CN105774649B - Method and apparatus for alerting vehicle barrier - Google Patents
Method and apparatus for alerting vehicle barrier Download PDFInfo
- Publication number
- CN105774649B CN105774649B CN201410820275.XA CN201410820275A CN105774649B CN 105774649 B CN105774649 B CN 105774649B CN 201410820275 A CN201410820275 A CN 201410820275A CN 105774649 B CN105774649 B CN 105774649B
- Authority
- CN
- China
- Prior art keywords
- image
- estimator
- vehicle
- information
- collision time
- 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.)
- Active
Links
- 230000011218 segmentation Effects 0.000 claims abstract description 38
- 238000003384 imaging method Methods 0.000 claims abstract description 37
- 230000000875 corresponding Effects 0.000 claims abstract description 12
- 238000003709 image segmentation Methods 0.000 claims abstract description 8
- 230000000694 effects Effects 0.000 claims description 10
- 230000012447 hatching Effects 0.000 claims description 8
- 238000010586 diagram Methods 0.000 description 4
- 206010057668 Cognitive disease Diseases 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 238000004040 coloring Methods 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 230000000414 obstructive Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000000034 method Methods 0.000 description 1
- 230000005236 sound signal Effects 0.000 description 1
- 230000001550 time effect Effects 0.000 description 1
- 238000002604 ultrasonography Methods 0.000 description 1
Abstract
The present invention relates to a kind of equipment for alerting vehicle barrier comprising is configured to shoot the image acquiring device of vehicle periphery image by imaging sensor.The current reference image that image segmentation device is configured to will acquire is divided with object block.Exercise estimator is configured to the two dimensional motion information of estimation object.Variation estimator is configured to the range information between estimation object and vehicle.Relative motion estimator be configured to by by two dimensional motion information and range information in conjunction with come the three dimensional relative movement information of estimating object.Collision time estimator is configured to estimate the collision time between object and vehicle by using three dimensional relative movement information.Image processor is configured to adjust the brightness of corresponding segmentation block.Display is configured to display image.
Description
Technical field
The present invention relates to the methods and apparatus for alerting barrier, and more particularly, to a kind of intuitively to driving
The method and apparatus for alerting barrier of the person's of sailing transmission obstacle information.
Background technique
To prevent accident, it is provided with various active safety systems in the car.Man-machine interface (HMI) technology by
Exploitation is for being effectively sent to driver for various warnings against danger.Recently, by using imaging sensor, HMI technology is
It is more widely used identification ambient enviroment.
Most common detection of obstacles/warning system is parking assistance system based on ultrasound.The parking assistance system
Cognitive disorders object, the distance by ultrasonic measurement vehicle away from barrier, and if the distance between vehicle and barrier are small,
Then with audible alert driver.Driver can be whether there is by warning cognitive disorders object, but the location information of barrier
And it is not provided with.In the prior art, the estimated location of barrier is informed by HMI.
When barrier is present in predefined region, detection of obstacles/warning system based on postposition camera passes through
The color (for example, red) for changing the entire profile of obstructions chart picture enables a driver to cognitive disorders object.However, due to existing
Technology changes the color on boundary, and only shows that barrier whether there is, so driver must directly confirm barrier
Position.
In the prior art, detection of obstacles/warning system based on overall view monitoring system (AVM) can check that barrier is
The no all directions for being present in vehicle, but driver must directly confirm the accurate location of barrier.
Highly saturated color is superimposed upon on the barrier detected using HMI or shows barrier in square by offer
In monitoring system.This method intuitively provides the position of the object detected and the presence of barrier.However, due to height
The use of the color of saturation is spent, which can provide inaccurate obstacle information, and when the barrier detected
When enormous amount, the complexity of HMI is increased by, therefore cannot guarantee safe driving.
Summary of the invention
An aspect of of the present present invention provides a kind of method and apparatus for alerting barrier, by catching to by imaging sensor
The image obtained, which carries out brightness regulation, improves the visibility and observability of barrier, and this method and equipment are intuitive by obstacle information
Ground is sent to driver.
The exemplary embodiment conceived according to the present invention, the equipment for alerting vehicle barrier include being configured to by extremely
The image acquiring device of the image of few two imaging sensors shooting vehicle periphery.Image segmentation device is configured to will be by least
Current reference image acquired in an imaging sensor in two imaging sensors is divided with object block.Exercise estimator is matched
It is set to the two dimensional motion information of the object in each segmentation block for estimating to be divided by image segmentation device.Change estimator configuration
Divide the distance between object and the vehicle in block information at estimation.Relative motion estimator is configured to by believing two dimensional motion
Breath and range information combine, the three dimensional relative movement information of the object in estimation segmentation block.Collision time estimator is configured to lead to
Cross the collision time estimated between object and vehicle in segmentation block using three dimensional relative movement information.Image processor is configured to
The brightness of corresponding segmentation block is adjusted with predefined ratio according to collision time.Display is configured to display and passes through image processor
The image of processing.
At least two imaging sensors can be installed on vehicle horizontally disposedly.
Object can be barrier.
Exercise estimator can be by relatively estimating two dimensional motion information by current reference image and previously with reference to image.
By using current reference image with by another imaging sensor obtain control image between correlation
Estimate to change, variation estimator can estimated distance information.
By using current reference image and previously with reference between image time difference and three dimensional relative movement information, collision
Timing estimator can be estimated to divide the collision time between the object and vehicle in block.
If collision time is less than threshold time, image processor can be to the profile addition shade effect of corresponding segmentation block
Fruit.
The present invention can further comprise warning output end, be configured to export police if collision time is less than threshold time
Accuse signal.
Another exemplary embodiment conceived according to the present invention, the method for alerting vehicle barrier include by extremely
The image of few two imaging sensors shooting vehicle periphery.An imaging sensor at least two imaging sensors will be passed through
Acquired current reference image is divided with object block.Estimate the two dimensional motion information of the object in each divided segmentation block
And range information.By the way that by two dimensional motion information and range information combination, the three dimensional relative movement of the object in block is divided in estimation
Information.By using three dimensional relative movement information, the collision time between object and vehicle in block is divided in estimation.Pass through basis
Collision time adjusts the brightness of corresponding segmentation block with predefined ratio to show image.
By comparing current reference image and previously with reference to image and can estimate two dimensional motion information.
By using current reference image with by another imaging sensor obtain control image between correlation
To estimate that variation can estimated distance information.
By using current reference image and previously with reference between image time difference and three dimensional relative movement information, estimation
The collision time between object and vehicle in the step of collision time estimation segmentation block.
This method further comprises adding shade to the profile of corresponding segmentation block if collision time is less than threshold time
Effect.
The step of this method further comprises, exports caution signal if collision time is less than threshold time.
According to the present invention, improve barrier and to brightness regulation is carried out from imaging sensor captured image
Property and observability, can intuitively be sent to driver for obstacle information.
The present invention can provide the presence of the location information of barrier and barrier.
In addition, by using increasing the high contrast effect of barrier brightness in image and using the profile of barrier
The eyes of driver can be directed to barrier by hatching effect, the present invention, and therefore improve the visibility of barrier and considerable
The property examined,
The eyes of driver further can be directed to the big barrier of risk of collision by warning based on collision time by the present invention
Hinder object.
Detailed description of the invention
In conjunction with attached drawing, the above and other target of the invention, feature and advantage will be more aobvious and easy from following detailed description
See.
Fig. 1 is the box configuration diagram for showing the equipment for alerting vehicle barrier for conceiving embodiment according to the present invention.
Fig. 2 is the flow chart for showing the method for alerting vehicle barrier for conceiving embodiment according to the present invention.
Fig. 3 is the exemplary diagram for showing the obstacle information display screen for conceiving embodiment according to the present invention.
Reference signs list:
10: image acquiring device
11: the first imaging sensors
12: the second imaging sensors
20: image segmentation device
30: exercise estimator
40: variation estimator
50: relative motion estimator
60: collision time estimator
70: image processor
80: display
90: warning output end
Specific embodiment
Hereinafter, the embodiment of present inventive concept will be described in detail with reference to the attached drawings.
According to the present invention, by adjusting the brightness by imaging sensor captured image, improve the visible of obstructions chart picture
Property and observability, so as to the information of barrier is intuitive, efficiently send to driver.
Fig. 1 is the box configuration diagram for showing the equipment for alerting vehicle barrier for conceiving embodiment according to the present invention.
As shown in Figure 1, the equipment according to the present invention for alerting vehicle barrier includes image acquiring device 10, image
Segmenting device 20, exercise estimator 30, variation estimator 40, relative motion estimator 50, collision time estimator 60, at image
Manage device 70, display 80 and warning output end 90.
Image acquiring device 10 is installed on vehicle, and obtains the image of vehicle periphery.Image acquiring device 10 includes
The first imaging sensor 11 and the second imaging sensor 12 horizontally disposed and be installed on vehicle.For example, the first camera
It can be horizontally arranged at vehicle front left side and forward right side respectively with the second camera.In the disclosure, the image acquiring device shown
10 include two imaging sensors 11,12, but image acquiring device 10 can have more than two imaging sensor.
An image in the image shot by the first imaging sensor 11 and the second imaging sensor 12 is used as and is worked as
Preceding reference picture.The exemplary embodiment explained below conceived according to the present invention will be shot by the first imaging sensor 11
Image be used as reference picture example.
Image segmentation device 20 will be by current acquired in image acquiring device 10 based on colouring information and profile information
Reference picture is divided with object block.Here, object means barrier.
By comparing the current reference image of each segmentation block with previously with reference to image, exercise estimator 30 estimates each point
Cut two dimensional motion (2D movement) information of block.That is, exercise estimator 30 estimates two of the barrier (object) in each segmentation block
Tie up motion information.
By estimating that the current reference image of each segmentation block is estimated with the variation between image, variation estimator 40 is compareed
Range information in each segmentation block between barrier and its vehicle.Here, control image is the image for not being selected as reference picture,
And it sends from the second imaging sensor 12.
By combining the two dimensional motion information of each segmentation block and range information, relative motion estimator 50 is estimated each
Divide the three dimensional relative movement information of block.In other words, relative motion estimator 50 extracts the three-dimensional phase of the object in segmentation block
To motion information.
By using current reference image and previously with reference between image time difference and three dimensional relative movement information, when collision
Between estimator 60 estimate it is each segmentation block in object and vehicle between collision time (time (the time to away from collision
Collision is abbreviated as TTC)).Here, collision time is until vehicle and barrier collide elapsed time.
The brightness of each segmentation block is actively adjusted according to collision time and arrives pre-set ratio by image processor 70.If collision
Time is less than threshold time, then image processor 70 adds hatching effect to the outmost profile of each segmentation block.I other words
If detected and the risk by the barrier collision in image acquired in image acquiring device 10, image processor 70
Just hatching effect is added to the profile of barrier in respective image.
The image that the output of display 80 is handled by image processor 70.Display 80 be embodied as liquid crystal display (LCD),
Head-up display (HUD), light emitting diode (LED) display, transparent display etc..
If collision time is less than threshold time, warning output end 90 just exports caution signal.At this point, caution signal can be real
Apply the combination of the light for light-emitting component, sound signal, voice signal etc..
Fig. 2 is the flow chart for showing the method for alerting vehicle barrier for conceiving embodiment according to the present invention, and Fig. 3
For the exemplary diagram for showing the obstacle information display screen for conceiving embodiment according to the present invention.Equipment for alerting vehicle barrier
(hereinafter referred to barrier apartment warning) obtains the image of vehicle periphery by least two imaging sensors, and will be obtained
An image in the image taken is set as current reference image (S11).For example, the selection of barrier apartment warning passes through the first image
The image of the first imaging sensor 11 input in sensor 11 and the second imaging sensor 12 is as current reference image.At this
In open, a reference picture is selected, but more than two reference picture also may be selected.
Based on colouring information and profile information, image segmentation device 20 will be inputted current by preset imaging sensor
Reference picture divides (S12) with object block.Here, object can be barrier.
Moreover, barrier apartment warning estimate each block divided by image segmentation device 20 two dimensional motion information and away from
From information (S13).Here, exercise estimator 30 estimates each segmentation by comparing current reference image and previously with reference to image
The two dimensional motion information of the object of block.By using it is each segmentation block current reference image with previously with reference to the phase between image
Guan Xing, the variation estimation variation of estimator 40, and by changing the object and the vehicle of oneself estimated to estimate each segmentation block
Between range information.I other words exercise estimator 30 estimates the two dimensional motion information and range information of object, and change estimation
Device 40 estimates vehicle to the distance of object.
By combining the two dimensional motion information of each segmentation block and range information, relative motion estimator 50 is estimated each
Divide the three dimensional relative movement information (S14) of block.In other words, by by the two dimensional motion information and range information of each object
In conjunction with relative motion estimator 50 estimates the three dimensional relative movement information of each object.
By using current reference image and previously with reference between image time difference and three dimensional relative movement information, collision
Timing estimator 60 estimates the collision time (time (TTC) away from collision) (S15) between object and vehicle in each segmentation block.
Image processor 70 actively adjusts the brightness (S16) of each segmentation block according to collision time with preset ratio.?
This, becomes smaller by the object (barrier) in image acquired in image acquiring device 10 and the collision time between vehicle,
Then each segmentation block just becomes brighter.
If collision time is less than threshold time, the just outmost profile addition to corresponding segmentation block of image processor 70
Hatching effect (S17, S18).In other words, as shown in figure 3, if being examined in through image acquired in image acquiring device 10
The object of risk of collision is measured, image processor 70 just adds hatching effect to the outmost profile of corresponding object.
Then, the image that image processor 70 crosses image procossing is output to display 80.At this point, if collision time is small
In threshold time, the just output caution signal of output end 90 is alerted.
As shown in figure 3, since the present invention identifies barrier in passing through image acquired in imaging sensor, and lead to
It overregulates brightness and improves the visibility and observability of barrier to corresponding barrier application hatching effect, so
Driver can intuitively and promptly identify the barrier in image.
Claims (12)
1. a kind of equipment for alerting vehicle barrier comprising:
Image acquiring device, described image acquisition device are configured to shoot vehicle periphery by least two imaging sensors
Image；
Image segmentation device, described image segmenting device is configured to will be by one at least two imaging sensor
Current reference image acquired in imaging sensor is divided with object block；
Exercise estimator, each segmentation block that the exercise estimator is configured to estimate to be divided by described image segmenting device
In the object two dimensional motion information；
Change estimator, it is described variation estimator be configured to estimate segmentation block in the object and the vehicle between away from
From information；
Relative motion estimator, the relative motion estimator are configured to by by the two dimensional motion information and the distance
Information is in conjunction with come the three dimensional relative movement information of estimating the object；
Collision time estimator, the collision time estimator are configured to estimate by using the three dimensional relative movement information
The collision time between object and the vehicle in the segmentation block；
Image processor, described image processor are configured to adjust corresponding point according to the collision time with predefined ratio
Cut the brightness of block；And
Display, the display are display configured to the image handled by described image processor,
Wherein if the collision time is less than threshold time, described image processor adds to the profile of the corresponding segmentation block
Add hatching effect.
2. equipment according to claim 1, wherein at least two imaging sensor is described in horizontally disposed be mounted on
On vehicle.
3. equipment according to claim 1, wherein the object is barrier.
4. equipment according to claim 1, wherein the exercise estimator passes through by the current reference image and previously
Reference picture relatively estimates the two dimensional motion information.
5. equipment according to claim 1, wherein the variation estimator is by using the current reference image and leads to
The correlation crossed between the control image of another imaging sensor input is estimated to change, to estimate the range information.
6. equipment according to claim 1, wherein the collision time estimator is by using the current reference image
With previously with reference between image time difference and the three dimensional relative movement information, come estimate it is described segmentation block in object and institute
State the collision time between vehicle.
7. equipment according to claim 1, wherein the equipment further includes warning output end, if being configured to described touch
Hitting the time just exports caution signal less than threshold time.
8. a kind of method for alerting vehicle barrier comprising following steps:
The image of vehicle periphery is shot by least two imaging sensors；
Current reference image acquired in an imaging sensor at least two imaging sensor will be passed through with object
Block segmentation；
Estimate the two dimensional motion information and range information of the object in each segmentation block；
By the way that by the two dimensional motion information and range information combination, the three-dimensional that the object in block is divided in estimation is opposite
Motion information；
By using the three dimensional relative movement information, touching between the object and the vehicle in the segmentation block is estimated
Hit the time；
If the collision time is less than threshold time, hatching effect just is added to the profile of corresponding segmentation block；And
By adjusting the brightness of corresponding segmentation block with predefined ratio according to the collision time, described image is shown.
9. according to the method described in claim 8, wherein by by the current reference image and previously with reference to image relatively come
Estimate the two dimensional motion information.
10. according to the method described in claim 8, wherein being passed by using the current reference image with by another image
Correlation between the control image of sensor input is estimated to change, to estimate the range information.
11. according to the method described in claim 8, the step of wherein estimating the collision time is by using the current reference
Image and previously with reference between image time difference and the three dimensional relative movement information come, estimate it is described segmentation block in object
With the collision time between the vehicle.
12. according to the method described in claim 8, its is further comprising the steps of: if the collision time is less than threshold time,
With regard to exporting caution signal.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR20140107907 | 2014-08-19 | ||
| KR10-2014-0107907 | 2014-08-19 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105774649A CN105774649A (en) | 2016-07-20 |
| CN105774649B true CN105774649B (en) | 2019-07-16 |
Family
ID=
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2013058128A1 (en) * | 2011-10-18 | 2013-04-25 | 本田技研工業株式会社 | Vehicle vicinity monitoring device |
| CN103223911A (en) * | 2012-01-30 | 2013-07-31 | 日立民用电子株式会社 | Vehicle collision risk prediction apparatus |
| CN103732480A (en) * | 2011-06-17 | 2014-04-16 | 罗伯特·博世有限公司 | Method and device for assisting a driver in performing lateral guidance of a vehicle on a carriageway |
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103732480A (en) * | 2011-06-17 | 2014-04-16 | 罗伯特·博世有限公司 | Method and device for assisting a driver in performing lateral guidance of a vehicle on a carriageway |
| WO2013058128A1 (en) * | 2011-10-18 | 2013-04-25 | 本田技研工業株式会社 | Vehicle vicinity monitoring device |
| CN103223911A (en) * | 2012-01-30 | 2013-07-31 | 日立民用电子株式会社 | Vehicle collision risk prediction apparatus |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5718942B2 (en) | Apparatus and method for assisting safe operation of transportation means | |
| JP6470335B2 (en) | Vehicle display system and method for controlling vehicle display system | |
| US9965957B2 (en) | Driving support apparatus and driving support method | |
| US10009580B2 (en) | Method for supplementing a piece of object information assigned to an object and method for selecting objects in surroundings of a vehicle | |
| JP5171629B2 (en) | Driving information providing device | |
| CN108263279A (en) | The pedestrian detection and pedestrian impact avoiding device and method integrated based on sensor | |
| US10504237B2 (en) | Method for determining a viewing direction of a person | |
| WO2013042767A1 (en) | Apparatus for monitoring surroundings of vehicle | |
| US20160144786A1 (en) | Method and apparatus for providing driving information | |
| KR101986734B1 (en) | Driver assistance apparatus in vehicle and method for guidance a safety driving thereof | |
| KR101914362B1 (en) | Warning system and method based on analysis integrating internal and external situation in vehicle | |
| JP2006338594A (en) | Pedestrian recognition system | |
| JP2007006006A (en) | Onboard information display device and onboard information display method | |
| JP2010044561A (en) | Monitoring device to be mounted on vehicle | |
| WO2018004421A1 (en) | Method and control unit for a digital rear view mirror | |
| JP2019185218A (en) | Alarm device | |
| US9704062B2 (en) | Method and apparatus for warning an obstacle of a vehicle | |
| KR20160017215A (en) | Safety system for vehicle | |
| KR20120086577A (en) | Apparatus And Method Detecting Side Vehicle Using Camera | |
| JP6914827B2 (en) | Oversight detectors, oversight detection methods, and programs | |
| JP2007133644A (en) | Pedestrian recognition device | |
| JP5192007B2 (en) | Vehicle periphery monitoring device | |
| JP5192009B2 (en) | Vehicle periphery monitoring device | |
| JPWO2017029740A1 (en) | Display control device, display device, and display control method | |
| CN105774649B (en) | Method and apparatus for alerting vehicle barrier |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |