CN105818746A - Calibration method and system of panoramic advanced driver assistance system - Google Patents
Calibration method and system of panoramic advanced driver assistance system Download PDFInfo
- Publication number
- CN105818746A CN105818746A CN201510007598.1A CN201510007598A CN105818746A CN 105818746 A CN105818746 A CN 105818746A CN 201510007598 A CN201510007598 A CN 201510007598A CN 105818746 A CN105818746 A CN 105818746A
- Authority
- CN
- China
- Prior art keywords
- fixed point
- coordinate
- image information
- point
- calibration
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Abstract
The present invention provides a calibration method and system of a panoramic advanced driver assistance system. The calibration method comprises the steps of setting corresponding screen detection areas according to preset sample car calibration point coordinates corresponding to each camera device by the calibration system, wherein the sample car calibration point coordinates are corresponding to physical calibration point coordinates on a calibration site; obtaining image information captured by each camera device, selecting angular points according with preset calibration point characteristic information in the corresponding screen detection area in the image information, and screening calibration points for calibration; and establishing relationship between coordinate systems of the image information corresponding to each camera device and coordinate systems where the physical calibration points are located on basis of relationship between the calibration point coordinates and the physical calibration point coordinates. A range of selecting the vehicle calibration point coordinates can be narrowed, and execution efficiency of calibration is improved.
Description
Technical field
The present invention relates to the calibration technique of a kind of DAS (Driver Assistant System), particularly relate to a kind of scaling method looking around senior DAS (Driver Assistant System) and system.
Background technology
In recent years, accepted by increasing user based on the senior DAS (Driver Assistant System) (also cry and look around ADAS system) looking around photographic head.By left and right before and after automobile, wide-angle camera is installed, gather the image of vehicle body surrounding, process through Algorithm Analysis, there is provided the various senior auxiliary parked including auxiliary, deviation warning, blind area vehicle detection include panorama to drive function to driver, have great importance for improving the active safety of automobile.
Before beginning to use and looking around ADAS system, need the corresponding relation setting up between image coordinate and physical coordinates.The foundation of described corresponding relation is that the demarcation by photographic head completes.The angle installed with photographic head due to the calibration result of photographic head, the most closely related, so actual production when, the demarcation of four-way CCD camera just can must complete after photographic head is installed on vehicle body.Although current calibration process is computer software automatic Calibration.But the automatic Calibration that its calibration process may be caused software by demarcating the outside cause such as space enrironment, light is inaccurate, it is necessary to manually repeatedly intervenes, adjust, or demarcate unsuccessfully, causes production efficiency to decline.On the production line of actual depot, general one is produced line only one of which demarcation room, this means that the demarcation of the four-way CCD camera of every chassis have to interior the shortest when be quickly accomplished, how can complete fast and accurately demarcate, be those skilled in the art need solve problem.
Summary of the invention
The shortcoming of prior art in view of the above, it is an object of the invention to provide a kind of scaling method looking around senior DAS (Driver Assistant System) and system, needs manual intervention repeatedly, inefficient problem for solving calibration process in prior art.
For achieving the above object and other relevant purposes, the present invention provides a kind of scaling method looking around senior DAS (Driver Assistant System), it is applied to be provided with in the vehicle of multiple camera head, including: detect region according to each sample car corresponding screen of fixed point setting coordinate of the default each camera head of correspondence, wherein, each described sample car fixed point coordinate is corresponding with Calibration Field each physics fixed point coordinate on the ground；Obtain in the image information that each described camera head is absorbed, and the corresponding screen detection region in each described image information, select the angle point meeting preset calibrations point characteristic information, and therefrom filter out the fixed point for demarcating；Corresponding relation based on each described fixed point coordinate Yu each described physics fixed point coordinate, sets up the coordinate system of image information corresponding to each described camera head and the corresponding relation of each described physics fixed point place coordinate system.
Preferably, the mode selecting the angle point meeting fixed point characteristic information described in includes: transfer the image information portion in described screen detection region to gray level image information part；The angle point meeting preset calibrations point characteristic information is chosen from described gray level image information.
Preferably, described therefrom filtering out includes for the mode of fixed point demarcated: centered by sample car fixed point coordinate, presetted pixel be that the angle point in the range of radius respectively filters out a fixed point.
Preferably, if the quantity of the angle point after Shai Xuan is less than the quantity of corresponding sample car fixed point, the mode of the described fixed point therefrom filtered out for demarcating also includes: shown by each described angle point of corresponding same camera head, so that technical staff intervenes, and each described angle point after reception technique personnel intervention；And, using each angle point after each angle point of being intervened and screening as the fixed point of corresponding described same camera head.
Based on above-mentioned purpose, the present invention also provides for a kind of calibration system looking around senior DAS (Driver Assistant System), it is applied to be provided with in the vehicle of multiple camera head, including: the first module, for each sample car fixed point setting coordinate corresponding screen detection region according to each camera head of correspondence preset, wherein, each described sample car fixed point coordinate is corresponding with Calibration Field each physics fixed point coordinate on the ground；Second module, for obtaining the image information that each described camera head is absorbed, and in the corresponding screen detection region in each described image information, selects the angle point meeting preset calibrations point characteristic information, and therefrom filters out the fixed point for demarcating；Three module, for corresponding relation based on each described fixed point coordinate Yu each described physics fixed point coordinate, sets up coordinate system and the corresponding relation of each described physics fixed point place coordinate system of image information corresponding to each described camera head.
Preferably, described second module includes: the 2nd 1 submodule, for transferring the image information portion in described screen detection region to gray level image information part；Two or two submodule, for choosing the angle point meeting preset calibrations point characteristic information from described gray level image information.
Preferably, described second module includes: the two or three submodule, for centered by sample car fixed point coordinate, presetted pixel be that the angle point in the range of radius respectively filters out a fixed point.
Preferably, if the quantity of the angle point after Shai Xuan is less than the quantity of corresponding sample car fixed point, described second module also includes: the two or four submodule, for each described angle point of corresponding same camera head is shown, so that technical staff intervenes, and reception technique personnel intervene after each described angle point, and, using each angle point after each angle point of being intervened and screening as the fixed point of corresponding described same camera head.
As mentioned above, the scaling method looking around senior DAS (Driver Assistant System) of the present invention and system, have the advantages that by coming setting screen detection region according to the sample car fixed point coordinate preset, the scope choosing described vehicle calibration point coordinates can greatly be reduced, improve the execution efficiency of demarcation；Simultaneously, utilize the corresponding relation of sample car fixed point coordinate and corresponding physics fixed point, it is easy to the deviation obtaining selected described vehicle calibration point coordinates place coordinate system more accurately with physics fixed point place coordinate system, the corresponding relation of Two coordinate system can be effectively improved, provide panorama service accurately for the auxiliary driving such as park of follow-up panorama；It addition, further screened according to the described vehicle calibration point coordinates selected by fixed point characteristic information, it is possible to remove demarcating the interference of the image of similar physical fixed point in place, improve the accuracy of selected described vehicle calibration point coordinates further；Further, when the discrepancy of quantity of selected described vehicle calibration point coordinates with corresponding sample car fixed point coordinate, the most also technical staff can be facilitated disposably to complete calibration experiment by manually adjusting；Additionally, use gray level image information it can be avoided that the ambient impact on calibration process, it is the most fixed to improve further.
Accompanying drawing explanation
Fig. 1 is shown as the flow chart of the scaling method looking around senior DAS (Driver Assistant System) of the present invention.
Fig. 2 is shown as the flow chart of a kind of preferred version of the scaling method looking around senior DAS (Driver Assistant System) of the present invention.
Fig. 3 is shown as the structural representation of the calibration system looking around senior DAS (Driver Assistant System) of the present invention.
Fig. 4 is shown as the structural representation of a kind of preferred version of the calibration system looking around senior DAS (Driver Assistant System) of the present invention.
Detailed description of the invention
Below by way of specific instantiation, embodiments of the present invention being described, those skilled in the art can be understood other advantages and effect of the present invention easily by the content disclosed by this specification.The present invention can also be carried out by the most different detailed description of the invention or apply, and the every details in this specification can also carry out various modification or change based on different viewpoints and application under the spirit without departing from the present invention.It should be noted that, in the case of not conflicting, the feature in following example and embodiment can be mutually combined.
Referring to Fig. 1, the present invention provides a kind of scaling method looking around senior DAS (Driver Assistant System).Described scaling method is mainly performed by calibration system.Described calibration system belongs to the part in the DAS (Driver Assistant System) installed in the car, comprises multiple camera head in described automobile.Common, described automobile comprises and lays respectively at two rearview mirrors of vehicle, headstock, the camera head of the tailstock.The image information that each camera head is absorbed is processed by the demarcation of described calibration system, it is possible to avoid ghost image, blind area, to realize the purpose representing vehicle's surroundings scene of panorama.
In step sl, described calibration system is according to each sample car fixed point setting coordinate corresponding screen detection region of each camera head of correspondence preset, and wherein, each described sample car fixed point coordinate is corresponding with Calibration Field each physics fixed point coordinate on the ground.
Here, each camera head is to absorbing Calibration Field multiple physics fixed points on the ground.In order to the fixed point of picked-up and described physics fixed point are carried out correspondence, each sample car fixed point coordinate of corresponding each camera head, for described calibration system place vehicle, is inputted in described calibration system by technical staff in advance.The each physics fixed point that each sample car fixed point coordinate inputted is Calibration Field ground screen coordinate in the absorbed image information of camera head of sample car.
Such as, it is positioned at vehicle left front on the ground at Calibration Field and lays 3 physics fixed points, calibration and usage is carried out for the camera head at vehicle left side and headstock, described technical staff calculates beforehand through simulation or true car is demarcated the modes such as test and obtained the coordinate in the image information that these 3 physics fixed points camera head respectively at described vehicle left side and headstock is absorbed, and inputs in described calibration system.
Then, described calibration system, according to each sample car fixed point coordinate pre-entered, sets its screen in respective image information detection region.Wherein, described screen detection region can comprise a described sample car fixed point coordinate, it is also possible to comprise multiple described sample car fixed point coordinate.Such as, described calibration system is according to the dense degree of sample car fixed point coordinate corresponding in same image information, along the scope that intensive multiple sample car fixed point coordinates are frame setting screen detection region.Preferably, its frame is slightly larger than each described sample car fixed point coordinate area defined.The frame in described screen detection region is than big 3-6 the pixel of frame of each described sample car fixed point coordinate area defined.
It should be noted that it should be appreciated by those skilled in the art that the range set mode in above-mentioned screen detection region is only for example.In fact technical staff can be adjusted according to vehicle, design requirement etc..
In step s 2, described calibration system obtains the image information that each described camera head is absorbed, and in the corresponding screen detection region in each described image information, select the angle point meeting preset calibrations point characteristic information, and therefrom filter out the fixed point for demarcating.
Specifically, the image information absorbed is passed to described calibration system by each camera head being positioned at vehicle's surroundings, described calibration system detects region according to the screen set by corresponding each camera head, each image information is carried out fixed point choose, i.e., in corresponding screen detection region, select to meet the angle point of fixed point characteristic information, coordinate system further according to respective image information determines the coordinate of selected each angle point, compare with corresponding sample car fixed point quantity further according to obtained angle point quantity, if it is consistent, then assert that selected angle point is fixed point.
Such as, the physics fixed point demarcating place is made up of relative square in angle, two angles, the most described calibration system using square structure relative for angle, angle as fixed point characteristic information, and mating in corresponding screen detection region according to described fixed point characteristic information blanket type.If it is consistent after consistent or bi-directional scaling, it is determined that to be described angle point at the two square diagonal angles that connect, and determine its coordinate in respective image information.Otherwise, then continue coupling, until finding all angle points meeting fixed point characteristic information, if the quantity of the angle point of each camera head of selected correspondence (or corresponding each image information) is identical with the quantity of sample car fixed point, then using obtained angle point as fixed point, and directly perform step S3.
Preferably, the mode selecting the angle point meeting fixed point characteristic information described in includes step S21, S22.(as shown in Figure 2)
In the step s 21, described calibration system transfers the image information portion in described screen detection region to gray level image information part.
In step S22, described calibration system chooses the angle point meeting preset calibrations point characteristic information from described gray level image information.
Wherein, described calibration system only can transfer the image information portion in described screen detection region to gray level image information part.First the image information absorbed can also be converted to gray level image, extract the gray level image information part of correspondence further according to described screen detection region part in respective image information.
Then, described calibration system chooses each described angle point met according to fixed point characteristic information from described gray level image information part.
Such as, after described calibration system transfers the image information portion in described screen detection region to gray level image information part, by described gray level image information part is carried out frequency domain conversion, obtain the characteristic information of described gray level image information part, and from described characteristic information, choose the part meeting preset calibrations point characteristic information, then obtain corresponding each angle point and coordinate thereof by inverse transformation.When the quantity of the fixed point coordinate of the obtained each camera head of correspondence is identical with the quantity of corresponding sample car fixed point coordinate, then can perform step S3.
Preferably, as in figure 2 it is shown, the mode that described calibration system therefrom filters out the fixed point for demarcating includes: step S23.(as shown in Figure 2)
In step S23, described calibration system centered by sample car fixed point coordinate, presetted pixel be that the angle point in the range of radius respectively filters out a fixed point.
Such as, described calibration system have chosen 7 angle points and coordinate from the image information that vehicle left side camera head is absorbed.In described calibration system, the quantity of the sample car fixed point of corresponding vehicle left side camera head position is 5, then with each sample car fixed point coordinate be respectively center, 3-5 pixel as radius in the range of select a fixed point, the angle point not meeting respective range is rejected.So, the fixed point identical with the sample car fixed point quantity of corresponding described vehicle left side camera head and coordinate thereof are obtained.
If through above-mentioned screening, when comprising multiple fixed point in the preset range of certain sample car fixed point coordinate, then choose at a distance of the nearest angle point of described sample car fixed point as fixed point, and reject other angle points.
If the quantity of the angle point after Shai Xuan is less than the quantity of corresponding sample car fixed point, the most described demarcation calibration system also performs step S24 (as shown in Figure 2), will each described angle point of corresponding same camera head be shown, so that technical staff intervenes, and each angle point after reception technique personnel intervention；And, using each angle point after each angle point of being intervened and screening as the fixed point of corresponding described same camera head.
Specifically, when the quantity of the angle point of the same camera head of correspondence after screening according to step S23 is less than corresponding sample car fixed point quantity, each angle point of corresponding same camera head is shown on the relevant position of screen with different colors from unchecked each angle point by described calibration system respectively, then technical staff can supplement the fixed point of volume residual in unchecked each angle point according to practical situation, and submits to described calibration system.Described calibration system obtains the fixed point identical with the quantity of sample car fixed point and coordinate thereof.
In step s3, described calibration system corresponding relation based on each described fixed point coordinate Yu each described physics fixed point coordinate, set up the coordinate system of image information corresponding to each described camera head and the corresponding relation of each described physics fixed point place coordinate system.
Specifically, described calibration system is according to the corresponding relation of each described sample car fixed point coordinate with each physics fixed point coordinate of Calibration Field ground, and the deviation between each described fixed point coordinate and corresponding sample car fixed point coordinate, set up coordinate system and the corresponding relation of the physical coordinates system of vehicle's surroundings in described demarcation place of the image information that the camera head of all directions is absorbed, complete to demarcate.
Or, described calibration system determines the corresponding relation of each fixed point coordinate and physics fixed point coordinate according to the hardware parameter of each camera head, further according to determined by corresponding relation set up the coordinate system of image information corresponding to each described camera head and the corresponding relation of each described physics fixed point place coordinate system.
Referring to Fig. 3, the present invention provides a kind of calibration system looking around senior DAS (Driver Assistant System).Described calibration system 1 belongs to the part in the DAS (Driver Assistant System) installed in the car, comprises multiple camera head in described automobile.Common, described automobile comprises and lays respectively at two rearview mirrors of vehicle, headstock, the camera head of the tailstock.The image information that each camera head is absorbed is processed by the demarcation of described calibration system 1, it is possible to avoid ghost image, blind area, to realize the purpose representing vehicle's surroundings scene of panorama.
Described calibration system 1 includes: first module the 11, second module 12, three module 13.
Described first module 11 is for each sample car fixed point setting coordinate corresponding screen detection region according to each camera head of correspondence preset, and wherein, each described sample car fixed point coordinate is corresponding with Calibration Field each physics fixed point coordinate on the ground.
Here, each camera head is to absorbing Calibration Field multiple physics fixed points on the ground.In order to the fixed point of picked-up and described physics fixed point are carried out correspondence, each sample car fixed point coordinate of corresponding each camera head, for described first module 11 place vehicle, is inputted in described first module 11 by technical staff in advance.The each physics fixed point that each sample car fixed point coordinate inputted is Calibration Field ground screen coordinate in the absorbed image information of camera head of sample car.
Such as, it is positioned at vehicle left front on the ground at Calibration Field and lays 3 physics fixed points, calibration and usage is carried out for the camera head at vehicle left side and headstock, described technical staff calculates beforehand through simulation or true car is demarcated the modes such as test and obtained the coordinate in the image information that these 3 physics fixed points camera head respectively at described vehicle left side and headstock is absorbed, and inputs in described first module 11.
Then, described first module 11, according to each sample car fixed point coordinate pre-entered, sets its screen in respective image information detection region.Wherein, described screen detection region can comprise a described sample car fixed point coordinate, it is also possible to comprise multiple described sample car fixed point coordinate.Such as, described first module 11 is according to the dense degree of sample car fixed point coordinate corresponding in same image information, along the scope that intensive multiple sample car fixed point coordinates are frame setting screen detection region.Preferably, its frame is slightly larger than each described sample car fixed point coordinate area defined.The frame in described screen detection region is than big 3-6 the pixel of frame of each described sample car fixed point coordinate area defined.
It should be noted that it should be appreciated by those skilled in the art that the range set mode in above-mentioned screen detection region is only for example.In fact technical staff can be adjusted according to vehicle, design requirement etc..
Described second module 12 is for obtaining the image information that each described camera head is absorbed, and in the corresponding screen detection region in each described image information, selects the angle point meeting preset calibrations point characteristic information, and therefrom filter out the fixed point for demarcating.
Specifically, the image information absorbed is passed to described second module 12 by each camera head being positioned at vehicle's surroundings, described second module 12 detects region according to the screen set by corresponding each camera head, each image information is carried out fixed point choose, i.e., in corresponding screen detection region, select to meet the angle point of fixed point characteristic information, coordinate system further according to respective image information determines the coordinate of selected each angle point, compare with corresponding sample car fixed point quantity further according to obtained angle point quantity, if it is consistent, then assert that selected angle point is fixed point.
Such as, the physics fixed point demarcating place is made up of relative square in angle, two angles, the most described second module 12 using square structure relative for angle, angle as fixed point characteristic information, and mating in corresponding screen detection region according to described fixed point characteristic information blanket type.If it is consistent after consistent or bi-directional scaling, it is determined that to be described angle point at the two square diagonal angles that connect.Otherwise, then continue coupling, until finding all angle points meeting fixed point characteristic information, if the quantity of the fixed point of each camera head of selected correspondence (or corresponding each image information) is identical with the quantity of sample car fixed point, the most then using obtained angle point as fixed point, and directly perform three module 13.
Preferably, described second module 12 includes: the 2nd 1 submodule the 121, the 2nd 2 submodule 122.(as shown in Figure 4)
Described 2nd 1 submodule 121 is for transferring the image information portion in described screen detection region to gray level image information part.
Described two or two submodule 122 for choosing the angle point meeting preset calibrations point characteristic information from described gray level image information part.
Wherein, described 2nd 1 submodule 121 only can transfer the image information portion in described screen detection region to gray level image information part.First the image information absorbed can also be converted to gray level image, extract the gray level image information part of correspondence further according to described screen detection region part in respective image information.
Then, described two or two submodule 122 chooses, according to mark fixed point characteristic information, each described angle point met from described gray level image information part.
Such as, after described 2nd 1 submodule 121 transfers the image information portion in described screen detection region to gray level image information part, by described two or two submodule 122 by described gray level image information part is carried out frequency domain conversion, obtain the characteristic information of described gray level image information part, and from described characteristic information, choose the part meeting preset calibrations point characteristic information, then obtain corresponding each angle point and coordinate thereof by inverse transformation.
When the quantity of the fixed point coordinate of the obtained each camera head of correspondence is identical with the quantity of corresponding sample car fixed point coordinate, then can perform three module 13.
Preferably, described second module 12 also includes: the two or three submodule 123.(as shown in Figure 4)
Described two or three submodule 123 is used for centered by sample car fixed point coordinate, presetted pixel is respectively to filter out a fixed point the angle point in the range of radius.
Such as, described two or two submodule 122 is extracted 7 angle points and coordinate thereof from the image information that vehicle left side camera head is absorbed.In described two or three submodule 123, the quantity of the sample car fixed point of corresponding vehicle left side camera head position is 5, then with each sample car fixed point coordinate be respectively center, 3-5 pixel as radius in the range of select an angle point, the angle point not meeting respective range is rejected.So, the fixed point identical with the sample car fixed point quantity of corresponding described vehicle left side camera head and coordinate thereof are obtained.
If comprising multiple fixed point in above-mentioned screening, the preset range of certain sample car fixed point coordinate, the most described two or three submodule 123 is chosen at a distance of the nearest angle point of described sample car fixed point as fixed point, and rejects other angle points.
If the quantity of the angle point after Shai Xuan is less than the quantity of corresponding sample car fixed point, the most described second module 12 also includes the two or four submodule 124 (as shown in Figure 4), for each described angle point of corresponding same camera head is shown, so that technical staff intervenes, and each angle point after reception technique personnel intervention；And, using each angle point after each angle point of being intervened and screening as the fixed point of corresponding described same camera head.
Specifically, when the quantity of the angle point according to the same camera head of correspondence after described two or three submodule 123 screening is less than corresponding sample car fixed point quantity, each angle point of corresponding same camera head is shown on the relevant position of screen with different colors from unchecked each angle point by described two or four submodule 124 respectively, then technical staff can supplement the fixed point of volume residual in unchecked each angle point according to practical situation, and submits to described two or four submodule 124.Described two or four submodule 124 obtains the fixed point identical with the quantity of sample car fixed point and coordinate thereof.
Described three module 13 for corresponding relation based on each described fixed point coordinate Yu each described physics fixed point coordinate, sets up coordinate system and the corresponding relation of each described physics fixed point place coordinate system of image information corresponding to each described camera head.
Specifically, described three module 13 is according to the corresponding relation of each described sample car fixed point coordinate with each physics fixed point coordinate of Calibration Field ground, and the deviation between each described vehicle calibration point coordinates and corresponding sample car fixed point coordinate, set up a coordinate system of the image information that direction camera head is absorbed and the corresponding relation of the physical coordinates system of vehicle's surroundings in described demarcation place, complete to demarcate.
Or, described three module 13 determines the corresponding relation of each fixed point coordinate and physics fixed point coordinate according to the hardware parameter of each camera head, further according to determined by corresponding relation set up the coordinate system of image information corresponding to each described camera head and the corresponding relation of each described physics fixed point place coordinate system.
In sum, the scaling method looking around senior DAS (Driver Assistant System) of the present invention and system, by coming setting screen detection region according to the sample car fixed point coordinate preset, it is possible to reduce the scope choosing described vehicle calibration point coordinates greatly, improve the execution efficiency of demarcation；Simultaneously, utilize the corresponding relation of sample car fixed point coordinate and corresponding physics fixed point, it is easy to the deviation obtaining selected described vehicle calibration point coordinates place coordinate system more accurately with physics fixed point place coordinate system, the corresponding relation of Two coordinate system can be effectively improved, provide panorama service accurately for the auxiliary driving such as park of follow-up panorama；It addition, further screened according to the described vehicle calibration point coordinates selected by fixed point characteristic information, it is possible to remove demarcating the interference of the image of similar physical fixed point in place, improve the accuracy of selected described vehicle calibration point coordinates further；Further, when the discrepancy of quantity of selected described vehicle calibration point coordinates with corresponding sample car fixed point coordinate, the most also technical staff can be facilitated disposably to complete calibration experiment by manually adjusting；Additionally, use gray level image information it can be avoided that the ambient impact on calibration process, it is the most fixed to improve further.So, the present invention effectively overcomes various shortcoming of the prior art and has high industrial utilization.
The principle of above-described embodiment only illustrative present invention and effect thereof, not for limiting the present invention.Above-described embodiment all can be modified under the spirit and the scope of the present invention or change by any person skilled in the art.Therefore, art has all equivalence modification or changes that usually intellectual is completed under without departing from disclosed spirit and technological thought such as, must be contained by the claim of the present invention.
Claims (8)
1. look around a scaling method for senior DAS (Driver Assistant System), be applied to be provided with in the vehicle of multiple camera head, it is characterised in that including:
Each sample car fixed point setting coordinate corresponding screen detection region according to the default each camera head of correspondence, wherein, each described sample car fixed point coordinate is corresponding with Calibration Field each physics fixed point coordinate on the ground；
Obtain in the image information that each described camera head is absorbed, and the corresponding screen detection region in each described image information, select the angle point meeting preset calibrations point characteristic information, and therefrom filter out the fixed point for demarcating；
Corresponding relation based on each described fixed point coordinate Yu each described physics fixed point coordinate, sets up the coordinate system of image information corresponding to each described camera head and the corresponding relation of each described physics fixed point place coordinate system.
The scaling method looking around senior DAS (Driver Assistant System) the most according to claim 1, it is characterised in that the mode selecting the angle point meeting fixed point characteristic information described in: includes:
Transfer the image information portion in described screen detection region to gray level image information part.
3. from described gray level image information, choose the angle point meeting preset calibrations point characteristic information.The scaling method looking around senior DAS (Driver Assistant System) according to claim 1, it is characterised in that: the mode of the described fixed point therefrom filtered out for demarcating includes:
Centered by sample car fixed point coordinate, presetted pixel be that the angle point in the range of radius respectively filters out a fixed point.
The scaling method looking around senior DAS (Driver Assistant System) the most according to claim 3, it is characterized in that: if the quantity of the angle point after Shai Xuan is less than the quantity of corresponding sample car fixed point, the mode of the described fixed point therefrom filtered out for demarcating also includes: shown by each described angle point of corresponding same camera head, so that technical staff intervenes, and each described angle point after reception technique personnel intervention；And, using each angle point after each angle point of being intervened and screening as the fixed point of corresponding described same camera head.
5. look around a calibration system for senior DAS (Driver Assistant System), be applied to be provided with in the vehicle of multiple camera head, it is characterised in that including:
First module, for each sample car fixed point setting coordinate corresponding screen detection region according to each camera head of correspondence preset, wherein, each described sample car fixed point coordinate is corresponding with Calibration Field each physics fixed point coordinate on the ground；
Second module, for obtaining the image information that each described camera head is absorbed, and in the corresponding screen detection region in each described image information, selects the angle point meeting preset calibrations point characteristic information, and therefrom filters out the fixed point for demarcating；
Three module, for corresponding relation based on each described fixed point coordinate Yu each described physics fixed point coordinate, sets up coordinate system and the corresponding relation of each described physics fixed point place coordinate system of image information corresponding to each described camera head.
The calibration system looking around senior DAS (Driver Assistant System) the most according to claim 5, it is characterised in that: described second module includes:
2nd 1 submodule, for transferring the image information portion in described screen detection region to gray level image information part；
Two or two submodule, for choosing the angle point meeting preset calibrations point characteristic information from described gray level image information.
The calibration system looking around senior DAS (Driver Assistant System) the most according to claim 5, it is characterised in that: described second module includes:
Two or three submodule, for centered by sample car fixed point coordinate, presetted pixel be that the angle point in the range of radius respectively filters out a fixed point.
The calibration system looking around senior DAS (Driver Assistant System) the most according to claim 7, it is characterized in that: if the quantity of the angle point after Shai Xuan is less than the quantity of corresponding sample car fixed point, described second module also includes: the two or four submodule, for each described angle point of corresponding same camera head is shown, so that technical staff intervenes, and each described angle point after reception technique personnel intervention, and, using each angle point after each angle point of being intervened and screening as the fixed point of corresponding described same camera head.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510007598.1A CN105818746A (en) | 2015-01-05 | 2015-01-05 | Calibration method and system of panoramic advanced driver assistance system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510007598.1A CN105818746A (en) | 2015-01-05 | 2015-01-05 | Calibration method and system of panoramic advanced driver assistance system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105818746A true CN105818746A (en) | 2016-08-03 |
Family
ID=56513383
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510007598.1A Pending CN105818746A (en) | 2015-01-05 | 2015-01-05 | Calibration method and system of panoramic advanced driver assistance system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105818746A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106447736A (en) * | 2016-11-23 | 2017-02-22 | 安徽慧视金瞳科技有限公司 | Multi-calibration-point detection method based on line-by-line division |
| CN106530354A (en) * | 2016-10-31 | 2017-03-22 | 纵目科技（上海）股份有限公司 | Image processing device, image correction method, system, calibration point search method and system |
| CN106600642A (en) * | 2016-10-31 | 2017-04-26 | 纵目科技（上海）股份有限公司 | Image processing device, image correction method and system, and fixed point searching method and system |
| CN106780616A (en) * | 2016-11-23 | 2017-05-31 | 安徽慧视金瞳科技有限公司 | A kind of projector calibrating method based on the mapping of many matrixes |
| CN108776470A (en) * | 2018-09-06 | 2018-11-09 | 中国科学院沈阳自动化研究所 | A kind of driving auxiliary calibration detecting system |
| CN111605481A (en) * | 2019-08-27 | 2020-09-01 | 北京纵目安驰智能科技有限公司 | Congestion car following system and terminal based on look around |
| CN114202588A (en) * | 2021-12-09 | 2022-03-18 | 纵目科技（上海）股份有限公司 | Method and device for quickly and automatically calibrating vehicle-mounted panoramic camera |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101216941A (en) * | 2008-01-17 | 2008-07-09 | 上海交通大学 | Motion estimation method under violent illumination variation based on corner matching and optic flow method |
| CN102074015A (en) * | 2011-02-24 | 2011-05-25 | 哈尔滨工业大学 | Two-dimensional image sequence based three-dimensional reconstruction method of target |
| CN103177439A (en) * | 2012-11-26 | 2013-06-26 | 惠州华阳通用电子有限公司 | Automatically calibration method based on black and white grid corner matching |
| CN103646403A (en) * | 2013-12-26 | 2014-03-19 | 北京经纬恒润科技有限公司 | Vehicle-mounted multi-camera calibration method, system and image processing device |
| CN103871070A (en) * | 2014-04-03 | 2014-06-18 | 深圳市德赛微电子技术有限公司 | Automatic calibration method of vehicle-mounted panoramic imaging system |
| CN103955916A (en) * | 2014-03-21 | 2014-07-30 | 贺徵廷 | Automobile panorama camera calibration method and apparatus |
| KR101427181B1 (en) * | 2013-01-09 | 2014-08-07 | 아진산업(주) | Calibration indicator used for calibration of onboard camera using variable ellipse pattern and calibration method of onboard camera using calibration indicator |
| CN103985118A (en) * | 2014-04-28 | 2014-08-13 | 无锡观智视觉科技有限公司 | Parameter calibration method for cameras of vehicle-mounted all-round view system |
| CN103996191A (en) * | 2014-05-09 | 2014-08-20 | 东北大学 | Detection method for black and white checkerboard image corners based on least square optimization |
| CN104008548A (en) * | 2014-06-04 | 2014-08-27 | 无锡观智视觉科技有限公司 | Feature point extraction method for vehicle-mounted around view system camera parameter calibration |
| CN104240225A (en) * | 2013-06-19 | 2014-12-24 | 德尔福电子(苏州)有限公司 | BEV camera one-step calibration method |
-
2015
- 2015-01-05 CN CN201510007598.1A patent/CN105818746A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101216941A (en) * | 2008-01-17 | 2008-07-09 | 上海交通大学 | Motion estimation method under violent illumination variation based on corner matching and optic flow method |
| CN102074015A (en) * | 2011-02-24 | 2011-05-25 | 哈尔滨工业大学 | Two-dimensional image sequence based three-dimensional reconstruction method of target |
| CN103177439A (en) * | 2012-11-26 | 2013-06-26 | 惠州华阳通用电子有限公司 | Automatically calibration method based on black and white grid corner matching |
| KR101427181B1 (en) * | 2013-01-09 | 2014-08-07 | 아진산업(주) | Calibration indicator used for calibration of onboard camera using variable ellipse pattern and calibration method of onboard camera using calibration indicator |
| CN104240225A (en) * | 2013-06-19 | 2014-12-24 | 德尔福电子(苏州)有限公司 | BEV camera one-step calibration method |
| CN103646403A (en) * | 2013-12-26 | 2014-03-19 | 北京经纬恒润科技有限公司 | Vehicle-mounted multi-camera calibration method, system and image processing device |
| CN103955916A (en) * | 2014-03-21 | 2014-07-30 | 贺徵廷 | Automobile panorama camera calibration method and apparatus |
| CN103871070A (en) * | 2014-04-03 | 2014-06-18 | 深圳市德赛微电子技术有限公司 | Automatic calibration method of vehicle-mounted panoramic imaging system |
| CN103985118A (en) * | 2014-04-28 | 2014-08-13 | 无锡观智视觉科技有限公司 | Parameter calibration method for cameras of vehicle-mounted all-round view system |
| CN103996191A (en) * | 2014-05-09 | 2014-08-20 | 东北大学 | Detection method for black and white checkerboard image corners based on least square optimization |
| CN104008548A (en) * | 2014-06-04 | 2014-08-27 | 无锡观智视觉科技有限公司 | Feature point extraction method for vehicle-mounted around view system camera parameter calibration |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106530354A (en) * | 2016-10-31 | 2017-03-22 | 纵目科技（上海）股份有限公司 | Image processing device, image correction method, system, calibration point search method and system |
| CN106600642A (en) * | 2016-10-31 | 2017-04-26 | 纵目科技（上海）股份有限公司 | Image processing device, image correction method and system, and fixed point searching method and system |
| CN106530354B (en) * | 2016-10-31 | 2020-01-14 | 纵目科技（上海）股份有限公司 | Image processing device, image correction method, and calibration point searching method and system |
| CN106447736A (en) * | 2016-11-23 | 2017-02-22 | 安徽慧视金瞳科技有限公司 | Multi-calibration-point detection method based on line-by-line division |
| CN106780616A (en) * | 2016-11-23 | 2017-05-31 | 安徽慧视金瞳科技有限公司 | A kind of projector calibrating method based on the mapping of many matrixes |
| CN106447736B (en) * | 2016-11-23 | 2019-05-31 | 安徽慧视金瞳科技有限公司 | A kind of more calibration point detecting methods divided on line by line basis |
| CN108776470A (en) * | 2018-09-06 | 2018-11-09 | 中国科学院沈阳自动化研究所 | A kind of driving auxiliary calibration detecting system |
| CN108776470B (en) * | 2018-09-06 | 2024-04-02 | 中国科学院沈阳自动化研究所 | Driving auxiliary calibration detection system |
| CN111605481A (en) * | 2019-08-27 | 2020-09-01 | 北京纵目安驰智能科技有限公司 | Congestion car following system and terminal based on look around |
| CN114202588A (en) * | 2021-12-09 | 2022-03-18 | 纵目科技（上海）股份有限公司 | Method and device for quickly and automatically calibrating vehicle-mounted panoramic camera |
| CN114202588B (en) * | 2021-12-09 | 2022-09-23 | 纵目科技（上海）股份有限公司 | Method and device for quickly and automatically calibrating vehicle-mounted panoramic camera |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105818746A (en) | Calibration method and system of panoramic advanced driver assistance system | |
| CN110381255B (en) | Vehicle-mounted video monitoring system and method applying 360-degree panoramic looking-around technology | |
| JP5739584B2 (en) | 3D image synthesizing apparatus and method for visualizing vehicle periphery | |
| EP2437494B1 (en) | Device for monitoring area around vehicle | |
| CN112965503A (en) | Multi-path camera fusion splicing method, device, equipment and storage medium | |
| CN104512328B (en) | Automobile looks around image generating method and automobile viewing system | |
| CN102045546B (en) | Panoramic parking assist system | |
| DE102013220669A1 (en) | Dynamic rearview indicator features | |
| JP4892965B2 (en) | Moving object determination system, moving object determination method, and computer program | |
| US20160005316A1 (en) | Around view system and operating method thereof | |
| DE112017004203T5 (en) | IMAGE PROCESSING DEVICE, INFORMATION GENERATION DEVICE AND INFORMATION GENERATION PROCESS | |
| CN105321160B (en) | The multi-camera calibration that 3 D stereo panorama is parked | |
| KR102057021B1 (en) | Panel transformation | |
| CN106994936A (en) | A kind of 3D panoramic parking assist systems | |
| CN110796711B (en) | Panoramic system calibration method and device, computer readable storage medium and vehicle | |
| KR102124298B1 (en) | Rear Cross Traffic-Quick Look | |
| CN107364393A (en) | Display methods, device, storage medium and the electronic equipment of vehicle rear view image | |
| JP5951785B2 (en) | Image processing apparatus and vehicle forward monitoring apparatus | |
| DE102018133030A1 (en) | VEHICLE REMOTE CONTROL DEVICE AND VEHICLE REMOTE CONTROL METHOD | |
| JP6337504B2 (en) | Image processing apparatus, moving body, robot, device control method and program | |
| CN105323447B (en) | More fish eye images processing methods, device and vehicle | |
| KR102235951B1 (en) | Imaging Apparatus and method for Automobile | |
| KR20160034681A (en) | Environment monitoring apparatus and method for vehicle | |
| WO2017158829A1 (en) | Display control device and display control method | |
| JP5305750B2 (en) | Vehicle periphery display device and display method thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160803 |