CN106778548A - Method and apparatus for detecting barrier - Google Patents
Method and apparatus for detecting barrier Download PDFInfo
- Publication number
- CN106778548A CN106778548A CN201611078768.6A CN201611078768A CN106778548A CN 106778548 A CN106778548 A CN 106778548A CN 201611078768 A CN201611078768 A CN 201611078768A CN 106778548 A CN106778548 A CN 106778548A
- Authority
- CN
- China
- Prior art keywords
- barrier
- map data
- streetscape map
- road
- detection
- 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
- G06V20/00—Scenes; Scene-specific elements
- G06V20/20—Scenes; Scene-specific elements in augmented reality scenes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Multimedia (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Remote Sensing (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
- Processing Or Creating Images (AREA)
Abstract
This application discloses the method and apparatus for detecting barrier.One specific embodiment of methods described includes：Obtain detection of obstacles model and target streetscape map data；Use barriers analyte detection model inspection simultaneously marks out the barrier in target streetscape map data；Wherein, detection of obstacles model is obtained as follows：Obtain the streetscape map data for marking out barrier in advance；Based on the barrier for being marked out, streetscape map data are processed；The part streetscape map data with the barrier for marking out are chosen from the streetscape map data after treatment as training data；Default detection of obstacles model is trained using training data, detection of obstacles model is obtained.The implementation method realizes making full use of and carrying out fast accurate detection to the barrier in streetscape map data for the streetscape map data to having marked out barrier.
Description
Technical field
The application is related to field of computer technology, and in particular to object detecting areas, more particularly to a kind of for detecting barrier
The method and apparatus for hindering thing.
Background technology
The automatic detection of barrier is significant for pedestrian visually impaired or automatic driving vehicle.Streetscape ground
Diagram data generally includes view data or cloud data, it comprises the obstacle information in the running environment of vehicles or pedestrians,
Such as pedestrian, vehicle, building etc..
Existing obstacle detection method first has to that the streetscape map data for gathering are identified and marked, to determine to adopt
The position of the barrier that the streetscape map data of collection include.Realized to above-mentioned streetscape map data using machine learning algorithm
Identification and mark when, first have to be trained machine learning algorithm, because the data volume needed for training is larger, cause to street
It is relatively costly that the processes such as collection, the mark of scape map datum are consumed, it is impossible to sufficiently using the streetscape map data for having gathered
Value.
The content of the invention
The purpose of the application is to propose a kind of to solve background above technology for detecting the method and apparatus of barrier
The technical problem that part is mentioned.
In a first aspect, this application provides a kind of method for detecting barrier, methods described includes：Obtain barrier
Detection model and target streetscape map data；Using the detection of obstacles model inspection and mark out the target streetscape map
Barrier in data；Wherein, the detection of obstacles model is obtained as follows：Acquisition marks out obstacle in advance
The streetscape map data of thing, the streetscape map data include the obstacle information in running environment；Based on the barrier for being marked out
Hinder thing, the streetscape map data are processed；Chosen with the obstacle for marking out from the streetscape map data after treatment
The part streetscape map data of thing are used as training data；Default detection of obstacles model is trained using the training data, is obtained
To the detection of obstacles model.
In certain embodiments, the streetscape map data also include the road information in running environment；And the side
Method also includes：The streetscape map data are imported the road Identification model of training in advance, in the identification streetscape map data
Road；The recognized road of mark, determines the road area in the streetscape map data.
In certain embodiments, it is described the streetscape map data are processed based on the barrier for being marked out, bag
Include：At least one barrier is chosen from the barrier for being marked out；Increase selected barrier in the road area；
By in the road area by the road information in region of increased barrier covering delete.
In certain embodiments, it is described the streetscape map data are processed based on the barrier for being marked out, bag
Include：At least one barrier is chosen from the barrier for being marked out；The position of selected barrier is moved to the road
Road region；The road information in the region of the barrier covering that will be moved in the road area is deleted.
In certain embodiments, the streetscape map data include view data；And it is described based on the barrier for being marked out
Hinder thing, the streetscape map data are processed, including：Whether include traffic lights in the marked out barrier of detection；Ring
Traffic lights should be included in the barrier for being marked out, change the color data of the traffic lights.
In certain embodiments, methods described also include it is following at least one：Streetscape map data after delete processing；It is defeated
Go out to mark out the target streetscape map data of barrier.
Second aspect, this application provides a kind of device for detecting barrier, described device includes：Acquiring unit,
For obtaining detection of obstacles model and target streetscape map data；Detection unit, for utilizing the detection of obstacles model
Detect and mark out the barrier in the target streetscape map data；Wherein, the detection of obstacles model is by training
What unit was obtained, the training unit includes：Acquisition module, the streetscape map data of barrier are marked out for obtaining in advance,
The streetscape map data include the obstacle information in running environment；Processing module, for based on the barrier for being marked out,
The streetscape map data are processed；Module is chosen, for being chosen with mark from the streetscape map data after treatment
The part streetscape map data of the barrier for going out are used as training data；Training module, for pre- using training data training
If detection of obstacles model, obtain the detection of obstacles model.
In certain embodiments, the streetscape map data also include the road information in running environment；And the dress
Putting also includes road Identification unit, and the road Identification unit includes：Identification module, for the streetscape map data to be imported
The road Identification model of training in advance, recognizes the road in the streetscape map data；Labeling module, for marking what is recognized
Road, determines the road area in the streetscape map data.
In certain embodiments, the processing module is further used for：At least one is chosen from the barrier for being marked out
Individual barrier；Increase selected barrier in the road area；By in the road area by the increased barrier of institute
The road information in the region of covering is deleted.
In certain embodiments, the processing module is further used for：At least one is chosen from the barrier for being marked out
Individual barrier；The position of selected barrier is moved to the road area；By what is moved in the road area
The road information in the region of barrier covering is deleted.
In certain embodiments, the streetscape map data include view data；And the processing module is further used
In：Whether include traffic lights in the marked out barrier of detection；Barrier in response to being marked out includes traffic lights, changes
Become the color data of the traffic lights.
In certain embodiments, described device also include it is following at least one：Unit is deleted, for the street after delete processing
Scape map datum；Output unit, the target streetscape map data of barrier are marked out for exporting.
The method and apparatus for detecting barrier that the application is provided, by the streetscape ground to marking out barrier in advance
Diagram data is processed, and the part streetscape map with the barrier for marking out is chosen from the streetscape map data after treatment
Data train default detection of obstacles model, obtain detection of obstacles model, and utilize above-mentioned detection of obstacles model inspection
And the barrier in label target streetscape map data, on the one hand realize streetscape map data to having marked out barrier
Make full use of, on the other hand realize carries out fast accurate detection to the barrier in streetscape map data.
Brief description of the drawings
By the detailed description made to non-limiting example made with reference to the following drawings of reading, the application other
Feature, objects and advantages will become more apparent upon：
Fig. 1 is the flow chart of one embodiment of the method for detecting barrier according to the application；
Fig. 2 is one embodiment of the training detection of obstacles model of the method for detecting barrier according to the application
Flow chart；
Fig. 3 is a schematic diagram for application scenarios of the method for detecting barrier according to the application；
Fig. 4 is the structural representation of one embodiment of the device for detecting barrier according to the application；
Fig. 5 is adapted for the structural representation for realizing the terminal device of the embodiment of the present application or the computer system of server
Figure.
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
Be easy to description, be illustrate only in accompanying drawing to about the related part of invention.
It should be noted that in the case where not conflicting, the feature in embodiment and embodiment in the application can phase
Mutually combination.Describe the application in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
With reference to Fig. 1, the flow 100 of one embodiment of the method for detecting barrier according to the application is shown.
The method for detecting barrier of the present embodiment, comprises the following steps：
Step 101, obtains detection of obstacles model and target streetscape map data.
In the present embodiment, above-mentioned detection of obstacles model can be it is various can be to being included in view data or cloud data
The model that builds of the various algorithms that are detected of barrier, such as convolutional neural networks (Convolutional Neural
Network, CNN), random forests algorithm etc..Above-mentioned barrier can be pedestrian or vehicle various obstacles in the process of moving
Thing, such as pedestrian, vehicle, building etc..Target streetscape map data can be barrier to be detected various view data or
Cloud data, wherein, cloud data can be obtained by laser radar, and view data can be obtained by image collecting device.
The electronic equipment (such as terminal device) for running the method for detecting barrier of the present embodiment can obtain the barrier being locally stored
Hinder analyte detection model and target streetscape map data, it is also possible to which artificial importing is obtained by wired connection or radio connection
Detection of obstacles model and target streetscape map data.
It is pointed out that above-mentioned radio connection can include but is not limited to 3G/4G connections, WiFi connections, bluetooth
Connection, WiMAX connections, Zigbee connections, UWB (ultra wideband) connections and other currently known or future develop
Radio connection.
Step 102, Use barriers analyte detection model inspection simultaneously marks out the barrier in target streetscape map data.
By in above-mentioned target streetscape map data input detection of obstacles model, you can in target streetscape map data
Barrier is detected, and the barrier that will be detected is marked out and.Wherein, above-mentioned detection of obstacles model is by Fig. 2 institutes
What the following steps training shown was obtained：
Step 201, obtains the streetscape map data for marking out barrier in advance.
In the present embodiment, the electronic equipment (such as terminal device) of the method for detecting barrier of operation the present embodiment
The streetscape map data for marking out barrier in advance being locally stored can be obtained, it is also possible to by wired connection or wireless connection
Mode obtains the artificial streetscape map data for marking out barrier.Above-mentioned mark refers to the barrier included to map datum
(such as pedestrian, vehicle), road sign (such as traffic lights, speed(-)limit sign) information are labeled, can using minimum circumscribed rectangle frame or
The frame of object to be marked is marked.It is understood that above-mentioned streetscape map data include the traveling ring of pedestrian or vehicle
The data such as size, shape, the color of the obstacle information in border, such as barrier.
Streetscape map data, based on the barrier for being marked out, are processed by step 202.
In the present embodiment, terminal device can be based on marked out barrier, streetscape map data be carried out various each
The treatment of sample, for example, changing marked out position, quantity, color of barrier etc..It is understood that the street after treatment
Scape map datum still includes the barrier for marking out.
Step 203, chooses the part streetscape map with the barrier for marking out from the streetscape map data after treatment
Data are used as training data.
After processing the above-mentioned streetscape map data for marking out barrier, equivalent to in original street view map datum
Comprising information excavated, the new streetscape map data differed with original street view map datum have been obtained, using new
Streetscape map data in choose with the part streetscape map data of barrier for marking out as training data.
Step 204, default detection of obstacles model is trained using training data, obtains detection of obstacles model.
Default detection of obstacles model is trained using above-mentioned training data, the obstacle mentioned in step 101 can be obtained
Analyte detection model.Wherein, above-mentioned default detection of obstacles model can be the unadjusted various algorithms of parameter, for example, obstacle
When analyte detection model is convolutional neural networks, default detection of obstacles model can be the initial volume for using newly-built function newly-built
Product neutral net.
With continued reference to Fig. 3, Fig. 3 is one of the application scenarios of the method for detecting barrier according to the present embodiment and shows
It is intended to.In the application scenarios of Fig. 3, automatic driving vehicle 301 is travelled on road, is provided with automatic driving vehicle 301 and taken the photograph
Camera 302, for gathering the obstacle information in running environment.Video camera 302, will after the image containing barrier is collected
Image imports detection of obstacles model 303, obtains marking out the image 304 of barrier.It is understood that above-mentioned obstacle quality testing
Survey model 303 may be mounted in vehicle-mounted computer, it is also possible to install in the server, vehicle is connected by network with server.
The method for detecting barrier that above-described embodiment of the application is provided, by marking out barrier in advance
Streetscape map data are processed, and the part street with the barrier for marking out is chosen from the streetscape map data after treatment
Scape map datum trains default detection of obstacles model, obtains detection of obstacles model, and utilize above-mentioned detection of obstacles mould
Barrier in type detection and label target streetscape map data, on the one hand realizes the streetscape map to having marked out barrier
Data make full use of, and on the other hand to realize carry out fast accurate detection to the barrier in streetscape map data.
In some optional implementations of the present embodiment, above-mentioned streetscape map data can also include vehicles or pedestrians
Road information in running environment, the above method can also include the following steps not shown in Fig. 2：
Streetscape map data are imported the road Identification model of training in advance, the road letter in identification streetscape map data
Breath；The recognized road of mark, determines the road area in streetscape map data.
In this implementation, when pedestrian or vehicle traveling are on road, in addition to obstacle information, in streetscape map data
Road information can also be included.Above-mentioned streetscape map data can be imported the road Identification model of training in advance, to recognize street
Road information in scape map datum.Above-mentioned road information can include the width of road, (one-way traffic track is two-way for type
Traveling lane), title etc..After road area is identified, the road area that can be will identify that is marked out and, so that it is determined that
Road area in above-mentioned streetscape map data.
In some optional implementations of the present embodiment, can be realized by the following steps not shown in Fig. 2
Treatment to streetscape map data：
At least one barrier is chosen from the barrier for being marked out；Increase selected obstacle in road area
Thing；By in road area by the road information in region of increased barrier covering delete.
In this implementation, one or more barriers can be chosen from the barrier for having marked out, and will be selected
Barrier increase to road area.It is understood that when pedestrian or vehicle traveling are on road, to its traveling shadow
Ring maximum should be the barrier on road, therefore, in the present embodiment, just selected barrier increases to road area
In, relative to original street view map datum, new streetscape map data increased barrier on road.So, using new
Streetscape map data when being trained to default detection of obstacles model, the detection of obstacles model after training can be caused
More can fast and accurately detect the position of barrier on road.After selected barrier is increased in road area,
Raw information in the road area that will can be marked out is deleted.
In some optional implementations of the present embodiment, can also be realized by the following steps not shown in Fig. 2
Treatment to streetscape map data：
At least one barrier is chosen from the barrier for being marked out；The position of selected barrier is moved to
Road region；The road information in the region of the barrier covering that will be moved in the road area is deleted.
Difference with above-mentioned implementation is, in this implementation, the treatment to streetscape map data stresses
In the position for changing the barrier for marking out, i.e., on the premise of the quantity of barrier in keeping streetscape map data is constant, lead to
The position of change barrier is crossed to form new streetscape map data.Carry what is marked out in the new streetscape map data of recycling
The part streetscape map data of barrier go to train default detection of obstacles model, can cause the detection of obstacles after training
Model is more sensitive to the barrier on road, it is easier to identify the barrier on road.
In some optional implementations of the present embodiment, above-mentioned streetscape map data can also include view data,
The treatment to streetscape map data can also be realized by the following steps not shown in Fig. 2 accordingly：
Whether include traffic lights in the marked out barrier of detection；Barrier in response to being marked out includes traffic
Lamp, changes the color data of traffic lights.
In this implementation, when streetscape map datum is the view data gathered by monocular or binocular camera, right
When streetscape map data are processed, whether include traffic lights in detection barrier first, when including traffic lights, by changing
The color data of traffic lights realizes the treatment to streetscape map data.So, when collection view data it is less or collection
When the color of traffic lights is red, green or yellow in view data, can just be obtained by the color data for changing traffic lights
To the view data of the traffic lights including other colors, such that it is able to preferably make the detection of obstacles model after training to traffic
The color of lamp is more sensitive, realizes the signal fast and accurately indicated by identification traffic lights.
In some optional implementations of the present embodiment, the above-mentioned method for detecting barrier can also include figure
Following steps not shown in 1：
Streetscape map data after delete processing.
In this implementation, processed to streetscape map data, and train the default detection of obstacles model of completion
Afterwards, timely the streetscape map data after treatment can be deleted, it is to avoid the streetscape map data after excessive treatment take deposits
Storage space, increases storage and maintenance cost.
In some optional implementations of the present embodiment, the above-mentioned method for detecting barrier can also include figure
Following steps not shown in 1：
Output marks out the target streetscape map data of barrier.
In this implementation, after the barrier included in detecting target streetscape map data, barrier can will be marked out
Hinder the target streetscape map data output of thing, so, staff can detect the accuracy of mark, preferably to safeguard obstacle
Analyte detection model.
With further reference to Fig. 4, as the realization to method shown in above-mentioned each figure, hinder for detecting this application provides one kind
Hinder one embodiment of the device of thing, the device embodiment is corresponding with the embodiment of the method shown in Fig. 1, the device specifically can be with
It is applied in various electronic equipments.
As shown in figure 4, the present embodiment for detecting that the device 400 of barrier includes：Acquiring unit 401, detection unit
402 and training unit 403.
Wherein, acquiring unit 401, for obtaining detection of obstacles model and target streetscape map data.
Detection unit 402, for the detection of obstacles model inspection obtained using acquiring unit 401 and marks out target street
Barrier in scape map datum.
In the present embodiment, detection of obstacles model is obtained by training unit 403, and training unit 403 includes：
Acquisition module 4031, the streetscape map data of barrier are marked out for obtaining in advance.
Above-mentioned streetscape map data include the obstacle information in running environment.
Processing module 4032, for based on the barrier for being marked out, processing streetscape map data.
Module 4033 is chosen, for choosing the part with the barrier for marking out from the streetscape map data after treatment
Streetscape map data are used as training data.
Training module 4034, for training default detection of obstacles model using training data, obtains the barrier
Detection model.
The device for detecting barrier that above-described embodiment of the application is provided, by processing module to marking out in advance
The streetscape map data of barrier are processed, and selection module is chosen from the streetscape map data after treatment and carries what is marked out
The part streetscape map data of barrier train default barrier as training data, training module using above-mentioned training data
Detection model, obtains detection of obstacles model, and detection unit is using above-mentioned detection of obstacles model inspection and marks acquiring unit
Barrier in the target streetscape map data for getting, on the one hand realizes the streetscape map data to having marked out barrier
Make full use of, on the other hand realize carries out fast accurate detection to the barrier in streetscape map data.
In some optional implementations of the present embodiment, above-mentioned streetscape map data also include the road in running environment
Road information, the above-mentioned device 400 for detecting barrier can also include the road Identification unit not shown in Fig. 4.Wherein, road
Road recognition unit can include identification module and labeling module.
Identification module, the road Identification model for streetscape map data to be imported training in advance, recognizes streetscape map number
Road in.
Labeling module, for marking recognized road, determines the road area in streetscape map data.
In some optional implementations of the present embodiment, above-mentioned processing module 4032 can be further used for：
At least one barrier is chosen from the barrier for being marked out；Increase selected obstacle in road area
Thing；By in road area by the road information in region of increased barrier covering delete.
In some optional implementations of the present embodiment, above-mentioned processing module 4032 can also be further used for：
At least one barrier is chosen from the barrier for being marked out；The position of selected barrier is moved to
Road region；The road information in the region of the barrier covering that will be moved in road area is deleted.
In some optional implementations of the present embodiment, above-mentioned streetscape map data include view data, above-mentioned place
Reason module 4032 can also be further used for：
Whether include traffic lights in the marked out barrier of detection；Barrier in response to being marked out includes traffic
Lamp, changes the color data of traffic lights.
It is above-mentioned for detecting that the device 400 of barrier can also be wrapped in some optional implementations of the present embodiment
Include deletion unit and/or the output unit not shown in Fig. 4.
Wherein, unit is deleted, for the streetscape map data after delete processing.
Output unit, the target streetscape map data of barrier are marked out for exporting.
It should be appreciated that for detect the unit 401 described in the device 400 of barrier to unit 403 respectively with refer to Fig. 1
And each step in the method described in Fig. 2 is corresponding.Thus, above with respect to for detecting the behaviour that the method for barrier is described
Make and feature is equally applicable to device 400 and the unit for wherein including, will not be repeated here.The corresponding units of device 400 can be with
Cooperate to realize the scheme of the embodiment of the present application with the unit in server.
Below with reference to Fig. 5, it illustrates the calculating for being suitable to terminal device or server for realizing the embodiment of the present application
The structural representation of machine system 500.
As shown in figure 5, computer system 500 includes CPU (CPU) 501, it can be according to storage read-only
Program in memory (ROM) 502 or be loaded into program in random access storage device (RAM) 503 from storage part 508 and
Perform various appropriate actions and treatment.In RAM 503, the system that is also stored with 500 operates required various programs and data.
CPU 501, ROM 502 and RAM 503 are connected with each other by bus 504.Input/output (I/O) interface 505 is also connected to always
Line 504.
I/O interfaces 505 are connected to lower component：Including the importation 506 of keyboard, mouse etc.；Penetrated including such as negative electrode
The output par, c 507 of spool (CRT), liquid crystal display (LCD) etc. and loudspeaker etc.；Storage part 508 including hard disk etc.；
And the communications portion 509 of the NIC including LAN card, modem etc..Communications portion 509 via such as because
The network of spy's net performs communication process.Driver 510 is also according to needing to be connected to I/O interfaces 505.Detachable media 511, such as
Disk, CD, magneto-optic disk, semiconductor memory etc., as needed on driver 510, in order to read from it
Computer program be mounted into as needed storage part 508.
Especially, in accordance with an embodiment of the present disclosure, the process above with reference to flow chart description may be implemented as computer
Software program.For example, embodiment of the disclosure includes a kind of computer program product, it includes being tangibly embodied in machine readable
Computer program on medium, the computer program includes the program code for the method shown in execution flow chart.At this
In the embodiment of sample, the computer program can be downloaded and installed by communications portion 509 from network, and/or from removable
Medium 511 is unloaded to be mounted.When the computer program is performed by CPU (CPU) 501, in execution the present processes
The above-mentioned functions of restriction.
Flow chart and block diagram in accompanying drawing, it is illustrated that according to the system of the various embodiments of the application, method and computer journey
The architectural framework in the cards of sequence product, function and operation.At this point, each square frame in flow chart or block diagram can generation
One part for module, program segment or code of table a, part for the module, program segment or code includes one or more
Executable instruction for realizing the logic function of regulation.It should also be noted that in some realizations as replacement, institute in square frame
The function of mark can also occur with different from the order marked in accompanying drawing.For example, two square frame reality for succeedingly representing
On can perform substantially in parallel, they can also be performed in the opposite order sometimes, and this is depending on involved function.Also
It is noted that the combination of the square frame in each square frame and block diagram and/or flow chart in block diagram and/or flow chart, Ke Yiyong
Perform the function of regulation or the special hardware based system of operation to realize, or can be referred to computer with specialized hardware
The combination of order is realized.
Being described in involved unit in the embodiment of the present application can be realized by way of software, it is also possible to by hard
The mode of part is realized.Described unit can also be set within a processor, for example, can be described as：A kind of processor bag
Include acquiring unit, detection unit and training unit.Wherein, the title of these units is not constituted to the unit under certain conditions
The restriction of itself, for example, acquiring unit is also described as " obtaining detection of obstacles model and target streetscape map data
Unit ".
As on the other hand, present invention also provides a kind of nonvolatile computer storage media, the non-volatile calculating
Machine storage medium can be the nonvolatile computer storage media included in device described in above-described embodiment；Can also be
Individualism, without the nonvolatile computer storage media allocated into terminal.Above-mentioned nonvolatile computer storage media is deposited
One or more program is contained, when one or more of programs are performed by an equipment so that the equipment：Obtain
Detection of obstacles model and target streetscape map data；Using the detection of obstacles model inspection and mark out the target street
Barrier in scape map datum；Wherein, the detection of obstacles model is obtained as follows：Obtain mark in advance
Go out the streetscape map data of barrier, the streetscape map data include the obstacle information in running environment；Based on being marked
The streetscape map data are processed by the barrier for going out；Choose to carry from the streetscape map data after treatment and mark out
Barrier part streetscape map data as training data；Default detection of obstacles mould is trained using the training data
Type, obtains the detection of obstacles model.
Above description is only the preferred embodiment and the explanation to institute's application technology principle of the application.People in the art
Member is it should be appreciated that involved invention scope in the application, however it is not limited to the technology of the particular combination of above-mentioned technical characteristic
Scheme, while should also cover in the case where the inventive concept is not departed from, is carried out by above-mentioned technical characteristic or its equivalent feature
Other technical schemes for being combined and being formed.Such as features described above has similar work(with (but not limited to) disclosed herein
The technical scheme that the technical characteristic of energy is replaced mutually and formed.
Claims (12)
1. a kind of method for detecting barrier, it is characterised in that methods described includes：
Obtain detection of obstacles model and target streetscape map data；
Using the detection of obstacles model inspection and mark out the barrier in the target streetscape map data；
Wherein, the detection of obstacles model is obtained as follows：
The streetscape map data for marking out barrier in advance are obtained, the streetscape map data include the barrier in running environment
Information；
Based on the barrier for being marked out, the streetscape map data are processed；
The part streetscape map data with the barrier for marking out are chosen from the streetscape map data after treatment as training
Data；
Default detection of obstacles model is trained using the training data, the detection of obstacles model is obtained.
2. method according to claim 1, it is characterised in that the streetscape map data also include the road in running environment
Road information；And
Methods described also includes：
The streetscape map data are imported the road Identification model of training in advance, the road in the streetscape map data is recognized
Road；
The recognized road of mark, determines the road area in the streetscape map data.
3. method according to claim 2, it is characterised in that described based on the barrier for being marked out, to the streetscape
Map datum is processed, including：
At least one barrier is chosen from the barrier for being marked out；
Increase selected barrier in the road area；
By in the road area by the road information in region of increased barrier covering delete.
4. method according to claim 2, it is characterised in that described based on the barrier for being marked out, to the streetscape
Map datum is processed, including：
At least one barrier is chosen from the barrier for being marked out；
The position of selected barrier is moved to the road area；
The road information in the region of the barrier covering that will be moved in the road area is deleted.
5. method according to claim 1, it is characterised in that the streetscape map data include view data；And
It is described the streetscape map data are processed based on the barrier for being marked out, including：
Whether include traffic lights in the marked out barrier of detection；
Barrier in response to being marked out includes traffic lights, changes the color data of the traffic lights.
6. according to the method that one of claim 1-5 is described, it is characterised in that methods described also include it is following at least one：
Streetscape map data after delete processing；
Output marks out the target streetscape map data of barrier.
7. a kind of device for detecting barrier, it is characterised in that described device includes：
Acquiring unit, for obtaining detection of obstacles model and target streetscape map data；
Detection unit, for using the detection of obstacles model inspection and marking out the barrier in the target streetscape map data
Hinder thing；
Wherein, the detection of obstacles model is obtained by training unit, and the training unit includes：
Acquisition module, the streetscape map data of barrier are marked out for obtaining in advance, and the streetscape map data include traveling
Obstacle information in environment；
Processing module, for based on the barrier for being marked out, processing the streetscape map data；
Module is chosen, for choosing the part streetscape map with the barrier for marking out from the streetscape map data after treatment
Data are used as training data；
Training module, for training default detection of obstacles model using the training data, obtains the detection of obstacles
Model.
8. device according to claim 7, it is characterised in that the streetscape map data also include the road in running environment
Road information；And
Described device also includes road Identification unit, and the road Identification unit includes：
Identification module, the road Identification model for the streetscape map data to be imported training in advance recognizes the streetscape ground
Road in diagram data；
Labeling module, for marking recognized road, determines the road area in the streetscape map data.
9. device according to claim 8, it is characterised in that the processing module is further used for：
At least one barrier is chosen from the barrier for being marked out；
Increase selected barrier in the road area；
By in the road area by the road information in region of increased barrier covering delete.
10. device according to claim 8, it is characterised in that the processing module is further used for：
At least one barrier is chosen from the barrier for being marked out；
The position of selected barrier is moved to the road area；
The road information in the region of the barrier covering that will be moved in the road area is deleted.
11. devices according to claim 7, it is characterised in that the streetscape map data include view data；And
The processing module is further used for：
Whether include traffic lights in the marked out barrier of detection；
Barrier in response to being marked out includes traffic lights, changes the color data of the traffic lights.
12. according to one of claim 7-11 described device, it is characterised in that described device also include it is following at least one：
Unit is deleted, for the streetscape map data after delete processing；
Output unit, the target streetscape map data of barrier are marked out for exporting.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611078768.6A CN106778548B (en) | 2016-11-30 | 2016-11-30 | Method and apparatus for detecting obstacles |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611078768.6A CN106778548B (en) | 2016-11-30 | 2016-11-30 | Method and apparatus for detecting obstacles |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106778548A true CN106778548A (en) | 2017-05-31 |
| CN106778548B CN106778548B (en) | 2021-04-09 |
Family
ID=58898990
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611078768.6A Active CN106778548B (en) | 2016-11-30 | 2016-11-30 | Method and apparatus for detecting obstacles |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106778548B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108805882A (en) * | 2018-05-29 | 2018-11-13 | 杭州视氪科技有限公司 | A kind of water surface and puddle detection method |
| CN109145680A (en) * | 2017-06-16 | 2019-01-04 | 百度在线网络技术（北京）有限公司 | A kind of method, apparatus, equipment and computer storage medium obtaining obstacle information |
| CN109544981A (en) * | 2018-12-29 | 2019-03-29 | 北京百度网讯科技有限公司 | Image processing method, device, equipment and medium |
| CN110377024A (en) * | 2018-04-13 | 2019-10-25 | 百度（美国）有限责任公司 | Automaticdata for automatic driving vehicle marks |
| CN111142150A (en) * | 2020-01-06 | 2020-05-12 | 中国石油化工股份有限公司 | Automatic intelligent obstacle avoidance design method for seismic exploration |
| CN111160360A (en) * | 2018-11-07 | 2020-05-15 | 北京四维图新科技股份有限公司 | Image recognition method, device and system |
| CN111325136A (en) * | 2020-02-17 | 2020-06-23 | 北京小马智行科技有限公司 | Method and device for labeling object in intelligent vehicle and unmanned vehicle |
| CN111401133A (en) * | 2020-02-19 | 2020-07-10 | 北京三快在线科技有限公司 | Target data augmentation method, device, electronic device and readable storage medium |
| CN112861573A (en) * | 2019-11-27 | 2021-05-28 | 宇龙计算机通信科技(深圳)有限公司 | Obstacle identification method and device, storage medium and intelligent lamp pole |
| WO2021134354A1 (en) * | 2019-12-30 | 2021-07-08 | 深圳元戎启行科技有限公司 | Path prediction method and apparatus, computer device, and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103176185A (en) * | 2011-12-26 | 2013-06-26 | 上海汽车集团股份有限公司 | Method and system for detecting road barrier |
| CN103793684A (en) * | 2012-10-30 | 2014-05-14 | 现代自动车株式会社 | Apparatus and method for detecting obstacle for around view monitoring system |
| CN103954275A (en) * | 2014-04-01 | 2014-07-30 | 西安交通大学 | Lane line detection and GIS map information development-based vision navigation method |
| CN105740802A (en) * | 2016-01-28 | 2016-07-06 | 北京中科慧眼科技有限公司 | Disparity map-based obstacle detection method and device as well as automobile driving assistance system |
| CN105957145A (en) * | 2016-04-29 | 2016-09-21 | 百度在线网络技术（北京）有限公司 | Road barrier identification method and device |
-
2016
- 2016-11-30 CN CN201611078768.6A patent/CN106778548B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103176185A (en) * | 2011-12-26 | 2013-06-26 | 上海汽车集团股份有限公司 | Method and system for detecting road barrier |
| CN103793684A (en) * | 2012-10-30 | 2014-05-14 | 现代自动车株式会社 | Apparatus and method for detecting obstacle for around view monitoring system |
| CN103954275A (en) * | 2014-04-01 | 2014-07-30 | 西安交通大学 | Lane line detection and GIS map information development-based vision navigation method |
| CN105740802A (en) * | 2016-01-28 | 2016-07-06 | 北京中科慧眼科技有限公司 | Disparity map-based obstacle detection method and device as well as automobile driving assistance system |
| CN105957145A (en) * | 2016-04-29 | 2016-09-21 | 百度在线网络技术（北京）有限公司 | Road barrier identification method and device |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109145680A (en) * | 2017-06-16 | 2019-01-04 | 百度在线网络技术（北京）有限公司 | A kind of method, apparatus, equipment and computer storage medium obtaining obstacle information |
| CN110377024A (en) * | 2018-04-13 | 2019-10-25 | 百度（美国）有限责任公司 | Automaticdata for automatic driving vehicle marks |
| CN108805882B (en) * | 2018-05-29 | 2021-09-03 | 杭州视氪科技有限公司 | Water surface and water pit detection method |
| CN108805882A (en) * | 2018-05-29 | 2018-11-13 | 杭州视氪科技有限公司 | A kind of water surface and puddle detection method |
| CN111160360B (en) * | 2018-11-07 | 2023-08-01 | 北京四维图新科技股份有限公司 | Image recognition method, device and system |
| CN111160360A (en) * | 2018-11-07 | 2020-05-15 | 北京四维图新科技股份有限公司 | Image recognition method, device and system |
| CN109544981A (en) * | 2018-12-29 | 2019-03-29 | 北京百度网讯科技有限公司 | Image processing method, device, equipment and medium |
| CN112861573A (en) * | 2019-11-27 | 2021-05-28 | 宇龙计算机通信科技(深圳)有限公司 | Obstacle identification method and device, storage medium and intelligent lamp pole |
| WO2021134354A1 (en) * | 2019-12-30 | 2021-07-08 | 深圳元戎启行科技有限公司 | Path prediction method and apparatus, computer device, and storage medium |
| CN111142150A (en) * | 2020-01-06 | 2020-05-12 | 中国石油化工股份有限公司 | Automatic intelligent obstacle avoidance design method for seismic exploration |
| CN111325136A (en) * | 2020-02-17 | 2020-06-23 | 北京小马智行科技有限公司 | Method and device for labeling object in intelligent vehicle and unmanned vehicle |
| CN111325136B (en) * | 2020-02-17 | 2024-03-19 | 北京小马慧行科技有限公司 | Method and device for labeling object in intelligent vehicle and unmanned vehicle |
| CN111401133A (en) * | 2020-02-19 | 2020-07-10 | 北京三快在线科技有限公司 | Target data augmentation method, device, electronic device and readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106778548B (en) | 2021-04-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106778548A (en) | Method and apparatus for detecting barrier | |
| US10817731B2 (en) | Image-based pedestrian detection | |
| CN108345822B (en) | Point cloud data processing method and device | |
| CN113748315B (en) | System for automatic lane marking | |
| US9652980B2 (en) | Enhanced clear path detection in the presence of traffic infrastructure indicator | |
| US9852357B2 (en) | Clear path detection using an example-based approach | |
| CN106097755A (en) | For identifying the method parked place and/or vacate place | |
| DE102016121864A1 (en) | Distinguish lane markers to be followed by a vehicle | |
| CN106092114A (en) | The automobile real scene navigation apparatus of a kind of image recognition and method | |
| EP3706072B1 (en) | Method, apparatus, and system for detecting degraded ground paint in an image | |
| CN112329553B (en) | Lane line marking method and device | |
| CN108520635B (en) | Mixed reality road display system | |
| CN107077606A (en) | For the method and apparatus for the travel situations for determining there is currently | |
| CN112418081B (en) | Method and system for quickly surveying traffic accidents by air-ground combination | |
| JP5522475B2 (en) | Navigation device | |
| CN116457800A (en) | Architecture for map change detection in an autonomous vehicle | |
| CN109903574A (en) | The acquisition methods and device of crossing traffic information | |
| CN116597690B (en) | Highway test scene generation method, equipment and medium for intelligent network-connected automobile | |
| CN110765224A (en) | Processing method of electronic map, vehicle vision repositioning method and vehicle-mounted equipment | |
| CN117612127B (en) | Scene generation method and device, storage medium and electronic equipment | |
| CN108711298B (en) | Mixed reality road display method | |
| JP6951782B2 (en) | Installation object recognition system and its program | |
| EP4155671A1 (en) | Method, apparatus, and system for confirming road vector geometry based on aerial images | |
| CN114359233B (en) | Image segmentation model training method and device, electronic equipment and readable storage medium | |
| CN114693722A (en) | Vehicle driving behavior detection method, detection device and detection equipment |
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 |