CN101079962A - A flicker restraint method - Google Patents
A flicker restraint method Download PDFInfo
- Publication number
- CN101079962A CN101079962A CNA2006100813816A CN200610081381A CN101079962A CN 101079962 A CN101079962 A CN 101079962A CN A2006100813816 A CNA2006100813816 A CN A2006100813816A CN 200610081381 A CN200610081381 A CN 200610081381A CN 101079962 A CN101079962 A CN 101079962A
- Authority
- CN
- China
- Prior art keywords
- flicker
- reference block
- image
- component
- pixel
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Picture Signal Circuits (AREA)
- Control Of Indicators Other Than Cathode Ray Tubes (AREA)
Abstract
This invention discloses a method for suppressing flash, which obtains a flash rule of an image from a stable part of an image, namely a reference block, then carries out image parameter correction computation to the entire image based on the sequence character of the flash rule. This invention can modify related image parameters of other pixels accurately according to the static data of an image parameter value of a reference block so as to suppress flash phenomenon effectively and the method can be realized by simple software computation. This invention also provides a method for optioning the current flash trend of a reference block to follow the change of image phase values.
Description
Technical field
The present invention relates to the digital imaging technique field, be specifically related to a kind of flicker restraint method.
Background technology
In the process of image collection, the frequency of the acquisition frame rate of (for example video camera, video first-class) and lighting source is asynchronous if image modalities, the scintillation on the display terminal will occur.With the video telephony applications is example, and radio phone terminal generally all is positioned at indoor, and fluorescent lamp is the main lighting source of indoor environment, and when the frequency of fluorescent lamp was the integral multiple of visual telephone video head acquisition frame rate, flicker did not take place.For example China adopts the 50Hz illuminator, and the acquisition frame rate of video heads is 25Hz, then can not glimmer in theory.But in practice, the frequency of light source may cause video heads can't collect the complete light source cycle in a collection period because small skew takes place near 50Hz in the influence of various factors, and at this moment scintillation will appear in radio phone terminal.Because scintillation has a strong impact on the quality and the effect of video image, common attention and research therefore have been subjected to.
O ' Rourke provides a kind of scheme that suppresses scintillation in application number is 6717621 U.S. Patent application " Method and apparatus forreducing flicker in a video image sequence ".The typical scene that this scheme is used is, under the prerequisite of camera acquisition frame per second and fluorescent light source frequency match, owing to there be second light source (display device of for example display of PC, or video telephone etc.), and also there is fixing refreshing frequency in the 2 one or two source, thus the flicker that causes.Under this scene, the frequency of flicker can be obtained by the frequency and the camera acquisition frame per second combined calculation of display.For this scene, the solution that this scheme provides is band limiting filter that centre frequency is a flicker frequency of design, can be with the flicker component filtering in the video sequence.The shortcoming of this method is that filter utilizes this frame and former frame corresponding pixel points to obtain do weighting when specific implementation, the simplest YR of being expressed as (t)=0.5y (t-1)+0.5YR (t) (YR (t) and y (t-1) represent the value of this frame and former frame pixel respectively), when bigger variation takes place in video content, this flicker restraint method that utilizes filter to realize can cause ghost phenomena, has a strong impact on subjective quality.
Chung provides the scheme of another kind of inhibition scintillation in application number is 6271884 U.S. Patent application " Image flicker reduction withfluorescent lighting ".This scheme mainly is the video camera that solves the 30Hz acquisition frame rate under the 50Hz power frequency supply environment or the video camera of the 25Hz acquisition frame rate problem of working under 60Hz power frequency supply environment.The method that this patent adopted mainly is the time for exposure of revising image modalities, make the time for exposure become the integral multiple in light source cycle, even the acquisition frame rate of image modalities and light source frequency do not match like this, but each time for exposure of gathering all mates with the light source cycle, the collection light source cycle that can be complete is so avoided the appearance of flicker.The shortcoming of this method is costly, because can be adaptive the change time for exposure collecting device often cost than higher, and this method only be applicable at the unmatched situation of frequency, can't solve because the flicker problem that little frequency shift (FS) causes takes place light source.
In addition, above-mentioned flicker suppresses scheme all at the fixing situation of flicker frequency, yet when skew took place light source frequency, side-play amount changed often over time, and this is all insurmountable problem of existing flicker inhibition scheme.
Summary of the invention
The object of the present invention is to provide a kind of can be simply, effectively and the low-cost flicker restraint method of realizing.
Further aim of the present invention is to provide a kind of flicker restraint method of situation about becoming can be applicable to flicker frequency the time.
To achieve the object of the present invention, the technical scheme of being taked is that a kind of flicker restraint method comprises: 1) choose in the image metastable part as the reference piece; 2) statistical-reference piece image parameter value is obtained the flicker rule of reference block image parameter value over time; 3) according to the flicker rule of reference block image parameter value present image is carried out the image parameter correction.
Usually because scintillation mainly impacts the brightness of image, therefore the image parameter of preferably using is brightness.At this moment, for ease of processing and the use of carrying out the reference block data, the pixel groups that can choose brightness value difference minimum in the metastable part of image is as the reference piece.
For the reference block data of guaranteeing to obtain effectively reflect actual conditions, the time that the reference block brightness value is added up is more preferably greater than a flicker cycle.
In the described step 3) image is carried out brightness correction and can adopt two kinds of methods, or respectively various brightness values are calculated corresponding DC component, utilize result of calculation that each pixel of present image is carried out brightness correction; Or respectively each pixel of present image is carried out DC component and calculate, utilize result of calculation that this pixel is carried out brightness correction.
Described flicker rule can be represented by the DC component of brightness value and the relation between the flicker parameter, the flicker parameter comprises that the flicker peak-to-peak value is worth mutually with the flicker bit, at this moment, the flicker rule shows as A: DC component=intrinsic brilliance value-flicker peak-to-peak value * flicker bit is worth mutually; B: DC component has linear relationship with the flicker peak-to-peak value.
Described step 3) can be carried out step by step by following: 3a) be worth variation tendency mutually according to the historical flicker bit of reference block and infer the flicker bit of present image and be worth mutually; 3b) according to described DC component and the DC component that concerns other pixel brightness values in the computed image between flicker peak-to-peak value and flicker bit are worth mutually; 3c) brightness value with each pixel in the present image is modified to corresponding DC component.
For reaching further aim of the present invention, at the 3a step by step of described step 3)) in preferred adopt such method: the flicker bit that the flicker bit of reference block is worth mutually as present image in present image is worth mutually.
Preferably, before described step 3), judge also whether reference block is effective) if then continue execution in step 3; If not, then or according to the variation tendency near the historical amendment record of present image present image is carried out brightness correction; Perhaps execution in step 2), add up again and refresh the reference block data.
In said method, when judging that reference block is invalid, and select again the statistical-reference blocks of data during, can carry out brightness correction to present image according to variation tendency near the historical amendment record of present image.
Perhaps, also choose in the image another metastable part as second reference block, to the same execution in step 2 of second reference block) obtain the second reference block data; When judging that reference block is invalid, and select again the statistical-reference blocks of data during, then according to the second reference block data execution in step 3).
Adopt technique scheme, beneficial technical effects of the present invention is:
1) the present invention's employing serves as according to the method for carrying out the image parameter correction reference block in the image is added up the flicker rule of obtaining, based on the flicker rule under Same Scene to image in the consistency of each pixel, can revise the respective image parameter of other pixels accurately according to the statistics of reference block image parameter value, thereby effectively suppress scintillation.And the inventive method can be calculated by simple software and realize, can reduce hardware cost.
2) the present invention further adopts before carrying out corrected Calculation, determine that according to the current flicker trend of reference block the flicker bit of present image is worth mutually, because the flicker trend of each pixel is consistent in next width of cloth picture of synchronization, therefore can still can correctly carry out under the situation that the assurance corrected Calculation becomes when flicker frequency by the tracking of reference block is judged that accurately the flicker bit of present image is worth mutually.
Description of drawings
Also the present invention is described in further detail in conjunction with the accompanying drawings below by embodiment.
Fig. 1 is a method flow schematic diagram among the embodiment one;
Fig. 2 is that the reference block brightness value changes schematic diagram in time among the embodiment one;
Fig. 3 is a method flow schematic diagram among the embodiment two;
Fig. 4 is that the reference block brightness value changes schematic diagram in time among the embodiment two;
Fig. 5 is a method flow schematic diagram among the embodiment three;
Embodiment
The invention provides a kind of flicker restraint method, its core concept is, earlier by statistics, from the metastable part of image, be the flicker rule of obtaining image in the reference block, utilize this rule to carry out the image parameter corrected Calculation of entire image based on the order of flicker rule and for the consistency of entire image then.Provide flicker rule to represent among the present invention and utilized method with big generality.Selecting brightness for use with image parameter is example, when carrying out corrected Calculation can be to calculate the pairing DC component of various brightness values earlier, and the brightness value according to each pixel carries out corresponding correction again, also can be that each pixel is calculated one by one.Situation about becoming when making corrected Calculation be applicable to flicker frequency, the present invention also provides the current flicker trend of preferably utilizing reference block to come the method for the variation that the tracking image flicker bit is worth mutually.In addition, the method for optimizing that also provides some to select reference block in the image among the present invention, and the method and the preferred process method when reference block causes losing efficacy owing to instability of reference block being carried out judgement of stability.Because brightness is the image parameter the most tangible parameter that is affected under the flicker situation, therefore, below all select for use brightness the inventive method to be elaborated as the image parameter of obtaining the flicker rule.
Embodiment one, a kind of flicker restraint method, flow process comprises as shown in Figure 1:
1) chooses in the image metastable part as the reference piece;
Owing to need add up obtaining the flicker related data reference block, so reference block need guarantee certain stability, like this this part carried out result that the statistics of a period of time obtains and is only significantly, normally selects part static relatively in the image.Concrete system of selection can be adopted multiple mode, for example, can be by image be carried out the statistical observation of a period of time, during the comparative statistics in the image variation of each pixel determine metastable part, this mode has more general applicability, is not subjected to the influence of concrete scene.For some typical application scenarioss, then can adopt simpler selection mode, for example in the majority with head shoulder image field scape for video telephony applications, upper left and upper right background often keeps static relatively in the image, can be elected to be reference block.Equally, in other application-specific scenes, also can scope that select reference block be set according to corresponding scene characteristics.
For ease of processing and the use of carrying out the reference block data, the pixel groups of further choosing brightness value difference minimum in the metastable part of image in this example is convenient to statistics is averaged processing to reduce error as the reference piece like this.Common this reference block is corresponding to the homochromy zone of a fritter of background parts in the image
As for the size of selected reference block, but the size that promptly comprises the number reference picture of pixel determines that generally speaking, in video telephony applications, the reference block of choosing size about 4 * 4 is relatively more suitable.
2) to the statistics of reference block and obtaining of flicker rule, comprise
2a) statistical-reference Block Brightness value promptly, writes down the brightness value in each pixel each two field picture in timing statistics in the reference block over time;
Owing to selected the immediate pixel groups of brightness value as the reference piece in the step 1), done as a whole consideration below with reference to piece for simplicity's sake.To the average treatment that each pixel statistics of reference block is carried out, can after each frame statistics, carry out, also can after the flicker rule that obtains each point, carry out again, below repeat no more.If find that in statistic processes unusual fluctuation has taken place some pixel, can also when averaging processing, abandon these unusual pixels, to guarantee the accuracy of reference block overall data.
Fig. 2 has provided a kind of under the flicker situation, and the reference block brightness value is schematic diagram over time.T is for gathering constantly among the figure; YR (t) is a t brightness value constantly; as seen from Figure 2; theoretically; as long as the statistical time range of reference block has covered high and minimum position t1 and the t2 of brightness value; just can determine the brightness situation of change of reference block; but because in practice; the moment that begins to add up is also uncertain; and do not get rid of the influence that can be subjected to various error components in the statistic processes; therefore be the effective complete reflection actual conditions of the reference block data guaranteeing to obtain, the time that the reference block brightness value is added up is more preferably greater than a flicker cycle T.
2b) obtain the flicker rule of reference block, i.e. the DC component of brightness value and flicker relationship between parameters;
With reference to Fig. 2, the DC component of reference block brightness value is exactly under the situation of not glimmering, the brightness value YRdc that reference block should have; The YRdc of reference block is by to step 2) in the statistics of record carry out analyzing and processing and obtain.The concrete computational methods of separating DC component from the periodicity discrete data have a lot, example adds the data segment between the adjacent extreme value is averaged, perhaps the data segment between three close values that occur is continuously averaged etc., the concrete visual applicable cases of computational methods is determined, is not construed as limiting the invention.
The brightness value YR (t) that the flicker parameter then reflects reference block is rule over time, adopts the situation of change of flicker component YRac (t) to describe this Changing Pattern in this example,
YRac(t)＝YR(t)-YRdc (1)
Obviously, the glimmer target that suppresses of image is removed YRac (t) composition among the YR (t) exactly, make the brightness value of pixel revert to YRdc.Under normal conditions, YRac (t) can be expressed as the product that flicker peak-to-peak value AYR and flicker bit are worth PR (t) mutually,
YRac(t)＝AYR×PR(t) (2)
Therefore adopt AYR and PR (t) in this example as the flicker parameter.AYR can obtain by YR (t) being carried out statistics of extremes, and YRdc and AYR have been arranged, and PR (t) can calculate by following formula easily,
PR(t)＝[YR(t)-YRdc]/AYR (3)
So far, three kinds of parameter YRdc, AYR, the PR (t) of reference block flicker rule have promptly been obtained to characterize.Obviously, the incompatible performance of parameter group that can also adopt other based on the characteristic and the principle of scintillation is as long as can reflect the variation of flicker and to the influence of image, concrete parameter set-up mode is not construed as limiting the invention.
3) utilization is carried out brightness correction from the flicker rule that reference block obtains to present image.Based on the principle of scintillation, the flicker rule has order and each pixel in the Same Scene is consistent, and therefore can utilize the flicker rule of the reference block that has obtained to revise the brightness value of present image.Based on selected parameter method for expressing in the present embodiment, corresponding to formula (1), (2) represented relation, can get t0 constantly, the correction formula of image pixel brightness value Y (t0) is
Ydc＝Y(t0)-AY×P(t0) (4)
Wherein Ydc is the DC component of expectation acquisition, and AY is the flicker peak-to-peak value under this brightness value, and P (t0) is worth mutually for t0 flicker bit constantly.Under the situation that flicker frequency is fixed, AY and P (t0) can be obtained by simple computation by flicker rule performance parameter YRdc, AYR, the PR (t) of reference block, specify as follows:
3a) being worth variation tendency mutually according to the historical flicker bit of reference block infers the flicker bit of present image and is worth mutually; Because flicker frequency fixes, can infer out follow-up t0 reference block flicker bit constantly according to the PR that adds up acquisition (t) and be worth PR (t0) mutually, and be consistent in the flicker trend of synchronization entire image, therefore have
P(t0)＝PR(t0) (5)
3b) calculate Ydc according to the relation between YRdc and AYR, the PR (t); First series of fortified passes between YRdc and AYR, the PR (t) system is shown in (3) formula, and this is the basic representation formula of scintillation, between Ydc and AY, the P (t0) identical expression is arranged also, shown in (4) formula; In addition, through experimental studies have found that, have the relation that is approximately linear between flicker peak-to-peak value and the respective straight flow component, the peak-to-peak value that promptly glimmers is linear increasing with the increase of DC component, and therefore, AYR and AY can be designated as A (YRdc) and A (Ydc), and have
A(Ydc)＝k×(Ydc-YRdc)+A(YRdc) (6)
Wherein k is the slope of linear change.(6) formula is second series of fortified passes system between DC component and the flicker peak-to-peak value, this relation is from intuitively being readily appreciated that, the brightness of pixel is big more, and it is also just obvious more naturally to the performance of flicker so, if the pixel intrinsic brightness is not high, the amplitude of then glimmering also can not be very big.Obtaining of k value can be adopted several different methods, experiment finds, for different scenario A (Ydc)～Ydc line have a fixing point (A (Y0), Y0), this point remains unchanged under different scenes substantially, therefore can determine the numerical value of k by this fixing point and acquired reference block YRdc, A (YRdc)
k＝[A(YRdc)-A(Y0)]/(YRdc-Y0) (7)
In addition, the numerical value of k also can obtain by real-time statistics,, by a reference block of choosing is in addition added up, obtains YRdc ' and A (YRdc ') that is, calculates k then
k＝[A(YRdc)-A(YRdc’)]/(YRdc-YRdc’) (8)
In order to guarantee the accuracy of k value, during a kind of method, the best and original reference block of choosing of another group reference block has bigger luminance difference after employing.
Comprehensively (4), (5), (6) formula promptly get desired DC component Ydc and are
Ydc＝[Y(t0)+k×PR(t0)×YRdc-PR(t0)×A(YRdc)]/[1+k×PR(t0)] (9)
3c) the brightness value Y (t0) with each pixel in the present image is modified to corresponding DC component Ydc.
Need to prove, image is carried out brightness correction, carry out, can also can calculate the correction value of the brightness value correspondence of each pixel one by one brightness value unified calculation correction values at different levels at brightness value.For example,, can calculate the correction value of brightness correspondences at different levels earlier, and then one by one image pixel be revised, judge that the current brightness value of this pixel is modified to corresponding correction value then for image with 256 grades of brightness; Can be the brightness value that obtains image pixel one by one also, calculate corresponding correction value, with correction value this pixel be revised again.In actual applications, the former has bigger superiority, at first this method can be avoided same brightness is carried out repeatedly double counting, and, the more important thing is, as seen from Figure 2, AYR is constant in the ordinary course of things, therefore, the variation that only is worth mutually with the flicker bit usually with the corresponding correction value of brightness at different levels changes, therefore when adopting first method, the correction value of each brightness degree correspondence can be under being preserved by integral body after the calculating very easily, if carry out being worth relevant corrected Calculation mutually once more with identical flicker bit, can needn't repeat the calculation procedure of correction value, only need to extract corresponding historical record and just can proceed makeover process.In the ideal case, the flicker bit that only needs to preserve one-period is worth corresponding accuracy table mutually, just follow-up corrected Calculation process can be reduced to the process of data extract and corresponding assignment, shorten the time of system handles corrected Calculation greatly, conserve system resources.
Embodiment two, a kind of flicker restraint method, flow process as shown in Figure 3, present embodiment method and embodiment one are basic identical, the difference part is, obtains the 3a step by step that the flicker bit is worth P (t0) mutually in the step 3)) what adopt is following method:
3a) be worth PR (t0) mutually and be worth P (t0) mutually as the flicker bit of present image with the pairing flicker bit of flicker component YRac (t0) of reference block in the present image.
Fig. 4 has provided a kind of the change under the situation when flicker frequency, the reference block brightness value is schematic diagram over time.TR is the statistical time range of reference block among the figure, the same Fig. 1 of all the other meaning of parameters.As seen from Figure 4, after the measurement period of reference block, change has taken place in flicker frequency, when t0 image is constantly revised, obviously is incorrect if still infer PR (t0) with PR (t) Changing Pattern of TR period.At this moment, need to use the method that provides in this example, follow the tracks of the variation that the flicker bit is worth PR (t0) mutually with the flicker component YRac (t0) of current reference block, (3) formula of utilization is calculated and is obtained PR (t0)
PR(t0)＝YRac(t0)/AYR＝[YR(t0)-YRdc]/AYR (10)
Carry out follow-up corrected Calculation with PR (t0) as P (t0) according to (5) formula again.
Compare with embodiment one, the outstanding advantage of present embodiment method is, can follow the tracks of the situation of change of flicker frequency immediately, the inhibition of still in real time, accurately glimmering under the situation about when flicker frequency, becoming, expand the scope of application of the inventive method greatly, remedied the defective that becomes situation when prior art can't be handled flicker frequency.
Embodiment three, a kind of flicker restraint method, flow process as shown in Figure 5, present embodiment method and embodiment two are basic identical, the difference part be, before step 3), also comprise:
A1) judge whether reference block is effective, if, then forward step 3) to, if not, execution in step a2 then);
Judgement to reference block validity can utilize multiple mode, for example, relatively the present frame reference block with the difference of former frame reference block, relatively resemble examine whether each pixel brightness value is consistent in the piece, whether current reference block brightness have big deviation etc. with historical record, various judgment modes can be independently or the mixing use.
A2) execution in step 2), add up again and refresh the reference block data; And during statistical-reference blocks of data again,, forward step 3c then to) according to calculating or extracting corresponding correction value near the variation tendency of the historical amendment record of present image.With Fig. 4 is example, if judge constantly that at t0 reference block lost efficacy, then infer t0 reference block data constantly (as can be known by the analysis among the embodiment two according to data near the reference block significant instant of t0, mainly be to infer the flicker bit to be worth mutually, the flicker frequency of inferring institute's foundation is historical record reflected in a period of time near t0 flicker frequency).
Increased in the present embodiment method reference block has been carried out the step that validity is judged, carried out purpose that validity judges and be whether can also reflecting effectively by the flicker rule that reference block obtains whether the reference block data of flicker rule under the current scene and present frame are available in order to determine.The former changes corresponding to scene or reference block moves, and it is permanent that consequent reference block lost efficacy, and comprises that promptly the reference block data of historical statistics value all lost efficacy; The latter is then temporarily unstable corresponding to the reference block that interference causes, for example, reference block be moved temporarily that object blocks or by moment light source produce to the temporary transient destruction of current scene etc., it is temporary transient that the reference block that produce this moment lost efficacy, after disturbing factor was loss, historical statistical data was still available.
Yet under normal conditions, be difficult to accurately know that by once judging reference block is because scene changes the permanent failure that causes or disturbs the temporary transient inefficacy that causes, therefore, usually can adopt dual mode to handle: the one, employed method in the present embodiment, promptly, lost efficacy in case judge reference block, no matter whether reality has taken place that scene changes or reference block such as moves at situation, all reference block is added up again; The one,, in one period judging reference block inefficacy beginning first, think that reference block just temporarily lost efficacy, do not carry out refreshing of reference block data, failure conditions judged in record simultaneously, after the continuous Failure count of reference block reaches certain value, then think the reference block permanent failure, carry out refreshing of reference block data.In addition, also can determine whether carrying out refreshing of reference block data according to the volume difference situation (or deviation situation of accumulation reference block data) of current reference block data, generally speaking, can carry out layout and selection according to concrete application scenarios and equipment situation to judgement and processing that reference block lost efficacy, concrete judgment processing mode is not construed as limiting the invention.
Present embodiment can strengthen the adaptive capacity that flicker restraint method of the present invention changes scene, when situations such as occurrence scene sudden change or switching, can be according to judging automatically that again the statistical-reference blocks of data (chooses under the situation of reference block adopting statistical method, even can reselect reference block), the effect that the reference block data influence flicker of avoiding losing efficacy suppresses.
Embodiment four, a kind of flicker restraint method, present embodiment method and embodiment three are basic identical, the difference part is, also chooses in the image another metastable part in the present embodiment as second reference block, to the same execution in step 2 of second reference block) obtain the second reference block data; Step a2 at embodiment four) in again the statistical-reference blocks of data during, be according to the second reference block data execution in step 3), carry out the correction operation of image.
The benefit of present embodiment is, adopt two reference blocks (when choosing, can consider to select physical location to be separated by and teach two pixel groups far away, for example, the upper left corner of image and the upper right corner), when one of can be therein being interfered interim the inefficacy, can be dependent on another reference block and finish and revise operation, strengthen the ageing and reliability of revising operation.
More than a kind of flicker restraint method provided by the present invention is described in detail, used specific case herein principle of the present invention and execution mode are set forth, the explanation of above embodiment just is used for helping to understand method of the present invention and core concept thereof; Simultaneously, for one of ordinary skill in the art, according to thought of the present invention, the part that all can change in specific embodiments and applications, in sum, this description should not be construed as limitation of the present invention.
Claims (11)
1, a kind of flicker restraint method is characterized in that, comprising:
1) chooses in the image metastable part as the reference piece;
2) statistical-reference piece image parameter value is obtained the flicker rule of reference block image parameter value over time;
3) according to the flicker rule of reference block image parameter value present image is carried out the image parameter correction.
2, flicker restraint method according to claim 1 is characterized in that: described image parameter is brightness.
3, flicker restraint method according to claim 2 is characterized in that: the pixel groups of choosing brightness value difference minimum in the metastable part of image is as the reference piece.
4, flicker restraint method according to claim 2 is characterized in that: the time that the reference block brightness value is added up is greater than a flicker cycle.
5, flicker restraint method according to claim 2 is characterized in that: in the described step 3) or respectively various brightness values are calculated corresponding DC component, utilize result of calculation that each pixel of present image is carried out brightness correction; Or
Respectively each pixel of present image is carried out DC component and calculate, utilize result of calculation that this pixel is carried out brightness correction.
6, according to any described flicker restraint method of claim 2～5, it is characterized in that: described flicker rule comprises the DC component of brightness value and the relation between the flicker parameter, described flicker parameter comprises that the flicker peak-to-peak value is worth mutually with the flicker bit, and described flicker rule shows as
A: DC component=intrinsic brilliance value-flicker peak-to-peak value * flicker bit is worth mutually
B: DC component has linear relationship with the flicker peak-to-peak value.
7, flicker restraint method according to claim 6 is characterized in that, described step 3) comprises:
3a) being worth variation tendency mutually according to the historical flicker bit of reference block infers the flicker bit of present image and is worth mutually;
3b) according to described DC component and the DC component that concerns other pixel brightness values in the computed image between flicker peak-to-peak value and flicker bit are worth mutually;
3c) brightness value with each pixel in the present image is modified to corresponding DC component.
8, flicker restraint method according to claim 6 is characterized in that, described step 3) comprises:
3a) flicker bit that is worth mutually as present image with the flicker bit of reference block in the present image is worth mutually;
3b) according to described DC component and the DC component that concerns other pixel brightness values in the computed image between flicker peak-to-peak value and flicker bit are worth mutually;
3c) brightness value with each pixel in the present image is modified to corresponding DC component.
9, flicker restraint method according to claim 6 is characterized in that: before described step 3), judge also whether reference block is effective, if then continue execution in step 3); If not, then or
According to the variation tendency near the historical amendment record of present image present image is carried out brightness correction; Perhaps
Execution in step 2), add up again and refresh the reference block data.
10, flicker restraint method according to claim 9, it is characterized in that: when judging that reference block is invalid, and select again the statistical-reference blocks of data during, according to the variation tendency near the historical amendment record of present image present image is carried out brightness correction.
11, flicker restraint method according to claim 9 is characterized in that: also choose in the image another metastable part as second reference block, to the same execution in step 2 of second reference block) obtain the second reference block data;
When judging that reference block is invalid, and select again the statistical-reference blocks of data during, according to the second reference block data execution in step 3).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200610081381A CN100591094C (en) | 2006-05-24 | 2006-05-24 | A flicker restraint method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200610081381A CN100591094C (en) | 2006-05-24 | 2006-05-24 | A flicker restraint method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101079962A true CN101079962A (en) | 2007-11-28 |
| CN100591094C CN100591094C (en) | 2010-02-17 |
Family
ID=38907166
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200610081381A Expired - Fee Related CN100591094C (en) | 2006-05-24 | 2006-05-24 | A flicker restraint method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100591094C (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102752513A (en) * | 2011-04-22 | 2012-10-24 | 奥林巴斯映像株式会社 | Camera and method for controlling camera |
| CN106559623A (en) * | 2015-09-24 | 2017-04-05 | 卡西欧计算机株式会社 | Camera head and flicker detection method |
| CN108174117A (en) * | 2015-08-10 | 2018-06-15 | 广东欧珀移动通信有限公司 | Eliminate method, system and the mobile terminal and medium product of flicker |
-
2006
- 2006-05-24 CN CN200610081381A patent/CN100591094C/en not_active Expired - Fee Related
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102752513A (en) * | 2011-04-22 | 2012-10-24 | 奥林巴斯映像株式会社 | Camera and method for controlling camera |
| CN102752513B (en) * | 2011-04-22 | 2015-04-22 | 奥林巴斯映像株式会社 | Camera and method for controlling camera |
| CN108174117A (en) * | 2015-08-10 | 2018-06-15 | 广东欧珀移动通信有限公司 | Eliminate method, system and the mobile terminal and medium product of flicker |
| CN108174117B (en) * | 2015-08-10 | 2021-04-02 | Oppo广东移动通信有限公司 | Method and system for eliminating flicker, mobile terminal and medium product |
| CN106559623A (en) * | 2015-09-24 | 2017-04-05 | 卡西欧计算机株式会社 | Camera head and flicker detection method |
| CN106559623B (en) * | 2015-09-24 | 2019-11-01 | 卡西欧计算机株式会社 | Photographic device and flicker detection method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100591094C (en) | 2010-02-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1254086C (en) | Image processor, image processing method and processing program | |
| CN1744674A (en) | Video electronic flutter-proof method | |
| CN1130069C (en) | False contour correcting apparatus and method | |
| CN1678024A (en) | Improved motion vector estimation at image borders | |
| CN1472954A (en) | Circuit and method for improving image quality by fram correlation | |
| CN1951121A (en) | Block noise reducing device and image display | |
| CN1441595A (en) | Contrast correction circuit | |
| CN1305317C (en) | Image brightness correcting method of video monitoring system | |
| CN104702964B (en) | A kind of image processing method and device | |
| CN100342710C (en) | Structure method for enhancing image | |
| CN1753501A (en) | Method of selecting H.264/AVC frame to frame motion estimation mode | |
| CN200956040Y (en) | Digital oscillograph | |
| CN101079962A (en) | A flicker restraint method | |
| CN1929552A (en) | Spatial domain pixel data processing method | |
| CN1863319A (en) | Fractional pixel interpolation method and encoding and decoding apparatus | |
| CN1254087C (en) | Image processor, image processing method and processing program | |
| CN1703095A (en) | Apparatus and method for detecting letter box, and mpeg decoding device having the same | |
| CN1245018C (en) | Method for eliminating scintilltion of moving picture | |
| CN1578470A (en) | Equipment and method for coding video frequency data | |
| CN1270527C (en) | Scan conversion appts. | |
| CN1649395A (en) | Image processing apparatus and method | |
| CN103428522B (en) | Method for improving image quality at low illumination level of web camera | |
| CN1607834A (en) | Device and method for estimating the noise of a video signal | |
| CN2838183Y (en) | Device for improving video image effect dynamically | |
| CN1193618C (en) | Pre-processing method for motion estimation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20100217 Termination date: 20170524 |