CN105635597B - The automatic explosion method and system of in-vehicle camera - Google Patents
The automatic explosion method and system of in-vehicle camera Download PDFInfo
- Publication number
- CN105635597B CN105635597B CN201510968340.8A CN201510968340A CN105635597B CN 105635597 B CN105635597 B CN 105635597B CN 201510968340 A CN201510968340 A CN 201510968340A CN 105635597 B CN105635597 B CN 105635597B
- Authority
- CN
- China
- Prior art keywords
- exposure
- frame
- time
- value
- next frame
- 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.)
- Expired - Fee Related
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/70—Circuitry for compensating brightness variation in the scene
- H04N23/73—Circuitry for compensating brightness variation in the scene by influencing the exposure time
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/70—Circuitry for compensating brightness variation in the scene
- H04N23/75—Circuitry for compensating brightness variation in the scene by influencing optical camera components
Abstract
A kind of automatic explosion method and system of in-vehicle camera, are arranged area-of-interest after initialization, are all handled just for area-of-interest every frame image；Start camera, the time for exposure of first frame uses initial exposure time, it carries out that the second frame is arranged by the quick adjusting of standard of gray threshold, to follow-up each frame, present frame and previous frame image gray average difference are not above gray scale shake saltus step range and then retain Time Exposure value using present frame, else if detecting that traffic sign then carries out the exposure based on image entropy and adjusts, else if the gradation of image mean value of present frame then directly uses the value of present frame in standard grayscale threshold range, otherwise gray average difference carries out being adjusted according to the PID of gray scale and exposure value ratio if within the scope of saltus step, otherwise carry out carrying out the Time Exposure discreet value of the exposure method acquisition next frame of fast zoom table using cmos sensor characteristic；Finally discreet value is controlled according to time for exposure adjustable range.
Description
Technical field
The present invention relates to technical field of intelligent traffic, provide a kind of automatic explosion method and system of in-vehicle camera.
Background technology
With the improvement of living standards with people's the accelerating rhythm of life, vehicle is more and more universal, the pressure of communications and transportation
Also the problem of therefore unprecedented increase, the active safety of automobile, is increasingly becoming focus of people's attention.
The it is proposed of unmanned vehicle is namely based on this demand and comes, and unmanned vehicle collection automatically controls, architecture, artificial intelligence
Can, vision numerous technologies such as calculate in one, be computer science, pattern-recognition and intelligent control technology high development production
Object, and an important symbol of national a research strength and industrial level is weighed, have in national defence and national economy field
Wide application prospect.
Pilotless automobile is to perceive road environment by vehicle-mounted sensor-based system, and automatic planning travelling line simultaneously controls vehicle
Reach the intelligent automobile of predeterminated target.Main sensing equipment of the vehicle-mounted digital camera as road vehicle context aware systems, by
Extensive attention is arrived.
And automatic exposure technology is the magazine focus technology of vehicle-mounted digital, is different from traditional automatic explosion method
It is using human eye vision as judgment criteria, the automatic exposure under vehicle environment is to enable the machine to automatic identification object, is such as known
The objects such as other traffic sign, traffic lights, lane line are only the purpose of automatic exposure.
The automatic explosion method of digital camera generally has two kinds of methods of hardware realization and software realization.Occur in digital camera
The early stage method that mostly uses hardware controls, it has many advantages, such as to control, and speed is fast, effect is good, but that there are hardware devices is of high cost
High disadvantage.With the innovation of production technology and the maturation of software exposure algorithm, can be obtained together using the method that software controls
The control effect of sample, and it is of low cost efficiently control exposure mainly have 3 kinds of methods：1 adjustment aperture, 2 control electronic shutters,
3 adjustment gain amplifiers.In the environment sensing of automatic driving car, latter two method application is more extensive.
Currently, there are mainly three types of the software approach of the auto-exposure control of digital camera：
1 automatic explosion method based on brightness of image mean value.
2 automatic explosion methods based on image brightness histogram.
The automatic exposure method of 3 image content-baseds.
This 3 kinds of methods respectively have its advantage and disadvantage, but due to the particular surroundings of vehicle environment, the high dynamic of external environment mutability
Under environment, such as there is situations such as backlight, road environment is complicated, surface gathered water cannot all meet program requirement very well.Based on these
Reason, the automatic explosion method of vehicular platform require have very high adaptability, high efficiency, hard real-time and stability.
Invention content
Different from the exposure method of camera and other monitoring cameras, the purpose of the present invention is put down from based on vehicle-mounted this
Platform sets out, and the main environment considered residing for automobile, vehicle condition is complicated and changeable, and accurately identification target is required in different locations
Object makes the image data of acquisition that can preferably identify object such as lane line, traffic sign, traffic lights etc.；And it is entirely
The complexity of system, the memory of distribution and processing time cannot be excessive；During high speed driving, how target is quickly identified
Object, exposure program are the foundation stone of whole image identification again, how to ensure that whole stability is the most important thing.
Technical solution of the present invention provides a kind of automatic explosion method of in-vehicle camera, includes the following steps：
Step 1, it initializes, including configuration initial exposure time, time for exposure adjustable range, standard grayscale threshold range
Gt, the saltus step range Gp and Gray Level Jump stabilization range Gc for meeting PID adjustings；
Step 2, area-of-interest is set, every frame image will all be handled just for area-of-interest；
Step 3, start camera, the time for exposure of first frame uses the initial exposure time that step 1 configures, in shooting first
Obtain first frame gradation of image mean value for area-of-interest after frame, according to the standard grayscale threshold range Gt of setting, carry out with
Gray threshold is the quick adjusting of standard, according to the time for exposure for adjusting result the second frame of setting, enters step 9；
Step 4, the gradation of image mean value of present frame is compared with the gradation of image mean value of former frame, if two frame gray scales
The gray scale shake saltus step range Gc that mean value difference DELTA G variation is not above step 1 setting then retains time exposure using present frame
Time Exposure discreet value of the light value as next frame, enters step 9, otherwise enters step 5；
Step 5, the exposure based on image entropy is carried out if detecting traffic sign from current frame image to adjust, under acquisition
The Time Exposure discreet value of one frame, enters step 9, otherwise enters step 6；
Step 6, if the gradation of image mean value of present frame is in step 1 preset standard gray threshold range Gt, directly
Time Exposure discreet value using the Time Exposure value of present frame as next frame, enters step 9, otherwise enters step 7；
Step 7, according to the difference DELTA of the gradation of image mean value of step 4 gained present frame and the gradation of image mean value of former frame
G further judges, if in the preset saltus step range Gp of step 1, carries out the PID tune according to gray scale and exposure value ratio
Section, obtains the Time Exposure discreet value of next frame, enters step 9, otherwise carries out step 8；
Step 8, the exposure method for carrying out being carried out fast zoom table using cmos sensor characteristic, the time for obtaining next frame are exposed
Light discreet value, enters step 9；
Step 9, to the Time Exposure discreet value of current gained next frame according to the time for exposure adjustable range in step 1 into
Row control is maximized if the Time Exposure discreet value of next frame is more than the maximum value of time for exposure adjustable range as most
The Time Exposure value of determining next frame eventually, if it is less than the minimum value of time for exposure adjustable range, then it is final to be minimized
The Time Exposure value of determining next frame, the Time Exposure of directly setting next frame is pre- if in time for exposure adjustable range
Valuation is the Time Exposure value of the next frame of final determination；It shoots to obtain according to finally determining Time Exposure value control camera
After next frame, return to step 4 is recycled next frame as new present frame.
Moreover, when area-of-interest is set in step 2, the useless automobile cabinet cover area in lower section in removal camera lens,
Top sky areas and both sides of the edge.
The present invention provides a kind of auto exposure system of in-vehicle camera, which is characterized in that comprises the following modules：
Initialization module, for configure initial exposure time, time for exposure adjustable range, standard grayscale threshold range Gt,
Meet the saltus step range Gp and Gray Level Jump stabilization range Gc of PID adjustings；
Region division module, for area-of-interest to be arranged, to every frame image all will be just for area-of-interest at
Reason；
Initial start module, for starting camera, the time for exposure of first frame uses the initial exposure of initialization module configuration
Between light time, first frame gradation of image mean value is obtained for area-of-interest after shooting first frame, according to the standard grayscale of setting
Threshold range Gt is carried out using gray threshold as the quick adjusting of standard, according to the time for exposure for adjusting result the second frame of setting, life
Control module is enabled to work；
Stabilization adjustment module, for comparing the gradation of image mean value of the gradation of image mean value of present frame and former frame
Compared with retaining if the gray scale shake saltus step range Gc that two frame gray average difference DELTA G variations are not above initialization module setting
Time Exposure discreet value using the Time Exposure value of present frame as next frame, command control module work, otherwise command diagram
As entropy adjustment module works；
Image entropy adjustment module, for carrying out the exposure based on image entropy if detecting traffic sign from current frame image
Light is adjusted, and obtains the Time Exposure discreet value of next frame, command control module work, otherwise command standard judgment module work；
Standard judgment module, if the gradation of image mean value for present frame is in initialization module preset standard gray threshold
In range Gt, then Time Exposure discreet value of the Time Exposure value of present frame as next frame, command control module are directly used
Work, otherwise order PID control adjustment module work；
PID control adjustment module, for according to the gradation of image mean value of present frame obtained by stabilization adjustment module with it is previous
The difference DELTA G of the gradation of image mean value of frame further judges, if in the preset saltus step range Gp of initialization module, carries out
It is adjusted according to the PID of gray scale and exposure value ratio, obtains the Time Exposure discreet value of next frame, command control module work is no
Then order cmos sensor adjustment module works；
Cmos sensor adjustment module, the exposure method for carried out using cmos sensor characteristic fast zoom table,
Obtain the Time Exposure discreet value of next frame, command control module work；
Control module, when for the Time Exposure discreet value to current gained next frame according to exposure in initialization module
Between adjustable range controlled, if the Time Exposure discreet value of next frame be more than time for exposure adjustable range maximum value,
The Time Exposure value being maximized as finally determining next frame then takes if it is less than the minimum value of time for exposure adjustable range
Minimum value is the Time Exposure value of finally determining next frame, and next frame is directly arranged if in time for exposure adjustable range
Time Exposure discreet value be final determination next frame Time Exposure value；According to finally determining Time Exposure value control
Camera is shot after obtaining next frame, and the work of return command stabilization adjustment module is followed next frame as new present frame
Ring.
Moreover, when area-of-interest is set in region division module, the useless automobile cabinet in lower section in removal camera lens
Cover area, top sky areas and both sides of the edge.
The various situations that composite demand of the present invention considers took slowly to solve full figure scanning, and subregion can lead to
It crosses division region to remove to save sweep time, combines road traffic sign detection module to carry out subregion exposure using region is divided
Light, rapidity are one of vehicular platform requirements, so using the photobehavior in conjunction with cmos sensor when initial adjust
It is quickly adjusted, while taking a gray threshold range to carry out auxiliary adjustment, and add grey scale change range, prevented
The flating that only large-scale saltus step is brought.
Also, the exposure method being adjusted based on gray threshold, governing speed is fast, but accuracy is not high, at certain
Exposure numerical value is less desirable in special circumstances a bit, but due to the rapidity requirement of vehicular platform, so adopting this method
It is adjusted, is adjusted using image entropy module in special circumstances on a large scale.The case where for backlight or frontlighting always vehicle
Where the difficult point for carrying automatic exposure, the evaluation for introducing image entropy is adjusted to more accurate to object progress under conditions of backlight
Exposure is adjusted.
Therefore, the invention has the advantages that：
1, there is very strong real-time, performance can reach 20 frames/second for exposure.
2, in-vehicle camera can be identified soon under the change over condition along backlight and rapidly be reacted exposure value.
3, in the case of some are special, such as by bridge opening before and after, surface gathered water seriously also shows in equal scene
Very strong adaptability.
4, due to adding the element of image entropy when identifying traffic mark board, to make discrimination rise.
5, whole flow process is all simplified very much in time complexity and space complexity, reduce occupy total system memory and
The time of consumption.
Description of the drawings
Fig. 1 is the flow chart of the embodiment of the present invention.
Specific implementation mode
Technical scheme of the present invention is described in detail below in conjunction with drawings and examples.
Referring to Fig. 1, the automatic explosion method workflow of the in-vehicle camera of the embodiment of the present invention is as follows：
Step 1, it initializes：When it is implemented, configuration file, including initial exposure time can be inputted, time for exposure tune
Adjusting range, standard grayscale threshold range Gt meet the saltus step range Gp, Gray Level Jump stabilization range Gc of PID adjustings.It is described
PID is that proportional integral differential is adjusted.Those skilled in the art can voluntarily preset initial exposure time and the time for exposure adjusts model
It encloses, the bound value of range Gt/Gp/Gc.When it is implemented, those skilled in the art can also voluntarily carry out other aspects
Initial setting, such as the parameters of the initial setting of gain and camera setting, generally comprise the closing of huge frame, IP's
Configuration, the setting of frame number, white balance modulation, hardware setting are fixed aperture number.
Step 2, region is divided：ROI (area-of-interest) is set.
The concrete function of in-vehicle camera is not whole choosing location, and in-vehicle camera only needs to identify set target,
And these targets are generally all distributed in each region of image, so the strategy of regional exposure can be used, that not only reduces takes,
Also ensure the Accuracy and high efficiency that exposure is adjusted.When carrying out the selection of area-of-interest, those skilled in the art can be voluntarily
Default selection principle, it is proposed that the useless automobile cabinet cover area in lower section, top sky areas and two sides in removal camera lens
Edge is finally gone into an inverted trapezoidal region.To estimate exposure value, the frame image subsequently obtained is all just for this area-of-interest
It is handled.When it is implemented, area coverage function can be pre-set, area-of-interest is extracted for process invocation；Or by
User voluntarily chooses area-of-interest manually.
Step 3, start camera, time for exposure of first frame is the initial exposure time configured using step 1, in shooting the
First frame gradation of image mean value (just for region of interesting extraction gray average, other follow-up frames are same), root are obtained after one frame
According to the standard grayscale threshold range Gt of setting, using gray threshold as the quick adjusting of standard, according to result is adjusted be arranged
To the time for exposure of the second frame, 9 are entered step.
When it is implemented, can be found in by the quick adjusting of standard specific implementation of gray threshold：Zhao Fang, Luan Xiaoming, Sun Yue
Deng.Exposure amount estimation method [J] based on gray level image.Applicating technology, 2009.The realization of this step also belongs to initial beam worker
Make, related content is omitted in Fig. 1.
Step 4, stabilization is adjusted：The gradation of image mean value that the gradation of image mean value of present frame and former frame retain is carried out
Compare, the gray scale shake saltus step range Gc that two frame gray average difference DELTA G variations are not above step 1 setting then retains using working as
Time Exposure discreet value of the Time Exposure value of previous frame as next frame, skips to step 9.Otherwise 5 are entered step, is proceeded by point
Block is adjusted.
Since the time for exposure of first frame is to configure the initial exposure time directly set, the second frame is fast according to step 3
Velocity modulation section and step 9 control acquired results, so since third frame, gradation of image mean value and the former frame reservation of present frame
Gradation of image mean value is compared, for example, when executing step 4 for the first time, frame is the second frame obtained by current shooting, to determine exposure
The next frame of time is third frame, and target is to estimate the exposure time values of third frame, then the gradation of image of current second frame is equal
Value is compared with the gradation of image mean value that first frame retains.It is follow-up return to step 4 mode it is identical.
Different from the camera shooting of usually household camera, in vehicular platform, since road environment is changeable, external circumstances are multiple
It is miscellaneous, so the numerical value that exposure is adjusted not is a fixed value, can constantly change in a region instead, and this variation
Sometimes saltus step can be generated as cam lens are blocked by tree shade, image may produce because of the saltus step of a very short time again
Raw flickering variation, the in order to prevent generation of such case add the adjusting limitation for preventing shake, to ensure image
The stability of acquisition.
Step 5, image entropy is adjusted：The exposure based on image entropy is carried out if detecting traffic sign from current frame image
It adjusts, obtains the Time Exposure discreet value of next frame, skip to step 9.6 are entered step if not detecting.
Related art can be used in the detections such as lane line, traffic sign, traffic lights when specific implementation, and the present invention pertains only to
Spectrum assignment is realized.Exposure based on image entropy, which is adjusted, can be found in：Yang Zuoyan, Ruan Ping, Zhai Bo etc..High dynamic based on image entropy
The automatic exposure algorithm [J] of range scene.Photon journal, 2013.
Step 6, if the gradation of image mean value of present frame in step 1 preset standard gray threshold range Gt (i.e.) then directly Time Exposure discreet value of the reservation using the Time Exposure value of present frame as next frame, skip to step
9.7 are entered step if exceeding.
Step 7, PID control is adjusted：According to the gradation of image of the gradation of image mean value and former frame of step 4 gained present frame
Mean value difference DELTA G further judges, if difference is smaller to meet preset saltus step range Gp (i.e.), i.e., also can be with
In the range of accuracy controlling, then the PID adjusting methods according to gray scale and exposure value ratio are carried out, the Time Exposure of next frame is obtained
Discreet value skips to step 9.If excessive more than accurate adjustable range, then step 8 is carried out.
It can be found in when it is implemented, adjusting specific implementation according to the PID of gray scale and exposure value ratio：Chen Shanwei, Hou Dexin,
Fast automatic exposure method [J] computer engineering in the lighter flame image analyses such as Ye Shuliang, 2011,37 (23):
200-202。
Step 8, cmos sensor is based on to adjust：Because present frame and former frame comparison saltus step are excessive, then CMOS is utilized to pass
Sensor characteristic carries out the exposure method of fast zoom table, achievees the effect that quick lock in exposure range, and the time for obtaining next frame exposes
Light discreet value, enters step 9.
When it is implemented, the exposure method specific implementation for being carried out fast zoom table using cmos sensor characteristic can be found in：
Vuong Q K,Yun S H,Kim S.A new auto exposure system to detect high dynamic
range conditions using CMOS technology[C].2008 Third International Conference
on Convergence and Hybrid Information Technology,Busan 2008:577-580.
Step 9, threshold exposure scope control：To the Time Exposure discreet value of current gained next frame according in step 1
Time for exposure adjustable range is controlled, if the Time Exposure discreet value of next frame is more than the maximum of time for exposure adjustable range
Value, then be maximized as the Time Exposure value of finally determining next frame, if it is less than the minimum value of time for exposure adjustable range,
It is then minimized the Time Exposure value of the next frame finally to determine, under being directly arranged if in time for exposure adjustable range
The Time Exposure discreet value of one frame is the Time Exposure value of the next frame of final determination, and control camera shoots to obtain next accordingly
After frame, return to step 4 starts to recycle next frame as new present frame, can carry out requiring to stop until user automatically.
To prove effect of the present invention, each method can be tested, judge effect according to time-consuming and frame per second, test table
It is as follows：
Each method test judges that effect, test table are as follows according to time-consuming and frame per second：
It can be seen that method provided by the present invention improves the significant effect of discrimination, and takes and will not extend.
When it is implemented, software technology, which may be used, in method provided by the present invention realizes automatic running, mould can also be used
Block mode provides corresponding system.The embodiment of the present invention also provides a kind of auto exposure system of in-vehicle camera, including with lower die
Block：
Initialization module, for configure initial exposure time, time for exposure adjustable range, standard grayscale threshold range Gt,
Meet the saltus step range Gp and Gray Level Jump stabilization range Gc of PID adjustings；
Region division module, for area-of-interest to be arranged, to every frame image all will be just for area-of-interest at
Reason；
Initial start module, for starting camera, the time for exposure of first frame uses the initial exposure of initialization module configuration
Between light time, first frame gradation of image mean value is obtained for area-of-interest after shooting first frame, according to the standard grayscale of setting
Threshold range Gt is carried out using gray threshold as the quick adjusting of standard, according to the time for exposure for adjusting result the second frame of setting, life
Control module is enabled to work；
Stabilization adjustment module, for comparing the gradation of image mean value of the gradation of image mean value of present frame and former frame
Compared with retaining if the gray scale shake saltus step range Gc that two frame gray average difference DELTA G variations are not above initialization module setting
Time Exposure discreet value using the Time Exposure value of present frame as next frame, command control module work, otherwise command diagram
As entropy adjustment module works；
Image entropy adjustment module, for carrying out the exposure based on image entropy if detecting traffic sign from current frame image
Light is adjusted, and obtains the Time Exposure discreet value of next frame, command control module work, otherwise command standard judgment module work；
Standard judgment module, if the gradation of image mean value for present frame is in initialization module preset standard gray threshold
In range Gt, then Time Exposure discreet value of the Time Exposure value of present frame as next frame, command control module are directly used
Work, otherwise order PID control adjustment module work；
PID control adjustment module, for according to the gradation of image mean value of present frame obtained by stabilization adjustment module with it is previous
The difference DELTA G of the gradation of image mean value of frame further judges, if in the preset saltus step range Gp of initialization module, carries out
It is adjusted according to the PID of gray scale and exposure value ratio, obtains the Time Exposure discreet value of next frame, command control module work is no
Then order cmos sensor adjustment module works；
Cmos sensor adjustment module, the exposure method for carried out using cmos sensor characteristic fast zoom table,
Obtain the Time Exposure discreet value of next frame, command control module work；
Control module, when for the Time Exposure discreet value to current gained next frame according to exposure in initialization module
Between adjustable range controlled, if the Time Exposure discreet value of next frame be more than time for exposure adjustable range maximum value,
The Time Exposure value being maximized as finally determining next frame then takes if it is less than the minimum value of time for exposure adjustable range
Minimum value is the Time Exposure value of finally determining next frame, and next frame is directly arranged if in time for exposure adjustable range
Time Exposure discreet value be final determination next frame Time Exposure value；According to finally determining Time Exposure value control
Camera is shot after obtaining next frame, and the work of return command stabilization adjustment module is followed next frame as new present frame
Ring.
Further, when area-of-interest is set in region division module, the useless automobile in lower section in removal camera lens
Cabinet cover area, top sky areas and both sides of the edge.
Above example is used for illustrative purposes only rather than limitation of the present invention, the technology people in relation to technical field
Member, without departing from the spirit and scope of the present invention, can also make various transformation or modification, therefore all equivalent
Technical solution both falls within protection scope of the present invention.
Claims (4)
1. a kind of automatic explosion method of in-vehicle camera, which is characterized in that include the following steps：
Step 1, it initializes, including configuration initial exposure time, time for exposure adjustable range, standard grayscale threshold range Gt, symbol
Close the saltus step range Gp and Gray Level Jump stabilization range Gc that PID is adjusted；
Step 2, area-of-interest is set, every frame image will all be handled just for area-of-interest；
Step 3, start camera, the time for exposure of first frame uses the initial exposure time that step 1 configures, after shooting first frame
First frame gradation of image mean value is obtained for area-of-interest, according to the standard grayscale threshold range Gt of setting, is carried out with gray scale
Threshold value is the quick adjusting of standard, according to the time for exposure for adjusting result the second frame of setting, enters step 9；
Step 4, the gradation of image mean value of present frame is compared with the gradation of image mean value of former frame, if two frame gray averages
The Gray Level Jump stabilization range Gc that difference DELTA G variation is not above step 1 setting then retains Time Exposure using present frame
It is worth the Time Exposure discreet value as next frame, enters step 9, otherwise enter step 5；
Step 5, the exposure based on image entropy is carried out if detecting traffic sign from current frame image to adjust, and obtains next frame
Time Exposure discreet value, enter step 9, otherwise enter step 6；
Step 6, it if the gradation of image mean value of present frame is in step 1 preset standard gray threshold range Gt, directly uses
Time Exposure discreet value of the Time Exposure value of present frame as next frame, enters step 9, otherwise enters step 7；
Step 7, according to the difference DELTA G of the gradation of image mean value of step 4 gained present frame and the gradation of image mean value of former frame into
One step judges, if in the preset saltus step range Gp of step 1, adjusted according to the PID of gray scale and exposure value ratio, obtains
The Time Exposure discreet value for taking next frame, enters step 9, otherwise carries out step 8；
Step 8, carry out carrying out the exposure method of fast zoom table using cmos sensor characteristic, the Time Exposure for obtaining next frame is pre-
Valuation enters step 9；
Step 9, the Time Exposure discreet value of current gained next frame is controlled according to the time for exposure adjustable range in step 1
System, if the Time Exposure discreet value of next frame is more than the maximum value of time for exposure adjustable range, it is final true to be maximized
The Time Exposure value of fixed next frame is then minimized if it is less than the minimum value of time for exposure adjustable range finally to determine
Next frame Time Exposure value, if in time for exposure adjustable range directly be arranged next frame Time Exposure discreet value
For the Time Exposure value of the next frame of final determination；It shoots to obtain according to finally determining Time Exposure value control camera next
After frame, return to step 4 is recycled next frame as new present frame.
2. the automatic explosion method of in-vehicle camera according to claim 1, it is characterised in that：Region of interest is set in step 2
When domain, the useless automobile cabinet cover area in lower section, top sky areas and both sides of the edge in removal camera lens.
3. a kind of auto exposure system of in-vehicle camera, which is characterized in that comprise the following modules：
Initialization module, for configuring initial exposure time, time for exposure adjustable range, standard grayscale threshold range Gt, meeting
The saltus step range Gp and Gray Level Jump stabilization range Gc that PID is adjusted；
Region division module will all be handled every frame image for area-of-interest to be arranged just for area-of-interest；
Initial start module, for starting camera, when the time for exposure of first frame uses the initial exposure of initialization module configuration
Between, first frame gradation of image mean value is obtained for area-of-interest after shooting first frame, according to the standard grayscale threshold value of setting
Range Gt is carried out using gray threshold as the quick adjusting of standard, according to the time for exposure for adjusting result the second frame of setting, order control
Molding block works；
Stabilization adjustment module, for the gradation of image mean value of present frame to be compared with the gradation of image mean value of former frame,
Retain if the Gray Level Jump stabilization range Gc that two frame gray average difference DELTA G variations are not above initialization module setting and adopts
Use the Time Exposure value of present frame as the Time Exposure discreet value of next frame, command control module works, otherwise order image
Entropy adjustment module works；
Image entropy adjustment module, for carrying out the exposure tune based on image entropy if detecting traffic sign from current frame image
Section obtains the Time Exposure discreet value of next frame, command control module work, otherwise command standard judgment module work；
Standard judgment module, if the gradation of image mean value for present frame is in initialization module preset standard gray threshold range
In Gt, then Time Exposure discreet value of the Time Exposure value of present frame as next frame, command control module is directly used to work,
Otherwise order PID control adjustment module works；
PID control adjustment module, for according to the gradation of image mean value and former frame of present frame obtained by stabilization adjustment module
The difference DELTA G of gradation of image mean value further judges, if in the preset saltus step range Gp of initialization module, carries out basis
The PID of gray scale and exposure value ratio is adjusted, and obtains the Time Exposure discreet value of next frame, otherwise command control module work is ordered
Cmos sensor adjustment module is enabled to work；
Cmos sensor adjustment module, the exposure method for carried out using cmos sensor characteristic fast zoom table are obtained
The Time Exposure discreet value of next frame, command control module work；
Control module, for the Time Exposure discreet value to current gained next frame according to the time for exposure tune in initialization module
Adjusting range is controlled, if the Time Exposure discreet value of next frame is more than the maximum value of time for exposure adjustable range, is taken most
Big value then takes minimum for the Time Exposure value of finally determining next frame if it is less than the minimum value of time for exposure adjustable range
Value for finally determining next frame Time Exposure value, be directly arranged if in time for exposure adjustable range next frame when
Between exposure discreet value be final determination next frame Time Exposure value；Camera is controlled according to finally determining Time Exposure value
After shooting obtains next frame, the work of return command stabilization adjustment module is recycled next frame as new present frame.
4. the auto exposure system of in-vehicle camera according to claim 3, it is characterised in that：Sense is set in region division module
When interest region, the useless automobile cabinet cover area in lower section, top sky areas and both sides of the edge in removal camera lens.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510968340.8A CN105635597B (en) | 2015-12-21 | 2015-12-21 | The automatic explosion method and system of in-vehicle camera |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510968340.8A CN105635597B (en) | 2015-12-21 | 2015-12-21 | The automatic explosion method and system of in-vehicle camera |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105635597A CN105635597A (en) | 2016-06-01 |
| CN105635597B true CN105635597B (en) | 2018-07-27 |
Family
ID=56050045
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510968340.8A Expired - Fee Related CN105635597B (en) | 2015-12-21 | 2015-12-21 | The automatic explosion method and system of in-vehicle camera |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105635597B (en) |
Families Citing this family (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107613190B (en) * | 2016-07-11 | 2020-11-03 | 中兴通讯股份有限公司 | Photographing method and terminal |
| CN106303279B (en) * | 2016-08-31 | 2019-08-09 | 北京数字精准医疗科技有限公司 | Multi-spectrum endoscopic automatic exposure imaging method |
| CN106791476A (en) * | 2017-01-25 | 2017-05-31 | 北京图森未来科技有限公司 | A kind of image-pickup method and device |
| CN111526300A (en) * | 2017-08-17 | 2020-08-11 | 吴彬 | Video shooting method capable of obtaining clear image quality |
| CN107592470A (en) * | 2017-09-21 | 2018-01-16 | 陕西土豆数据科技有限公司 | A kind of exposure algorithm applied to more lens sensors |
| CN107592469B (en) * | 2017-10-23 | 2021-01-05 | 努比亚技术有限公司 | Exposure value adjusting method, mobile terminal and readable storage medium |
| CN108174114B (en) * | 2017-12-08 | 2020-06-30 | 上海集成电路研发中心有限公司 | Entropy calculation device and entropy calculation method |
| CN108600622B (en) * | 2018-04-12 | 2021-12-24 | 联想(北京)有限公司 | Video anti-shake method and device |
| CN108900782B (en) * | 2018-08-22 | 2020-01-24 | Oppo广东移动通信有限公司 | Exposure control method, exposure control device and electronic equipment |
| CN109639994B (en) * | 2019-01-03 | 2020-08-07 | 湖北工业大学 | Dynamic adjusting method for exposure time of embedded vehicle-mounted camera |
| CN110708473B (en) * | 2019-11-14 | 2022-04-15 | 深圳市道通智能航空技术股份有限公司 | High dynamic range image exposure control method, aerial camera and unmanned aerial vehicle |
| CN113160590B (en) * | 2020-01-23 | 2023-02-03 | 华为技术有限公司 | Control method and device of intelligent automobile and related equipment |
| CN111432134A (en) * | 2020-03-17 | 2020-07-17 | 广东博智林机器人有限公司 | Method and device for determining exposure time of image acquisition equipment and processor |
| CN112153360B (en) * | 2020-08-24 | 2022-03-18 | 青岛小鸟看看科技有限公司 | Method, device and equipment for determining exposure parameters of head-mounted equipment |
| CN112351217A (en) * | 2020-11-04 | 2021-02-09 | 中冶赛迪技术研究中心有限公司 | Image enhancement type waterproof holder image acquisition system and method |
| CN114866703A (en) * | 2021-02-03 | 2022-08-05 | 浙江舜宇智能光学技术有限公司 | Active exposure method and device based on TOF imaging system and electronic equipment |
| WO2022222028A1 (en) * | 2021-04-20 | 2022-10-27 | Baidu.Com Times Technology (Beijing) Co., Ltd. | Traffic light detection and classification for autonomous driving vehicles |
| CN113657427B (en) * | 2021-06-29 | 2024-01-23 | 东风汽车集团股份有限公司 | In-vehicle multi-source image fusion recognition method and device |
| CN115278095A (en) * | 2022-05-11 | 2022-11-01 | 岚图汽车科技有限公司 | Vehicle-mounted camera control method and device based on fusion perception |
| CN117615244B (en) * | 2024-01-22 | 2024-04-02 | 四川群源科技有限公司 | Driving method and device of image pickup equipment and electronic equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1547075A (en) * | 2003-12-05 | 2004-11-17 | Method for improving automatic exposure under low light level | |
| CN101119439A (en) * | 2007-08-22 | 2008-02-06 | 东风汽车有限公司 | Automatic control method for clear lane image shooting by vehicle-mounted camera |
| CN102369723A (en) * | 2011-08-29 | 2012-03-07 | 华为终端有限公司 | Auto exposure method and device, and imaging device |
| CN104301622A (en) * | 2014-09-30 | 2015-01-21 | 上海微小卫星工程中心 | Self-adaption space-based space target imaging exposure time control method and device |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011056994A2 (en) * | 2009-11-04 | 2011-05-12 | Board Of Regents, The University Of Texas System | Adaptive automatic exposure apparatus and method for digital images |
| FR2954034B1 (en) * | 2009-12-16 | 2012-07-20 | St Microelectronics Grenoble 2 | CONTROL OF A DYNAMIC IMAGE SENSOR |
-
2015
- 2015-12-21 CN CN201510968340.8A patent/CN105635597B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1547075A (en) * | 2003-12-05 | 2004-11-17 | Method for improving automatic exposure under low light level | |
| CN101119439A (en) * | 2007-08-22 | 2008-02-06 | 东风汽车有限公司 | Automatic control method for clear lane image shooting by vehicle-mounted camera |
| CN102369723A (en) * | 2011-08-29 | 2012-03-07 | 华为终端有限公司 | Auto exposure method and device, and imaging device |
| CN104301622A (en) * | 2014-09-30 | 2015-01-21 | 上海微小卫星工程中心 | Self-adaption space-based space target imaging exposure time control method and device |
Non-Patent Citations (2)
| Title |
|---|
| 一种面向无人驾驶汽车的高效交通标志识别方法;宋文杰等;《机器人》;20150115;第37卷(第1期);第102-111页 * |
| 基于图像熵的高动态范围场景的自动曝光算法;杨作廷等;《光子学报》;20130531;第42卷(第6期);第742-746页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105635597A (en) | 2016-06-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105635597B (en) | The automatic explosion method and system of in-vehicle camera | |
| CN101739827B (en) | Vehicle detecting and tracking method and device | |
| US8908924B2 (en) | Exterior environment recognition device and exterior environment recognition method | |
| CN103810469B (en) | Car external environment recognition device | |
| CN100579174C (en) | Motion detection method and device | |
| CN105518702A (en) | Method, device and robot for detecting target object | |
| CN110858295A (en) | Traffic police gesture recognition method and device, vehicle control unit and storage medium | |
| CN101996410A (en) | Method and system of detecting moving object under dynamic background | |
| CN102201121A (en) | System and method for detecting article in video scene | |
| CN110610150B (en) | Tracking method, device, computing equipment and medium of target moving object | |
| CN105812674A (en) | Signal lamp color correction method, monitoring method, and device thereof | |
| CN102542244A (en) | Face detection method and system and computer program product | |
| CN109326125B (en) | Picture quality diagnosis system and method based on embedded system | |
| CN111038488A (en) | Energy optimization control method and device for hybrid electric vehicle | |
| CN102339390A (en) | Method and system for updating target template of video monitoring system | |
| CN106815587A (en) | Image processing method and device | |
| CN114143940A (en) | Tunnel illumination control method, device, equipment and storage medium | |
| CN109146923A (en) | The processing method and system of disconnected frame are lost in a kind of target following | |
| CN107045630B (en) | RGBD-based pedestrian detection and identity recognition method and system | |
| CN102799861A (en) | Method for rapidly identifying reading of instrument by using color | |
| CN109996377A (en) | A kind of method for controlling street lamps, device and electronic equipment | |
| CN104282013A (en) | Image processing method and device for detecting foreground target | |
| CN102238366A (en) | Camera of realizing image tracking and monitoring and method thereof | |
| CN109509368A (en) | A kind of parking behavior algorithm based on roof model | |
| CN103065145A (en) | Vehicle movement shadow eliminating method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180727 Termination date: 20201221 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |