CN101995239B - System and method for determining obstacle by using aerial view images - Google Patents
System and method for determining obstacle by using aerial view images Download PDFInfo
- Publication number
- CN101995239B CN101995239B CN2009101617455A CN200910161745A CN101995239B CN 101995239 B CN101995239 B CN 101995239B CN 2009101617455 A CN2009101617455 A CN 2009101617455A CN 200910161745 A CN200910161745 A CN 200910161745A CN 101995239 B CN101995239 B CN 101995239B
- Authority
- CN
- China
- Prior art keywords
- image
- bird
- eye view
- vehicle
- barrier
- 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
Images
Landscapes
- Image Processing (AREA)
- Image Analysis (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The invention provides a system and a method for determining an obstacle by using aerial view images, relates to technology for determining the obstacle around a vehicle by aerial view image technology. A plurality of image acquisition devices which are arranged at different positions are used for acquiring a plurality of images around a vehicle body respectively, and the plurality of images are at least provided with an overlapping region; an image processor performs independent aerial view image conversion on the images; and an obstacle calculating unit compares the correlation of repeat regions of two independent aerial view images and judges whether the obstacle exists or not according to the size of the correlation. In addition, a mapping table of a group of space conversion data is constructed and comprises driving state data such as the steering wheel angle, gear or speed of the vehicle, and the like, so when the driving state is changed, a panoramic aerial view image at an appropriate viewing angle is quickly converted through the mapping table, and the position of the obstacle measured by the obstacle calculating unit is marked on the panoramic aerial view image; therefore, the safe driving of a user is facilitated.
Description
Technical field
The relevant obstacle detection system of the present invention, especially a kind of employed obstacle detection system when being applied to steering vehicle.
Background technology
The cost of video camera day by day descends, and photographic goods have been widely used in vehicle driving at present and have assisted, and the right only simple mostly passive image that vehicle periphery is provided is avoided the generation of accident with the driver assistance person.Because the driver is difficult to keep watch on simultaneously a plurality of photographic pictures; And the visual dead angle that the conventional planar image is applied to stop when auxiliary is still big; Therefore also having at present the dealer to develop can be through being placed in vehicle body a plurality of video cameras on every side; Its a plurality of images of obtaining are integrated and changed its visual angle and become the virtual ring field overlooked by roof and get a bird's eye view image system, the technological means that this comparatively meets the human eye vision custom and also can further overcome visual dead angle.
But; Generally system is got a bird's eye view when the conversion plane image is got a bird's eye view image for the ring field in the ring field; Because of compound many images of need while,, then can can't show that this has barrier highly on a ring image because of the dead angle of image stitching therefore if when image overlap area occurs having the object of height; So convention ring field image technology still has it must improve part.
In view of this; For the barrier and the assistant that can detect the image suture remind the driver assistance person that there is barrier at this place with picture caution or sound caution; Then conventional art uses radar or ultrasonic unit to detect barrier to combine with photographic image, get a bird's eye view the disappearance of image technology to supply convention ring field more.Such as applying for a patent of patent publication No. JP2007041791 (A) announcement; The system and method for image is got a bird's eye view in a kind of demonstration; It is earlier with the image around a plurality of video camera collection vehicle, its image is stitched into a ring gets a bird's eye view image, whether has barrier near learning the vehicle front, rear, left and right with image; And be to overcome at the image suture to be difficult to show the problem with height barrier, this patent then detects this dead angle area with four ultrasonic units and solves.Though but this kind mode can overcome the problems referred to above, can increase the cost of ultrasonic unit relatively.
Moreover; Two dimension gets a bird's eye view that ring image system is many to be set in the vehicle top with virtual view in the known techniques; Yet, to advance, retreat at vehicle or turn when moving, the imagery zone visual field size of the required moving direction of driver is also from different; But known techniques is only used the picture of getting a bird's eye view at a fixing virtual visual angle, is difficult to meet driver's demand.
Therefore, the present invention proposes a kind of utilization and gets a bird's eye view image disturbance in judgement thing system and method, can effectively solve the shortcoming of prior art.
Summary of the invention
Fundamental purpose of the present invention is got a bird's eye view the system and method for image disturbance in judgement thing for proposing a kind of utilization, and its collection has two of repeat region and gets a bird's eye view image, and relatively the correlativity of its repeat region image the position occurs with the barrier of judging the tool height.
Another object of the present invention is got a bird's eye view the system and method for image disturbance in judgement thing for proposing a kind of utilization, to get a bird's eye view the direct disturbance in judgement object location of image, need not increase other detection of obstacles instruments of radar or ultrasound wave, can reduce production/assembly cost.
A purpose more of the present invention is got a bird's eye view the system and method for image disturbance in judgement thing for proposing a kind of utilization, when signal of video signal import, is used for reaching quick conversion with saving system operation time by making of mapping table.
Another purpose of the present invention is got a bird's eye view the system and method for image disturbance in judgement thing for proposing a kind of utilization; It is by the arrange in pairs or groups mapping table of one group of corresponding steering wheel angle and virtual camera depression angle of traveling state of vehicle; Get a bird's eye view image and export different ring fields, get a bird's eye view image to the driver with what different visual angles were provided.
A purpose more of the present invention is got a bird's eye view the system and method for image disturbance in judgement thing for proposing a kind of utilization; It combines the ring field at convertible visual angle to get a bird's eye view image and utilization to get a bird's eye view image disturbance in judgement thing technology, and then can have more flexible detection of obstacles zone and get a bird's eye view the visual field to driver's reference turning or provide during reversing.
Disclose the system that image disturbance in judgement thing is got a bird's eye view in a kind of utilization for reaching above-mentioned purpose the present invention, comprise:
A plurality of image collection devices, it is located on the vehicle, and these image collection devices are obtained the image of a plurality of these vehicle peripheries;
One image processor converts these images to a plurality of images of independently getting a bird's eye view respectively, and wherein two adjacent these are independently got a bird's eye view image and comprised a repeat region; And
One barrier computing unit, relatively two adjacent these are independently got a bird's eye view the correlativity of this repeat region of image, have judged whether barrier by this correlativity.
A plurality of image collection devices by being set up on the vehicle are gathered image; And to see through image processor be a plurality of images of independently getting a bird's eye view with these video conversion; Whether calculate the two adjacent repeat regions of independently getting a bird's eye view image to obtain the related coefficient of repeat region by barrier computing unit comparison again, judging had the barrier appearance in the zone; These a plurality of images of getting a bird's eye view of image processor combination simultaneously are that image is got a bird's eye view in a vehicle ring field, when the photography zone detects barrier, obstruction marking are got a bird's eye view on the image in this ring field, use and remind the driver.
In addition; The present invention also discloses the method that image disturbance in judgement thing is got a bird's eye view in a kind of utilization; Comprise the vehicle body signal of gathering a vehicle, and select pairing mapping table, and gather near a plurality of images this vehicle; Set up a ring field according to this mapping table and get a bird's eye view image, wherein two these adjacent images comprise a repeat region;
Be a plurality of images of independently getting a bird's eye view with these video conversion respectively;
Judge that adjacent this independently get a bird's eye view the correlativity of this repeat region of image;
Judge whether this repeat region has barrier; And
If this barrier is arranged, this obstruction marking is got a bird's eye view in the image in this ring field.
It gathers earlier a plurality of images around vehicle body signal and the vehicle body, and wherein adjacent image comprises a repeat region, and picks out the mapping table of relevant space conversion information according to the steering wheel angle signal, encircles the field and gets a bird's eye view video conversion; Again these images are converted into respectively and independently get a bird's eye view image; Continue, judge that adjacent this independently get a bird's eye view the correlativity of the repeat region of image; Judge according to this correlativity size whether this repeat region has barrier to exist; At last, if repeat region was then got a bird's eye view this obstruction marking in the image in Huan Chang when having barrier, getting a bird's eye view the image visual angle to change the detection of obstacles scope according to the adjustment of steering wheel angle signal simultaneously.
Beneficial effect of the present invention is; Improve the shortcoming that barrier takes place can't show because of the image dead angle at the image suture known techniques; Utilize the characteristic of when there is the image overlap place in barrier, encircling the distortion that to be distorted after video conversion is got a bird's eye view in the field, further be converted into the means of disturbance in judgement object location, will have two raw videos of overlap; Convert into individually and get a bird's eye view striograph; In two correlativitys of getting a bird's eye view its repeat region of image of comparison, the little person of correlativity then has barrier, and the subregion of analyzing thin portion accurately concludes this barrier position.
Thus, the present invention saves other checkout equipments such as ultrasound wave or radar are installed, and utilizes image technology can record the barrier minimum distance purely.In addition; Drive a vehicle and turn or sensing range when retreating in order to strengthen the driver; In the video conversion technology, the present invention encircles the field with a plurality of mapping tables and gets a bird's eye view video conversion, and wherein each mapping table comprises the space conversion information under each driving states; So getting a bird's eye view the virtual visual angle of image, ring of the present invention field can change according to the driving situation; And native system shows that the detection of obstacles scope also because of the change at virtual visual angle changes, is more humane driving servicing unit, so the driver is when using technological means of the present invention; Image is got a bird's eye view in the ring field that different visual angles under the situation such as can obtain to keep straight on, turn, retreat, thereby can more safely drive and set out on a journey.
Beneath by the appended graphic explanation in detail of specific embodiment cooperation, when the effect that is easier to understand the object of the invention, technology contents, characteristics and is reached.
Description of drawings
Fig. 1 is the system schematic of one embodiment of the invention;
Fig. 2 (a) is the synoptic diagram of image camera coverage of the present invention to Fig. 2 (e);
Fig. 3 is the synoptic diagram of the related coefficient of calculating repeat region;
Fig. 4 is got a bird's eye view the synoptic diagram of image for the present invention judges the ring field of clear;
Fig. 5 is got a bird's eye view the synoptic diagram of image for the present invention judges the ring field that barrier is arranged;
Fig. 6 is got a bird's eye view the image synoptic diagram for the ring field at the convertible virtual visual angle of the present invention;
Fig. 7 is the flow chart of steps of one embodiment of the invention.
Description of reference numerals: 10-image collection device; 12-vehicle body signal picker; The 14-image processor; The 141-mapping table; The 16-display; 18-barrier computing unit; The 20-alarming device; 211～216-repeat region; Image is got a bird's eye view in 22-ring field; The 24-vehicle body; 28-related coefficient γ is less than 0.5 zone; The 30-lane line; 32-is got a bird's eye view the image suture line; 34, the 36-raw video; 38,40-is independently got a bird's eye view image; 381, the 401-repeat region; 42, the 44-vehicle; 46-cylinder type barrier; 48, the 50-raw video; 52,54-is independently got a bird's eye view image; 521, the 541-repeat region; The 60-vehicle body; Image capturing range is got a bird's eye view in 62-ring field; The virtual visual angle of 64-; The 66-barrier.
Embodiment
The video conversion that major technology means of the present invention are gathered different video camera is two independently to get a bird's eye view image; Again to this two independently get a bird's eye view image repeat region do correlativity comparison, comparison result is if the high person of its correlativity judges that then there is not the barrier of tool height in this zone; Otherwise, if the low person of correlativity then judges the barrier that has the tool height.Come accurately to judge whether to have barrier and its position and distance with the method.
At first please refer to Fig. 1; System architecture diagram for one embodiment of the invention; In the present embodiment, gather vehicle body image on every side with six image collection devices of being located on the car 10, all there is partly overlapping the same area image in the image that wherein every adjacent image collection device 10 is gathered; Unification is called repeat region under this area image, and image collection device 10 is charge coupled cell (CCD) video camera or complementary metal oxide semiconductor (CMOS) video camera.Present embodiment also is combined with vehicle body signal picker 12 to obtain the driving situation, and the vehicle body signal comprises steering wheel angle angle, GES or gear signal, the data that is provided by image processor 14 reception raw videos and vehicle body signal again.Mainly judge the turn inside diameter situation in the present embodiment with the steering wheel angle signal; Therefore according to the steering wheel angle signal of vehicle body; Image processor 14 selects correct suitable mapping table 141 to carry out the image space conversion; Wherein mapping table 141 establishes the space conversion information of relevant camera position and steering wheel angle angle earlier before present embodiment is carried out; Mapping table 141 is also noted down simultaneously has skew control, have an X-rayed information such as conversion and image stitching, is to get a bird's eye view image with raw video through the ring field that mapping table can be converted into multiple visual angle fast.And image processor 14 is inserted in raw video in the mapping table 141 and export the ring field general view at suitable virtual visual angle, again it is output in a display 16, and this display 16 shows the driving boost line or the parking boost line of this vehicle periphery.
Detect in order to carry out the vehicle body peripheral obstacle; Image processor 14 of the present invention is except changing a holistic ring general view; Must six raw videos be converted into individually independently to get a bird's eye view image and then obtain six and independently get a bird's eye view behind the image and compare its repeat region and judge that again two independently get a bird's eye view the correction of image property of the repeat region of image, with this as the means that detect barrier.If wherein correlativity is high, there is not the barrier of tool height in the expression repeat region; Otherwise,, represent that then there is the barrier of tool height in repeat region if correlativity is low.And present embodiment is got a bird's eye view measured its position of barrier demonstration on the image in Huan Chang, further can install alarming device 20 additional, and it can be light-emittingdiode, hummer or loudspeaker, reminds driver's attention deficit disorder thing to send alarm signal.
Continue, the repeat region how further explain the present invention independently gets a bird's eye view image by comparison two is to judge the technological means of barrier.Known techniques obtains at least four raw videos of vehicle's surroundings earlier; Can get a ring field of overlooking after synthesizing via image processing again and get a bird's eye view image; But in the image calculation processes, need ground is regarded as a flat surfaces, also promptly can only the object of tool height be regarded as planar object and carry out image processing.So the object of the tool height phenomenon that when getting a bird's eye view video conversion, will be distorted causes obtaining effectively ring and gets a bird's eye view image, this is the shortcoming of known techniques.The present invention then uses gets a bird's eye view the unique point that image is distorted, and comes embodiment of the present invention to detect the technology of barrier.
The present invention gathers the vehicle body image in six zones on every side by six image collection devices 10, converts a ring field into through image processor 14 again and gets a bird's eye view image 22, and wherein this image collection device 10 is a video camera.Please refer to Fig. 2 (a), the present invention gathers forward and backward a-quadrant of vehicle body and B zone respectively by first video camera and second video camera that are placed on the vehicle 24; The 3rd video camera and four cameras are gathered left and right C zone of vehicle body 24 and D zone respectively, shown in Fig. 2 (b); And the 5th video camera and the 6th video camera gather vehicle body 24 left and right E zone and F zone respectively, shown in Fig. 2 (c).The image of wherein above-mentioned per two adjacent areas all has a repeat region; Promptly the hatched example areas in Fig. 2 (d) 211,212,213,214,215, shown in 216; This also is the zone of detection of obstacles; Ring field image capturing range after those repeat regions are sewed up through image processing then is vehicle body 24 near zones 22, promptly shown in Fig. 2 (e).
Continue, image processor 14 converts the raw video in above-mentioned A, B, C, D, E, six zones of F separately into and compares two adjacent repeat regions of getting a bird's eye view image respectively by barrier computing unit 18 again after independently getting a bird's eye view image, and judges its correlativity.Again with the a-quadrant among Fig. 2 (a) independently get a bird's eye view image with Fig. 2 (c) in the image of independently getting a bird's eye view in E zone be example, this a-quadrant can comprise the identical repeat region 212 shown in Fig. 2 (e) with the E zone.Please continue with reference to figure 3; (s t) represents the repeat region GTG value of a-quadrant, w (s to f among the figure; T) represent the regional repeat region GTG value of E; Image processor 14 is cut into nine sub regions with repeat region, calculate the related coefficient γ of the relative subregion GTG value in two zones, and the computing formula of related coefficient does
Wherein f and w are the average GTG value of a-quadrant and E zone repeat region image.The meaning of this formula is got a bird's eye view image with each single-point difference in zone and the difference value of accumulating every single-point for calculating two, and then obtains two related coefficient γ of getting a bird's eye view image.If two images of getting a bird's eye view repeat region between image are identical, then related coefficient is 1; If the image in two repetition zones is uncorrelated fully, its related coefficient then is 0.With the threshold value setting of disturbance in judgement thing related coefficient is 0.5 to be example, if on behalf of this subregion, related coefficient less than this threshold values, have barrier.Like the zone among Fig. 3 28, because of related coefficient γ less than 0.5, be that this zone of decidable 28 has barrier to exist with this.Be with, getting a bird's eye view image during for flat surfaces when repeat region can distortion distortion, and the related coefficient γ of gained representes then that greater than threshold values 0.5 repeat region does not have barrier to exist.
Please continue and consult Fig. 4, judge the synoptic diagram of clear in order to explanation the present invention.As shown in the figure, vehicle periphery does not have barrier to exist among the figure, and vehicle 42 sides are two to have four and get a bird's eye view image suture line 32 in the lane line 30 on ground and figure.When beginning to judge when clear is arranged; Be positioned at video camera elder generation collection vehicle 42 the place ahead raw videos 36 of vehicle front side; And the video camera collection vehicle 42 right side raw videos 34 on right side, do not convert into through 14 of image processors again and independently get a bird's eye view image 38,40, the repeat region of getting a bird's eye view image 381,401 in Fig. 4 after visible the place ahead and the right side conversion; And judge the correlativity size of the repeat region get a bird's eye view image here via the barrier computing unit; Because this routine repeat region 381,401 be smooth road surface, so judged result is that correlativity is big, then represents clear existence here.
Fig. 5 then judges the synoptic diagram that barrier is arranged for the present invention; Be example with vehicle 44 the place ahead images and side image equally; When a cylinder type barrier 46 appears in vehicle 44 right fronts, the raw video 48,50 that image collection device 10 is gathered convert into two independently get a bird's eye view image 52,54 after because this cylinder type barrier 46 has highly; Therefore carrying out video conversion for after getting a bird's eye view image; Cylinder type barrier 46 bodies can distortion distortion, causes the correlativity of two repeat regions 521,541 of getting a bird's eye view image 52 and 54 low, can distinguish that to this this zone has barrier and exists.
So via above-mentioned image processing process, the present invention can be set in repeat region driver's visual dead angle place, see through the present invention and judge the result of repeat region can know whether there is barrier.
In addition, the present invention also the related coefficient of repeat region capable of using change and come the disturbance in judgement object location.With the above embodiment of the present invention is example, calculates earlier the related coefficient of nine sub regions in two repeat regions of getting a bird's eye view image respectively, if if there is a barrier to be arranged in the parton zone of this repeat region, its related coefficient can present little value; Otherwise other that barrier does not fall in this repeat region are subregion partly, and its related coefficient then presents big value.In other words, the subregion of the clear in this repeat region and the subregion intersection that barrier is arranged, its related coefficient die-offs because barrier occurs, and then utilizes the present invention of this characteristic can calculate this barrier of acquisition from the nearest distance of vehicle body.More than be principal character of the present invention technology, to the dead angle of video camera and the shortcoming of picture distortion after getting a bird's eye view video conversion, the present invention compares the correlativity after image is got a bird's eye view in conversion, with acquired disturbance object location accurately.Compared to known techniques, the present invention more can strengthen the effect of getting a bird's eye view the relevant detection of obstacles of image technology.In addition, to get a bird's eye view image output and more press close to human nature in order to make, the present invention also proposes a new improvement to the disposal route that image is got a bird's eye view in the ring field.
The technology that connecting needle is got a bird's eye view image to ring of the present invention field further specifies; The invention is characterized in and increased one group of mapping table; In this embodiment; Mapping table for the total system device set up finish after; Set each camera post earlier in the spatial information of car body, and set up the coordinate transitional information of the pairing virtual view conversion of the separated steering wheel angle of each (5 degree, 10 degree, 15 degree ...), so mapping table has write down the transfer process that a unit point of raw video corresponds to another unit point of getting a bird's eye view image.Therefore; When image collection device 10 collects image; Can cooperate steering wheel angle signal at that time and then select pairing mapping table; To obtain again image be inserted in this mapping table can be with the signal of video signal fast processing, wherein the present invention is got a bird's eye view the video conversion computing method and is utilized video camera coordinate transition matrix, virtual view coordinate transition matrix, perspective conversion formula and anti-perspective conversion to carry out video conversion, and then obtains a two-dimentional vehicle ring field general view.
Below lift instance explanation: when the driver controls vehicle and bends to right; As shown in Figure 6; Image processor receives vehicle body 60 signals; Image is seen through the mapping table conversion in advance configured, get a bird's eye view image capturing range 62 to the driver with the ring field of the right front maximum region of headstock that virtual visual angle 64 is appeared that the inclined left rear is provided, and detect being present in this barrier of getting a bird's eye view image capturing range 66.Above-mentioned is an advantage of the present invention; Be because the barrier of vehicle periphery is not all can threaten to vehicle completely; And barrier computing unit of the present invention does not need the four corner of monitoring round of vehicle yet; The direction that the present invention can be only advances to vehicle and adjust the detection of obstacles zone, making detection of obstacles scope and method more is wisdom.
Be that flow chart of steps of the present invention is also simultaneously with reference to figure 1 please with reference to Fig. 7.At first; Step S10 is that vehicle body signal picker 12 is gathered the vehicle body signal earlier; Wherein this vehicle body signal can be steering wheel for vehicle angular signal, gear signal or GES; And step S12 is by six raw videos that are arranged on around image collection device 10 collection vehicle on the vehicle, and the image that wherein adjacent image collection device 10 is gathered has at least one overlapping region; After obtaining image and vehicle body signal, like step S14, image processor 14 is selected pairing mapping table 141 according to the vehicle body signal, and the mapping table of wherein selecting carries out skew control, perspective conversion to image and sew up image getting a bird's eye view image to produce a ring field; Obtain this ring field and get a bird's eye view image, like step S20; Image is got a bird's eye view in this rings field of display 16 outputs, like step S22; Continue, image processor 14 converts six raw videos into respectively else independently gets a bird's eye view image, like step S24; Shown in step S26, barrier computing unit 18 is judged the two adjacent correlativitys of independently getting a bird's eye view its repeat region of image, and present embodiment is cut into a plurality of subregions with repeat region earlier, and the related coefficient that calculates subregion wherein the formula system of related coefficient do
Wherein, and f (s, t) (s t) is respectively two adjacent these and independently gets a bird's eye view this subregion GTG value of this repeat region of image, and f and w are respectively two adjacent these and independently get a bird's eye view the average GTG value of this subregion of this repeat region of image with w; Whether whether come the disturbance in judgement thing to exist less than threshold values according to related coefficient, wherein on behalf of clear, related coefficient then occur greater than threshold values, related coefficient representes that then barrier exists, like step S28 less than threshold values; If there is barrier to exist, then gets a bird's eye view sign barrier position in the image, and see through alarming device 20 alarm and reminding drivers, like step S30 in this ring field; Exist like clear, then this flow process finishes operation.
Via above-mentioned explanation; At the image suture shortcoming of barrier takes place can't to show because of the image dead angle it is thus clear that the present invention improves known techniques, utilization of the present invention encircles the characteristic of the distortion that must be distorted after video conversion is got a bird's eye view in the field when there is the image overlap place in barrier, further be converted into the means of disturbance in judgement object location; Two raw videos that will have overlap; Convert into individually and get a bird's eye view striograph, in two correlativitys of getting a bird's eye view its repeat region of image of comparison, the little person of correlativity then has barrier; And the subregion of analyzing thin portion accurately concludes this barrier position; Thus, the present invention saves other checkout equipments such as ultrasound wave or radar are installed, and utilizes image technology can record the barrier minimum distance purely.In addition; Drive a vehicle and turn or sensing range when retreating in order to strengthen the driver; In the video conversion technology, the present invention encircles the field with a plurality of mapping tables and gets a bird's eye view video conversion, and wherein each mapping table comprises the space conversion information under each driving states; So getting a bird's eye view the virtual visual angle of image, ring of the present invention field can change according to the driving situation; And native system shows that the detection of obstacles scope also because of the change at virtual visual angle changes, is more humane driving servicing unit, so the driver is when using technological means of the present invention; Image is got a bird's eye view in the ring field that different visual angles under the situation such as can obtain to keep straight on, turn, retreat, thereby can more safely drive and set out on a journey.
The above description of this invention is illustrative, and nonrestrictive, and those skilled in the art is understood, and within spirit that claim limits and scope, can carry out many modifications, variation or equivalence to it, but they will fall in protection scope of the present invention all.
Claims (10)
1. the system of image disturbance in judgement thing is got a bird's eye view in a utilization, it is characterized in that, comprises:
A plurality of image collection devices, it is located on the vehicle, and these image collection devices are obtained the image of a plurality of these vehicle peripheries;
One image processor converts these images to a plurality of images of independently getting a bird's eye view respectively, and wherein two adjacent these are independently got a bird's eye view image and comprised a repeat region; And
One barrier computing unit, relatively two adjacent these are independently got a bird's eye view the correlativity of this repeat region of image, have judged whether barrier by this correlativity.
2. the system of image disturbance in judgement thing is got a bird's eye view in utilization as claimed in claim 1; It is characterized in that; More comprise the vehicle body signal picker, it gathers steering wheel angle signal, GES or the gear signal of this vehicle, the signal that this image processor is gathered according to this vehicle body signal picker; Change out a ring field and get a bird's eye view image, the steering wheel angle signal transformation of the virtual visual angle of image along with this vehicle is got a bird's eye view in this ring field.
3. the system of image disturbance in judgement thing is got a bird's eye view in utilization as claimed in claim 2; It is characterized in that; This image processor more comprises a plurality of mapping tables; It is the space conversion information of position of steering wheel angle angle and these image collection devices of relevant this vehicle, and this mapping table with this image carry out that image distortion is proofreaied and correct, the conversion of image perspective and image sew up, and gets a bird's eye view image and convert this ring into.
4. the system of image disturbance in judgement thing is got a bird's eye view in utilization as claimed in claim 2; It is characterized in that; More comprise a display, electrically connect this image processor, image is got a bird's eye view in this ring field of this display output; And the position display of this barrier is got a bird's eye view in the image in this ring field, and this display shows the driving boost line or the parking boost line of this vehicle periphery.
5. the system of image disturbance in judgement thing is got a bird's eye view in utilization as claimed in claim 1; It is characterized in that; This barrier computing unit cuts into a plurality of subregions respectively with two these adjacent these repeat regions of independently getting a bird's eye view image; And calculate the two the related coefficient of this subregion that two adjacent these are independently got a bird's eye view this repeat region of image, and the formula of this related coefficient does
F (s; T) with w (s; T) be respectively two adjacent these and independently get a bird's eye view this subregion GTG value of image; F and w are respectively two adjacent these and independently get a bird's eye view the average GTG value of this subregion of image, judge with this related coefficient whether this subregion has this barrier, and then obtain the distance of this vehicle and this barrier.
6. the method for image disturbance in judgement thing is got a bird's eye view in a utilization, it is characterized in that said method comprises step:
Gather the vehicle body signal of a vehicle, and select pairing mapping table, and gather near a plurality of images this vehicle, set up a ring field according to this mapping table and get a bird's eye view image, wherein two these adjacent images comprise a repeat region;
Be a plurality of images of independently getting a bird's eye view with these video conversion respectively;
Judge that adjacent this independently get a bird's eye view the correlativity of this repeat region of image;
Judge whether this repeat region has barrier; And
If this barrier is arranged, this obstruction marking is got a bird's eye view in the image in this ring field.
7. the method for image disturbance in judgement thing is got a bird's eye view in utilization as claimed in claim 6, it is characterized in that, this ring virtual view of getting a bird's eye view image is according to this steering wheel angle signal transformation.
8. the method for image disturbance in judgement thing is got a bird's eye view in utilization as claimed in claim 6; It is characterized in that; Judge that two adjacent these independently get a bird's eye view the step of this correlativity of this repeat region of image; More comprise two these adjacent these repeat regions of independently getting a bird's eye view image are cut into a plurality of subregions respectively, and calculate the two the related coefficient of this subregion that two adjacent these are independently got a bird's eye view this repeat region of image, the formula of this related coefficient does
F (s; T) with w (s; T) be respectively two adjacent these and independently get a bird's eye view this subregion GTG value of this repeat region of image; F and w are respectively two adjacent these and independently get a bird's eye view the average GTG value of this subregion of this repeat region of image, judge with this related coefficient whether this subregion has this barrier, and then obtain the distance of this vehicle and this barrier.
9. the method for image disturbance in judgement thing is got a bird's eye view in utilization as claimed in claim 6, it is characterized in that the space conversion information of this mapping table for setting up according to this vehicle body signal and these image collection device positions.
10. the method for image disturbance in judgement thing is got a bird's eye view in utilization as claimed in claim 6, it is characterized in that, this vehicle body signal comprises steering wheel angle signal, gear signal or GES.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009101617455A CN101995239B (en) | 2009-08-11 | 2009-08-11 | System and method for determining obstacle by using aerial view images |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009101617455A CN101995239B (en) | 2009-08-11 | 2009-08-11 | System and method for determining obstacle by using aerial view images |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101995239A CN101995239A (en) | 2011-03-30 |
| CN101995239B true CN101995239B (en) | 2012-05-23 |
Family
ID=43785707
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009101617455A Active CN101995239B (en) | 2009-08-11 | 2009-08-11 | System and method for determining obstacle by using aerial view images |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101995239B (en) |
Families Citing this family (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102970460B (en) * | 2011-08-31 | 2015-08-26 | 帷享科技有限公司台中分公司 | The method of adjustment vehicle image device and system thereof |
| CN103987582B (en) * | 2011-12-15 | 2017-02-22 | 松下知识产权经营株式会社 | Drive assistance device |
| CN102774324B (en) * | 2012-07-06 | 2014-12-10 | 广东好帮手电子科技股份有限公司 | Panoramic parking assist system and implementation method thereof |
| KR101916467B1 (en) * | 2012-10-30 | 2018-11-07 | 현대자동차주식회사 | Apparatus and method for detecting obstacle for Around View Monitoring system |
| US9834143B2 (en) * | 2013-05-23 | 2017-12-05 | GM Global Technology Operations LLC | Enhanced perspective view generation in a front curb viewing system |
| US9886636B2 (en) * | 2013-05-23 | 2018-02-06 | GM Global Technology Operations LLC | Enhanced top-down view generation in a front curb viewing system |
| TWI533694B (en) * | 2013-12-18 | 2016-05-11 | 歐特明電子股份有限公司 | Obstacle detection and display system for vehicle |
| CN104811663B (en) * | 2015-04-17 | 2018-09-04 | 浙江吉利汽车研究院有限公司 | The image processing method and processing unit of vehicle |
| CN104760537A (en) * | 2015-04-21 | 2015-07-08 | 重庆大学 | Novel vehicle-mounted safe driving assistance system |
| JP6561824B2 (en) * | 2015-12-18 | 2019-08-21 | 株式会社デンソー | Display control device |
| TWI585723B (en) * | 2015-12-31 | 2017-06-01 | 合盈光電科技股份有限公司 | Vehicle monitoring system and method thereof |
| TWI607911B (en) * | 2016-04-28 | 2017-12-11 | H P B Optoelectronic Co Ltd | Vehicle safety protection system and method |
| TWI650018B (en) * | 2017-07-18 | 2019-02-01 | 晶睿通訊股份有限公司 | Method for providing user interface for scene stitching of scene and electronic device thereof |
| CN110406472A (en) * | 2018-04-27 | 2019-11-05 | 华创车电技术中心股份有限公司 | Parking assistance system and parking assistance method |
| CN110525360B (en) * | 2019-08-26 | 2022-04-26 | 合创汽车科技有限公司 | Vehicle-mounted terminal-based driving assisting method, device and system and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006253872A (en) * | 2005-03-09 | 2006-09-21 | Toshiba Corp | Apparatus and method for displaying vehicle perimeter image |
| JP2007036668A (en) * | 2005-07-27 | 2007-02-08 | Nissan Motor Co Ltd | System and method for displaying bird's-eye view image |
| CN101442618A (en) * | 2008-12-31 | 2009-05-27 | 葛晨阳 | Method for synthesizing 360 DEG ring-shaped video of vehicle assistant drive |
-
2009
- 2009-08-11 CN CN2009101617455A patent/CN101995239B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006253872A (en) * | 2005-03-09 | 2006-09-21 | Toshiba Corp | Apparatus and method for displaying vehicle perimeter image |
| JP2007036668A (en) * | 2005-07-27 | 2007-02-08 | Nissan Motor Co Ltd | System and method for displaying bird's-eye view image |
| CN101442618A (en) * | 2008-12-31 | 2009-05-27 | 葛晨阳 | Method for synthesizing 360 DEG ring-shaped video of vehicle assistant drive |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101995239A (en) | 2011-03-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101995239B (en) | System and method for determining obstacle by using aerial view images | |
| US11971475B2 (en) | 2D MIMO vehicular radar sensing system | |
| JP7342197B2 (en) | Imaging device and method of controlling the imaging device | |
| US8315433B2 (en) | Obstacle determination system and method implemented through utilizing bird's-eye-view images | |
| US11657604B2 (en) | Systems and methods for estimating future paths | |
| CN102745138B (en) | Dual view-field dynamic-trajectory reverse image system | |
| US8559674B2 (en) | Moving state estimating device | |
| US20130162825A1 (en) | Avm top view based parking support system | |
| EP1895766B1 (en) | Camera with two or more angles of view | |
| US11027667B2 (en) | Display method and apparatus | |
| US20130208118A1 (en) | Vehicular infrared night assistant driving system | |
| JP2005045602A (en) | Field of view monitor system for vehicle | |
| CN105015418B (en) | Vehicle-carrying display screen video switching method | |
| CN106169258A (en) | For the method and apparatus assisting trailer reversing process | |
| JP5951785B2 (en) | Image processing apparatus and vehicle forward monitoring apparatus | |
| CN102303563A (en) | System and method for prewarning front vehicle collision | |
| JP3960153B2 (en) | Vehicle periphery monitoring device | |
| CN202847561U (en) | Car-backing image system | |
| CN102616184A (en) | Vehicle backup image system and vehicle with same | |
| CN104670089A (en) | Panoramic driving monitoring and alarming system | |
| CN111753741A (en) | Special vehicle auxiliary driving method and device based on deep learning and storage medium | |
| Farooq et al. | Decisive Data using Multi-Modality Optical Sensors for Advanced Vehicular Systems | |
| CN217598448U (en) | Visual monitored control system of vehicle blind area and car | |
| WO2021132227A1 (en) | Information processing device, sensing device, moving body, and information processing method | |
| WO2020153262A1 (en) | Measurement device and ranging device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |