calculating
Machine readable storage medium storing program for executing either the two any combination.Computer readable storage medium for example can be --- but it is unlimited
In system, device or the device of --- electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor, or any above combination.It calculates
The more specific example of machine readable storage medium storing program for executing can include but is not limited to: have the electrical connection, portable of one or more conducting wires
Formula computer disk, hard disk, random access storage device (RAM), read-only memory (ROM), erasable programmable read only memory
(EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory device or
The above-mentioned any appropriate combination of person.In this application, computer readable storage medium can be it is any include or storage program
Tangible medium, which can be commanded execution system, device or device use or in connection.And in this Shen
Please in, computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal,
In carry computer-readable program code.The data-signal of this propagation can take various forms, including but not limited to
Electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be computer-readable
Any computer-readable medium other than storage medium, the computer-readable medium can send, propagate or transmit for by
Instruction execution system, device or device use or program in connection.The journey for including on computer-readable medium
Sequence code can transmit with any suitable medium, including but not limited to: wireless, electric wire, optical cable, RF etc. are above-mentioned
Any appropriate combination.
Flow chart and block diagram in attached drawing are illustrated according to the system of the various embodiments of the application, method and computer journey
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part of one module, program segment or code of table, a part of the module, program segment or code include one or more use
The executable instruction of the logic function as defined in realizing.It should also be noted that in some implementations as replacements, being marked in box
The function of note can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are actually
It can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it to infuse
Meaning, the combination of each box in block diagram and or flow chart and the box in block diagram and or flow chart can be with holding
The dedicated hardware based system of functions or operations as defined in row is realized, or can use specialized hardware and computer instruction
Combination realize.
Being described in unit involved in the embodiment of the present application can be realized by way of software, can also be by hard
The mode of part is realized.Described unit also can be set in the processor, for example, can be described as: a kind of processor packet
Include determination unit, fitting unit and line style determination unit.Wherein, the title of these units is not constituted to this under certain conditions
The restriction of unit itself, for example, determination unit is also described as " determining multiple pixels where at least one lane line
Location information unit ".
As on the other hand, present invention also provides a kind of computer-readable medium, which be can be
Included in device described in above-described embodiment;It is also possible to individualism, and without in the supplying device.Above-mentioned calculating
Machine readable medium carries one or more program, when said one or multiple programs are executed by the device, so that should
Device: the color of each pixel in multiple pixels based on target image, in multiple pixels in the position of target image
In information, the location information of multiple pixels where at least one lane line is determined;To more where at least one lane line
The location information of a pixel carries out linear fit, to determine each lane in target image, at least one lane line
The location information of at least two positions on line segment where line;Based at least two on the line segment where each lane line
The location information set, the region where determining each lane line in the target image, and determine the line of lane line in each region
Type.
Above description is only the preferred embodiment of the application and the explanation to institute's application technology principle.Those skilled in the art
Member is it should be appreciated that invention scope involved in the application, however it is not limited to technology made of the specific combination of above-mentioned technical characteristic
Scheme, while should also cover in the case where not departing from foregoing invention design, it is carried out by above-mentioned technical characteristic or its equivalent feature
Any combination and the other technical solutions formed.Such as features described above has similar function with (but being not limited to) disclosed herein
Can technical characteristic replaced mutually and the technical solution that is formed.
Claims (14)
1. a kind of method for handling image, comprising:
The color of each pixel in multiple pixels based on target image, in the multiple pixel in the target image
Location information in, determine the location information of multiple pixels where at least one lane line;
Linear fit is carried out to the location information of multiple pixels where at least one lane line, to determine in the target figure
As, the location informations of at least two positions on line segment where each lane line at least one lane line;
Based on the location information of at least two positions on the line segment where each lane line, determine each lane line in the mesh
Region where in logo image, and determine the line style of lane line in each region.
2. the method according to claim 1 for handling image, wherein described to more where at least one lane line
The location information of a pixel carries out linear fit, to determine in the target image, described at least one lane line
The location information of at least two positions on line segment where each lane line, comprising:
Linear fit is carried out to the location information of multiple pixels where at least one lane line, obtains described at least one
The function of the line segment where each lane line in lane line;
For every lane line, at least two positions of function determining target image, meeting line segment where the lane line
Information.
3. the method according to claim 1 for handling image, wherein multiple pixels based on target image
In the color of each pixel determine at least one vehicle in the multiple pixel in the location information of the target image
The location information of multiple pixels where diatom, comprising:
The location information input of the multiple pixels and the multiple pixel of target image in the target image is preparatory
Trained color classification model, wherein the color classification model is to carry out color classification simultaneously to the pixel for including color
Output belongs to the pixel and location information of the colour type of lane line;
Each colour type is more at least one colour type that obtain the color classification model output, belonging to lane line
Location information of the multiple pixels of a pixel and the colour type in the target image.
4. the method according to claim 1 for handling image, wherein the location information of at least two position is
The location information of two endpoints of the line segment where lane line and the location information at the midpoint for two broadsides in region, each vehicle
Region where diatom is rectangle;And
The location information of at least two positions on the line segment based on where each lane line determines each lane line in institute
Region where stating in target image, comprising:
For every lane line, the position at the midpoint of two broadsides based on the region where the line segment where the lane line is believed
Breath, in the target image determine width be predetermined width and length be described two endpoint lines length rectangle region
Domain.
5. the method according to claim 1 for handling image, wherein the line of lane line in described determining each region
Type, comprising:
By the line style disaggregated model trained in advance of the region input where each lane line, the line style disaggregated model output is obtained
Each region in include lane line line style, wherein the line style disaggregated model divides to the line style to lane line
Class.
6. the method according to claim 1 for handling image, wherein the method also includes:
Straight-line detection and/or frequency domain detection are carried out to the image formerly obtained, determine the son where the lane line of acquired image
Image, and the subgraph is determined as target image.
7. a kind of for handling the device of image, comprising:
Determination unit is configured to the color of each pixel in multiple pixels based on target image, in the multiple picture
Vegetarian refreshments determines the location information of multiple pixels where at least one lane line in the location information of the target image;
Fitting unit is configured to carry out linear fit to the location information of multiple pixels where at least one lane line,
To determine at least two on the line segment where each lane line in the target image, described at least one lane line
The location information of position;
Line style determination unit is configured to the location information based at least two positions on the line segment where each lane line,
Region where determining each lane line in the target image, and determine the line style of lane line in each region.
8. according to claim 7 for handling the device of image, wherein the fitting unit, comprising:
Fitting module is configured to carry out Linear Quasi to the location information of multiple pixels where at least one lane line
It closes, obtains the function of the line segment where each lane line at least one lane line;
Determining module, is configured to for every lane line, letter determining target image, meeting line segment where the lane line
At least two several location informations.
9. according to claim 7 for handling the device of image, wherein the determination unit, comprising:
Input module is configured to multiple pixels of target image and the multiple pixel in the target image
Location information input color classification model trained in advance, wherein the color classification model is to the pixel including color
Point carries out color classification and exports the pixel and location information for belonging to the colour type of lane line;
Output module is configured at least one colour type that obtain the color classification model output, belonging to lane line
In each colour type multiple pixels and the colour type multiple pixels in the target image position letter
Breath.
10. according to claim 7 for handling the device of image, wherein the location information of at least two position
For two endpoints of the line segment where lane line location information and be region two broadsides midpoint location information, each item
Region where lane line is rectangle;And
The line style determination unit is further configured to:
For every lane line, the position at the midpoint of two broadsides based on the region where the line segment where the lane line is believed
Breath, in the target image determine width be predetermined width and length be described two endpoint lines length rectangle region
Domain.
11. according to claim 7 for handling the device of image, wherein the line style determination unit further configures
For:
By the line style disaggregated model trained in advance of the region input where each lane line, the line style disaggregated model output is obtained
Each region in include lane line line style, wherein the line style disaggregated model divides to the line style to lane line
Class.
12. according to claim 7 for handling the device of image, wherein described device further include:
Pretreatment unit is configured to carry out straight-line detection and/or frequency domain detection to the image formerly obtained, determines acquired figure
As lane line where subgraph, and the subgraph is determined as target image.
13. a kind of electronic equipment, comprising:
One or more processors;
Storage device, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processors are real
Now such as method as claimed in any one of claims 1 to 6.
14. a kind of computer readable storage medium, is stored thereon with computer program, wherein when the program is executed by processor
Realize such as method as claimed in any one of claims 1 to 6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711337034.XA CN109960959B (en) | 2017-12-14 | 2017-12-14 | Method and apparatus for processing image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711337034.XA CN109960959B (en) | 2017-12-14 | 2017-12-14 | Method and apparatus for processing image |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109960959A true CN109960959A (en) | 2019-07-02 |
CN109960959B CN109960959B (en) | 2020-04-03 |
Family
ID=67017831
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711337034.XA Active CN109960959B (en) | 2017-12-14 | 2017-12-14 | Method and apparatus for processing image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109960959B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110688971A (en) * | 2019-09-30 | 2020-01-14 | 上海商汤临港智能科技有限公司 | Method, device and equipment for detecting dotted lane line |
CN112507852A (en) * | 2020-12-02 | 2021-03-16 | 上海眼控科技股份有限公司 | Lane line identification method, device, equipment and storage medium |
CN113066153A (en) * | 2021-04-28 | 2021-07-02 | 浙江中控技术股份有限公司 | Method, device and equipment for generating pipeline flow chart and storage medium |
CN113688721A (en) * | 2021-08-20 | 2021-11-23 | 北京京东乾石科技有限公司 | Method and device for fitting lane line |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6212287B1 (en) * | 1996-10-17 | 2001-04-03 | Sgs-Thomson Microelectronics S.R.L. | Method for identifying marking stripes of road lanes |
CN102298693A (en) * | 2011-05-18 | 2011-12-28 | 浙江大学 | Expressway bend detection method based on computer vision |
CN103295420A (en) * | 2013-01-30 | 2013-09-11 | 吉林大学 | Method for recognizing lane line |
CN104318258A (en) * | 2014-09-29 | 2015-01-28 | 南京邮电大学 | Time domain fuzzy and kalman filter-based lane detection method |
CN106203273A (en) * | 2016-06-27 | 2016-12-07 | 开易(北京)科技有限公司 | The lane detection system of multiple features fusion, method and senior drive assist system |
-
2017
- 2017-12-14 CN CN201711337034.XA patent/CN109960959B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6212287B1 (en) * | 1996-10-17 | 2001-04-03 | Sgs-Thomson Microelectronics S.R.L. | Method for identifying marking stripes of road lanes |
CN102298693A (en) * | 2011-05-18 | 2011-12-28 | 浙江大学 | Expressway bend detection method based on computer vision |
CN103295420A (en) * | 2013-01-30 | 2013-09-11 | 吉林大学 | Method for recognizing lane line |
CN104318258A (en) * | 2014-09-29 | 2015-01-28 | 南京邮电大学 | Time domain fuzzy and kalman filter-based lane detection method |
CN106203273A (en) * | 2016-06-27 | 2016-12-07 | 开易(北京)科技有限公司 | The lane detection system of multiple features fusion, method and senior drive assist system |
Non-Patent Citations (1)
Title |
---|
鞠乾翱: "基于机器视觉的快速车道线辨识研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110688971A (en) * | 2019-09-30 | 2020-01-14 | 上海商汤临港智能科技有限公司 | Method, device and equipment for detecting dotted lane line |
WO2021063228A1 (en) * | 2019-09-30 | 2021-04-08 | 上海商汤临港智能科技有限公司 | Dashed lane line detection method and device, and electronic apparatus |
CN110688971B (en) * | 2019-09-30 | 2022-06-24 | 上海商汤临港智能科技有限公司 | Method, device and equipment for detecting dotted lane line |
CN112507852A (en) * | 2020-12-02 | 2021-03-16 | 上海眼控科技股份有限公司 | Lane line identification method, device, equipment and storage medium |
CN113066153A (en) * | 2021-04-28 | 2021-07-02 | 浙江中控技术股份有限公司 | Method, device and equipment for generating pipeline flow chart and storage medium |
CN113688721A (en) * | 2021-08-20 | 2021-11-23 | 北京京东乾石科技有限公司 | Method and device for fitting lane line |
CN113688721B (en) * | 2021-08-20 | 2024-03-05 | 北京京东乾石科技有限公司 | Method and device for fitting lane lines |
Also Published As
Publication number | Publication date |
---|---|
CN109960959B (en) | 2020-04-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108073910B (en) | Method and device for generating human face features | |
CN109960959A (en) | Method and apparatus for handling image | |
CN108090916B (en) | Method and apparatus for tracking the targeted graphical in video | |
CN109063653A (en) | Image processing method and device | |
CN108898185A (en) | Method and apparatus for generating image recognition model | |
CN104239465B (en) | A kind of method and device scanned for based on scene information | |
CN109344762A (en) | Image processing method and device | |
CN109034069A (en) | Method and apparatus for generating information | |
CN108510472A (en) | Method and apparatus for handling image | |
CN109308681A (en) | Image processing method and device | |
CN108345387A (en) | Method and apparatus for output information | |
CN110390237A (en) | Processing Method of Point-clouds and system | |
CN110533055A (en) | A kind for the treatment of method and apparatus of point cloud data | |
CN108364209A (en) | Methods of exhibiting, device, medium and the electronic equipment of merchandise news | |
CN108734185A (en) | Image verification method and apparatus | |
CN107742128A (en) | Method and apparatus for output information | |
CN108491825A (en) | information generating method and device | |
CN110349158A (en) | A kind of method and apparatus handling point cloud data | |
CN109242801A (en) | Image processing method and device | |
CN108595448A (en) | Information-pushing method and device | |
CN108960110A (en) | Method and apparatus for generating information | |
CN108182457A (en) | For generating the method and apparatus of information | |
CN108882025A (en) | Video frame treating method and apparatus | |
CN109214501A (en) | The method and apparatus of information for identification | |
CN109118456A (en) | Image processing method and device |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |