CN102137221A - Method and system for sharpening digital video image - Google Patents
Method and system for sharpening digital video image Download PDFInfo
- Publication number
- CN102137221A CN102137221A CN201010023153XA CN201010023153A CN102137221A CN 102137221 A CN102137221 A CN 102137221A CN 201010023153X A CN201010023153X A CN 201010023153XA CN 201010023153 A CN201010023153 A CN 201010023153A CN 102137221 A CN102137221 A CN 102137221A
- Authority
- CN
- China
- Prior art keywords
- new
- sharpening
- current pixel
- pixel
- overshoot
- 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
Images
Landscapes
- Picture Signal Circuits (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a method and a system for sharpening a digital video image. The method comprises the following steps of: performing sharpening filtration on the present pixel to obtain a sharpening gain delta; if the delta is more than or equal to 0, performing positive sawtooth control and positive overshoot width control, otherwise, performing negative sawtooth control and negative overshoot width control to obtain a new sharpening gain delta new; adding the sharpening gain delta new and the brightness value of the present pixel to obtain the updated present pixel; and if the updated present pixel is more than the maximum value of the pixels in the neighborhood of the present pixel or less than the minimum value of the pixels in the neighborhood of the present pixel, performing overshoot amplitude on the updated present pixel, otherwise, keeping the updated present pixel unchanged. By the method and through the system, the amplitude and the width of positive overshoot and negative overshoot can be convenient to adjust independently; good balance is performed in a sharpness and sawtooth effect time; and a sharpened result image is natural and smooth while the definition becomes better.
Description
Technical field
The invention belongs to the digital video signal processing neighborhood, specifically, relate to a kind of sharpening method and system of digital video image.
Background technology
The sharpening of digital video image is called definition usually again and strengthens, refers to edge of image is promoted so that image looks more clear sharp keen, and be one of most important image enhancement technique.In the Digital Video Processing chip, because sharpening longitudinally needs a lot of row cache resources, common way is only to carry out horizontal sharpening to handle.The present invention also handles as the problem of discussing with horizontal sharpening.In the Digital Video Processing chip design of reality, if resource allows, the technology that the horizontal sharpening of introducing among the present invention is handled can be applied to fully in vertical sharpening processing and go.
For digital video image, sharpening is only handled its luminance signal.Usually digital video image is represented with yuv space in the Digital Video Processing chip, and for another kind of common format, i.e. the image represented of rgb space can be transformed into the space of YUV by simple space conversion.In view of this space conversion is well-known technology at the Digital Video Processing neighborhood, we do not do detailed description at this, and the sharpening among hypothesis the present invention is the brightness at image, and promptly Y-signal is handled.Simultaneously we suppose to handle to as if the digital picture of 8 bits, so wherein the brightness Y value of pixel in interval [0,255].
Fig. 1 is the schematic diagram of the edge configuration of adjacent area in the image.In Fig. 1, originally the brightness Y1 from image-region 1 is very slow to the border transition district the brightness Y2 of image-region 2, that is to say that the slope slope_old of this transition region is smaller.After the sharpening processing, the slope of the brightness curve in this transition region has become bigger slope_new.The lifting of luminance transition slope of a curve is sharpening in the transition region, or claims the core content that definition strengthens.When slope promoted, we saw that also the easement curve of brightness has presented the phenomenon of the negative overshoot of positive hyperharmonic.Under normal conditions, the sensation that the appearance of the negative overshoot of positive hyperharmonic can get a promotion for a kind of image definition of beholder really, but excessive overshoot meeting makes the edges of regions in the image " two-wire " (double-line occur, refer to since wide strong excessively just/negative overshoot makes edges of regions side by side a bright line and a concealed wire occur) thereby make the image very nature that seems, the fatigue of this factitious sharpening result also very easily causes image-watching person's eyes.
On the other hand, sharpening when improving edge sharpness, also might produce a kind of sawtooth effect in that slow border transition district is narrowed on the border that tilts.As shown in Figure 2, when edges of regions is strict vertical the time, we can allow the sharpening operation does obtains a very sharp keen border very by force.Yet when edges of regions had certain angle of inclination, strong excessively sharpening is handled will bring the sawtooth effect shown in the figure.This sawtooth effect is the another kind of side effect that sharpening is handled, and it can make video image seem very not untrue naturally.
In sum, in the Video processing neighborhood, simply not the brightness slope that promotes the edge transition district to the image sharpening effect of requirement, also require the result images after the sharpening when definition improves, to seem natural and tripping.This just requires the sharpener can be independently and regulate the amplitude and the width of the negative overshoot of positive hyperharmonic and the balance of making easily between acutance and sawtooth effect.For the adjusting of overshoot, common cognition is that the amplitude of positive overshoot is littler than negative overshoot amplitude, this be because human eye to bright excessively edge fatiguability more.In addition, the width of overshoot can not be too big, and not so the sensation of " two-wire " can be very strong and the image nature that also can therefore seem.For sawtooth effect, we then must sacrifice some acutancees in the position of the transition region central authorities that keep to the side and eliminate.In a word, in the Digital Video Processing chip design, have independent adjustable just/negative overshoot amplitude and width, and the sharpener of sawtooth effect power has been generally believed it is to reach a kind of necessity of superb image quality and strong means.The present invention just will introduce a kind of sharpener that satisfies this requirement.
Summary of the invention
The object of the present invention is to provide a kind of sharpening method and system of digital video image, to overcome when the brightness slope that promotes the image border transition region carries out sharpening the not good technical problem of natural and tripping property of image.
In order to achieve the above object, technical scheme of the present invention is as follows:
A kind of sharpening method of digital video image, comprise the steps: to set be positioned on the same image line with current pixel P
XBe N the pixel at center, wherein N is the odd number more than or equal to 3; Utilize a described N pixel that current pixel is carried out obtaining the sharpening increment Delta after the sharpening filtering of N rank; In Δ 〉=0 o'clock, carry out obtaining new sharpening increment Delta new after positive sawtooth control and the positive overshoot width control; Otherwise obtain new sharpening increment Delta new after bearing sawtooth control and bearing the control of overshoot width; The current pixel P that obtains upgrading after the added luminance with this sharpening increment Delta new and current pixel
XNew; If the current pixel that upgrades is greater than the maximum Gmax of pixel in the neighborhood of current pixel, perhaps the current pixel of Geng Xining is less than the minimum value Gmin of pixel in the neighborhood of current pixel, the current pixel of this renewal carried out the overshoot amplitude control; Otherwise the current pixel that upgrades remains unchanged.
A kind of sharpening system of digital video image comprises: the sharpening filter, be used on the same image line of being positioned at of video image line with current pixel P
XFor N the pixel at center described current pixel is carried out obtaining the sharpening increment Delta after the sharpening filtering of N rank, wherein N is the odd number more than or equal to 3; Positive sawtooth control unit that links to each other and positive overshoot width control unit in Δ 〉=0 o'clock, carry out obtaining new sharpening increment Delta new after positive sawtooth control and the positive overshoot width control; The negative sawtooth control unit and the negative overshoot width control unit that link to each other in Δ＜0 o'clock, bear sawtooth control and are born and obtain new sharpening increment Delta new after the overshoot width is controlled; Adder, the current pixel P that obtains upgrading after the added luminance with described sharpening increment Delta new and current pixel
XNew; Overshoot amplitude control unit receives the current pixel P from the renewal of adder
XIf new is the current pixel P that upgrades
XNew is greater than the maximum Gmax of pixel in the neighborhood of current pixel, then by a positive excess modulation control register current pixel of this renewal carried out positive excess modulation control; If the current pixel P that upgrades
XNew then bears the control of overshoot amplitude by a negative overshoot amplitude control register to the current pixel of this renewal less than the minimum value Gmin of pixel in the neighborhood of current pixel; Otherwise the current pixel that upgrades remains unchanged.
Adopt said method of the present invention and system, relevant register is provided or has adjusted coefficient to make the user independently to regulate the edge sharpening effect of wanting to reach the width and the amplitude of positive and negative overshoot.In addition, in the adjusting algorithm to the overshoot width, we have added positive and negative sawtooth control register jaggyKp and jaggyKn again, can also control the power of edge sawtooth like this in width that carries out positive and negative overshoot and amplitude control.Thus, can be independently and regulate the amplitude and the width of the negative overshoot of positive hyperharmonic easily, and the balance of between acutance and sawtooth effect, making, it is natural and tripping that the result images after the sharpening seems when definition improves.
Description of drawings
Fig. 1 is the schematic diagram of the edge configuration of adjacent area in the image;
Fig. 2 is the schematic diagram of sharpening and edge sawtooth effect;
Fig. 3 is a sharpening handling process commonly used;
Fig. 4 is the sharpening operating process among the present invention.
Embodiment
According to Fig. 3 and Fig. 4, provide preferred embodiment of the present invention, and described in detail below, enable to understand better function of the present invention, characteristics.
Fig. 3 has described sharpening handling process commonly used.We only handle at horizontal sharpening herein, and suppose that it is to carry out in the delegation of image that horizontal sharpening is handled, and after the image pixel in the delegation is handled, carry out the sharpening of next line image pixel again and handle, and finish dealing with up to the sharpening of entire image.But the method for introducing among the present invention both had been applicable to and laterally also had been used in image sharpening processing longitudinally.When carrying out image sharpening processing longitudinally, the video image among Fig. 3 is about to become the video image row.At this, as the example of a narration, we only utilize the single file view data to carry out horizontal sharpening and handle.
In sharpening commonly used is handled, the sharpening filter can be designed voluntarily by the user, to produce a sharpening increment to original image signal, the big I of sharpening increment is regulated by the sharpening strength control by the user, when this increment has just produced the effect of sharpening during by the add-back original image signal.In addition, the user also can carry out the operation of a low-pass filtering on original image signal, and purpose is to eliminate certain noise, but this low pass filter is selectable, is not enforceable.Method among the present invention also adopts the flow process among similar Fig. 3, but has carried out extra judgement and processing for the sharpening increment Delta so that the result images after the sharpening not only definition get a promotion, and seem natural and tripping.
Sharpening operating process among the present invention has been described among Fig. 4.Wherein " positive sawtooth " refers to the sawtooth of a brighter side that produces on the zone boundary, and " negative sawtooth " is the sawtooth of the darker side that produces on the zone boundary.In the elimination of sawtooth effect, need will a brighter side and the sawtooth of a darker side all smoothly fall just can make the border level and smooth nature that seems.Because different beholders is different with the sensitivity of the sawtooth of a darker side to the sawtooth of a brighter side, so we provide two separate control registers in the control procedure of sawtooth effect, be positive sawtooth control register jaggyKp ∈ [0,255], with negative sawtooth control register jaggyKn ∈ [0,255], wherein the positive sawtooth of the big more correspondence of jaggyKp and jaggyKn and negative sawtooth are weak more, be that sawtooth effect is weak more, otherwise then sawtooth effect is strong more.Similarly, previously described by us, we also provide separate positive overshoot width control register Xp ∈ [0,255] and negative overshoot width control register Xn ∈ [0,255], wherein the width of the negative overshoot of the positive hyperharmonic of the big more correspondence of Xp and Xn is big more, otherwise then corresponding overshoot width is more little; And positive excess modulation control register Ap ∈ [0,255] and negative overshoot amplitude control register An ∈ [0,255], wherein the amplitude of the negative overshoot of the positive hyperharmonic of the big more correspondence of Ap and An is more little, on the contrary then corresponding overshoot amplitude is big more.By these registers, the user can regulate the power of the edge sawtooth effect of being introduced by the sharpening operation easily, and just/bear the amplitude and the width of overshoot, to reach the sharpening result who oneself wants.
Below we illustrate one by one to the module among Fig. 4.Because only carrying out horizontal sharpening in the delegation of image, we handle, so we suppose that the part image pixel is:
……p1?p2?p3?(p4)?p5?p6?p7……
Wherein p4 is current processed pixels.Here we suppose that the sharpening filter is that 7 rank are for the linear filter with current pixel centering position, so the sharpening computing of each current pixel p4 need be 7 pixels, the i.e. p1～p7 at center with p4.In actual use, the user can design the sharpening filter voluntarily, and the present invention does not limit the type and the exponent number of sharpening filter.The local pixel here can set that to be positioned on the same image line be N the pixel at center with current pixel PX; Wherein N gets final product more than or equal to 3 odd number.For convenience of explanation, in the present embodiment N is set at 7, current pixel P
XCorrespondingly be p4.
The linear sharpening filter in 7 rank of supposing us is F=[f1, f2, f3, f4, f3, f2, f1], have so:
Δ＝K*[f1*(p1+p7)+f2*(p2+p6)+f3*(p3+p5)+f4*p4]
Wherein K is a sharpening strength control register, K 〉=0, and the corresponding sharpening intensity of big more K value is big more.Theoretically, K can be any number more than or equal to 0, but in actual applications, excessive K does not have meaning, so can limit the upper limit of K.K for example can be set be one 8 register, 4 decimals of 4 integers, the scope of K has just become 0～15.9375 like this.The concrete F filter of the scope that K needs and user is also relevant, thus K herein 〉=0, and the user can set up the upper limit of K parameter according to the actual needs on their own and change precision.
After obtaining the sharpening increment Delta, we divide two passages to handle the situation of Δ 〉=0 and the situation of Δ＜0 respectively.In Δ 〉=0 o'clock, the brightness value of current pixel p4 will be increased, and this is corresponding to the situation of positive sawtooth and positive overshoot.Opposite, when Δ＜0, the brightness value of p4 will reduce, corresponding to the situation of negative sawtooth and negative overshoot.The operation of finishing in the positive sawtooth control unit is:
Calculating variable hd=max (| p3-p5|, | p4-p5|, | p3-p4|)
yp＝hd*(jaggyKp+1.0)/256
if yp＞1.0
then
yp＝1.0
Similarly, the situation in Δ＜0, the operation of finishing in the negative sawtooth control unit is:
Calculating variable hd=max (| p3-p5|, | p4-p5|, | p3-p4|)
yp＝hd*(jaggyKn+1.0)/256
if yp＞1.0
then
yp＝1.0
For the control of overshoot width, what positive overshoot width control unit was finished is operating as:
if yp＜(1-Xp/256)/2
then
Δnew＝Δ*(yp+Xp/256)
else
Δnew＝Δ*(1-yp)
What negative overshoot width control unit was finished is operating as:
if yp＜(1-Xn/256)/2
then
Δnew＝Δ*(yp+Xn/256)
else
Δnew＝Δ*(1-yp)
After finishing sawtooth control and the control of overshoot width, we have obtained the sharpening increment Delta new through revising.The brightness value L P of the current pixel after this increment Delta new and the low-pass filtering (p4) addition obtains:
p4new＝LP(p4)+Δnew
Next p4new is carried out the judgement and the control of overshoot amplitude.At first we define the neighborhood of current pixel p4
G＝(p2，p3，p4，p5，p6)
Overshoot in order to decision current pixel position.Here be noted that the example in just the present invention's explanation chosen of as above neighborhood G, the present invention is the neighborhood of limited subscriber definition oneself not, the physical significance of this neighborhood is: the overshoot at current pixel place is to be that the basis defines with the maximum of pixel among the neighborhood G and minimum value, the positive excess peaked part that is adjusted to greater than pixel among the neighborhood G, and negative overshoot is the part less than the minimum value of pixel among the neighborhood G.Thus, just/negative overshoot amplitude control unit in:
Gmax＝max(p2，p3，p4，p5，p6)
Gmin＝min(p2，p3，p4，p5，p6)
if p4new＞Gmax
then
p4new＝p4new-(p4new-Gmax)*Ap/256
else?if p4new＜Gmin
then
p4new＝p4new+(Gmin-p4new)*An/256
else
P4new remains unchanged
Handle with the sharpening that p4new replaces the brightness value of current pixel p4 just to finish p4.Be the next pixel in the same image line then, finish, carry out the processing of next image line again, until the sharpening of entire image is finished dealing with up to the processes pixel of this image line.
Correspondingly, consult Fig. 4, a kind of sharpening system of digital video image comprises: the sharpening filter receives that capable being positioned on the same image line is that 7 pixels at center are to obtain the sharpening increment Delta with current pixel p4 from video image; Positive sawtooth control unit that links to each other and positive overshoot width control unit in Δ 〉=0 o'clock, carry out obtaining new sharpening increment Delta new after positive sawtooth control and the positive overshoot width control; The negative sawtooth control unit and the negative overshoot width control unit that link to each other in Δ＜0 o'clock, bear sawtooth control and are born and obtain new sharpening increment Delta new after the overshoot width is controlled; Adder, the current pixel p4new that obtains upgrading after the added luminance with described sharpening increment Delta new and current pixel, the brightness value of current pixel herein also can be selected to carry out earlier after the low-pass filtering and sharpening increment Delta new addition again, this sharpening system also needs to comprise a selectable low pass filter under this kind situation, and current pixel p4 is carried out low-pass filtering to obtain through the current pixel LP of low-pass filtering (p4); Just/negative overshoot amplitude control unit, reception is from the current pixel p4new of the renewal of adder, if the current pixel p4new that upgrades is greater than the maximum Gmax of pixel in the neighborhood of current pixel, then the current pixel of this renewal is carried out positive excess modulation control by a positive excess modulation control register; If the current pixel p4new that upgrades less than the minimum value Gmin of pixel in the neighborhood of current pixel, then bears the control of overshoot amplitude by a negative overshoot amplitude control register to the current pixel of this renewal; Otherwise the current pixel that upgrades remains unchanged.
Described positive sawtooth control unit comprises: the module of calculating variable hd: hd=max (| p3-p5|, | p4-p5|, | p3-p4|); Calculate the module of variable yp: yp=hd* (jaggyKp+1.0)/256, wherein jaggyKp is the adjustment coefficient that positive sawtooth control register is provided with, jaggyKp ∈ [0,255]; And judge module, if yp＞1.0 make yp=1.0.
Described positive overshoot width control unit comprises: judge that yp is whether less than the module of (1-Xp/256)/2; With the assignment module; If yp＜(1-Xp/256)/2, Δ new=Δ * (yp+Xp/256); Otherwise Δ new=Δ * (1-yp); Wherein Xp is the adjustment coefficient that positive overshoot width control register is provided with, Xp ∈ [0,255].
Described negative sawtooth control unit comprises: the module of calculating variable hd: hd=max (| p3-p5|, | p4-p5|, | p3-p4|); Calculate the module of variable yp: yp=hd* (jaggyKn+1.0)/256, wherein jaggyKn is the adjustment coefficient of negative sawtooth control register setting, jaggyK ∈ [0,255]; And judge module, if yp＞1.0 make yp=1.0.
Described negative overshoot width control unit comprises: judge that yp is whether less than the module of (1-Xn/256)/2; With the assignment module; If yp＜(1-Xn/256)/2, Δ new=Δ * (yp+Xn/256); Otherwise Δ new=Δ * (1-yp); Wherein Xn is the adjustment coefficient of negative overshoot width control register setting, Xn ∈ [0,255].
Described sharpening filtering is the linear sharpening filtering in N rank.
The overshoot amplitude control unit of the current pixel that upgrades during greater than the maximum Gmax of pixel in the neighborhood of current pixel comprises: the module that recomputates the current pixel p4new of renewal: p4new=p4new-(p4new-Gmax) * Ap/256; Wherein Ap is the adjustment coefficient that positive excess modulation control register is provided with, Ap ∈ [0,255].
The overshoot amplitude control unit of the current pixel that upgrades during less than the minimum value Gmin of pixel in the neighborhood of current pixel comprises: the module that recomputates the current pixel p4new of renewal: p4new=p4new+ (Gmin-p4new) * An/256; Wherein An is the adjustment coefficient of negative overshoot amplitude control register setting, An ∈ [0,255].
In a word, emphasis of the present invention is: sharpening always produces positive and negative overshoot, and the user needs and can independently regulate the edge sharpening effect of wanting to reach to the width and the amplitude of positive and negative overshoot, and the present invention just provides relevant register to realize this point.In addition, in the adjusting algorithm to the overshoot width, we have added positive and negative sawtooth control register jaggyKp and jaggyKn again, can also control the power of edge sawtooth like this in width that carries out positive and negative overshoot and amplitude control.For the two-wire effect,, then more or less all can there be the two-wire effect to produce, always because have the existence of positive and negative overshoot as long as carry out sharpening.Method of the present invention provides the means of positive and negative overshoot width of free adjustment and amplitude to the user, and user's pattern that just can adjust the positive and negative overshoot of oneself liking is come like this, just the pattern of two-wire.For overcoming of sawtooth effect, lie in the control of overshoot width, promptly produce corresponding yp (referring to the calculating of top yp), and the relation (referring to the calculating of top Δ new) of the size of yp and positive and negative overshoot width control register Xp and Xn has determined the numerical value of Δ new by positive and negative sawtooth control register jaggyKp and jaggyKn.The Δ new has here comprised to sawtooth with to the control of overshoot width.
The front provides the description to preferred embodiment, so that any technical staff in this neighborhood can use or utilize the present invention.To this preferred embodiment, the technical staff in this neighborhood can make various modifications or conversion on the basis that does not break away from the principle of the invention.Should be appreciated that these modifications or conversion do not break away from protection scope of the present invention.
Claims (18)
1. the sharpening method of a digital video image comprises the steps:
Setting be positioned on the same image line with current pixel P
XN pixel for the center; Wherein N is the odd number more than or equal to 3;
Utilize a described N pixel that described current pixel is carried out obtaining the sharpening increment Delta after the sharpening filtering of N rank;
In Δ 〉=0 o'clock, carry out obtaining new sharpening increment Delta new after positive sawtooth control and the positive overshoot width control; Otherwise obtain new sharpening increment Delta new after bearing sawtooth control and bearing the control of overshoot width;
The current pixel P that obtains upgrading after the added luminance with this sharpening increment Delta new and current pixel
XNew;
If the current pixel that upgrades is greater than the maximum Gmax of pixel in the neighborhood of current pixel, perhaps the current pixel of Geng Xining is less than the minimum value Gmin of pixel in the neighborhood of current pixel, the current pixel of this renewal carried out the overshoot amplitude control; Otherwise the current pixel that upgrades remains unchanged.
2. the sharpening method of digital video image as claimed in claim 1 is characterized in that, described positive sawtooth control comprises the steps:
Calculating variable hd:hd=max (| P
X-1-P
X+1|, | P
X-P
X+1|, | P
X-1-P
X|); P wherein
X-1Be current pixel P
XLast pixel, P
X+1Be current pixel P
XA back pixel;
Calculate variable yp:yp=hd* (jaggyKp+1.0)/256, wherein jaggyKp is positive sawtooth control coefrficient, jaggyKp ∈ [0,255];
If yp＞1.0 make yp=1.0.
3. the sharpening method of digital video image as claimed in claim 2 is characterized in that, described positive overshoot width control comprises the steps:
If yp＜(1-Xp/256)/2, Δ new=Δ * (yp+Xp/256); Otherwise Δ new=Δ * (1-yp); Wherein Xp is positive overshoot width control coefrficient, Xp ∈ [0,255].
4. the sharpening method of digital video image as claimed in claim 1 is characterized in that, described negative sawtooth control comprises the steps:
Calculating variable hd:hd=max (| P
X-1-P
X+1|, | P
X-P
X+1|, | P
X-1-P
X|); P wherein
X-1Be current pixel P
XLast pixel, P
X+1Be current pixel P
XA back pixel;
Calculate variable yp:yp=hd* (jaggyKn+1.0)/256, wherein jaggyKn is negative sawtooth control coefrficient, jaggyKn ∈ [0,255];
If yp＞1.0 make yp=1.0.
5. the sharpening method of digital video image as claimed in claim 4 is characterized in that, described negative overshoot width control comprises the steps:
If yp＜(1-Xn/256)/2, Δ new=Δ * (yp+Xn/256); Otherwise Δ new=Δ * (1-yp); Wherein Xn is negative overshoot width control coefrficient, Xn ∈ [0,255].
6. the sharpening method of digital video image as claimed in claim 1 is characterized in that, described sharpening filtering is the linear sharpening filtering in N rank.
7. as the sharpening method of the described digital video image of arbitrary claim in the claim 1 to 6, it is characterized in that the overshoot amplitude control of the current pixel of renewal during greater than the maximum Gmax of pixel in the neighborhood of current pixel comprises the steps:
Recomputate the current pixel P of renewal
XNew:P
XNew=P
XNew-(P
XNew-Gmax) * Ap/256; Wherein Ap is positive excess modulation control coefrficient, Ap ∈ [0,255].
8. as the sharpening method of the described digital video image of arbitrary claim in the claim 1 to 6, it is characterized in that the overshoot amplitude control of the current pixel of renewal during less than the minimum value Gmin of pixel in the neighborhood of current pixel comprises the steps:
Recomputate the current pixel P of renewal
XNew:P
XNew=P
XNew+ (Gmin-P
XNew) * An/256; Wherein An is negative overshoot amplitude control coefrficient, An ∈ [0,255].
9. the sharpening method of digital video image as claimed in claim 1 is characterized in that, before the added luminance with this sharpening increment Delta new and current pixel, also comprises the step of current pixel being carried out low-pass filtering.
10. the sharpening system of a digital video image comprises:
The sharpening filter, be used on the same image line of being positioned at of video image line with current pixel P
XFor N the pixel at center described current pixel is carried out obtaining the sharpening increment Delta after the sharpening filtering of N rank, wherein N is the odd number more than or equal to 3;
Positive sawtooth control unit that links to each other and positive overshoot width control unit in Δ 〉=0 o'clock, carry out obtaining new sharpening increment Delta new after positive sawtooth control and the positive overshoot width control;
The negative sawtooth control unit and the negative overshoot width control unit that link to each other in Δ＜0 o'clock, bear sawtooth control and are born and obtain new sharpening increment Delta new after the overshoot width is controlled;
Adder, the current pixel P that obtains upgrading after the added luminance with described sharpening increment Delta new and current pixel
XNew;
Overshoot amplitude control unit receives the current pixel P from the renewal of adder
XIf new is the current pixel P that upgrades
XNew is greater than the maximum Gmax of pixel in the neighborhood of current pixel, then by a positive excess modulation control register current pixel of this renewal carried out positive excess modulation control; If the current pixel P that upgrades
XNew then bears the control of overshoot amplitude by a negative overshoot amplitude control register to the current pixel of this renewal less than the minimum value Gmin of pixel in the neighborhood of current pixel; Otherwise the current pixel that upgrades remains unchanged.
11. the sharpening system of digital video image as claimed in claim 10 is characterized in that, described positive sawtooth control unit comprises:
The module of calculating variable hd: hd=max (| P
X-1-P
X+1|, | P
X-P
X+1|, | P
X-1-P
X|); P wherein
X-1Be current pixel P
XLast pixel, P
X+1Be current pixel P
XA back pixel;
Calculate the module of variable yp: yp=hd* (jaggyKp+1.0)/256, wherein jaggyKp is positive sawtooth control coefrficient, jaggyKp ∈ [0,255];
Judge module is if yp＞1.0 make yp=1.0.
12. the sharpening system of digital video image as claimed in claim 11 is characterized in that, described positive overshoot width control unit comprises:
Judge that yp is whether less than the module of (1-Xp/256)/2; With the assignment module;
If yp＜(1-Xp/256)/2, Δ new=Δ * (yp+Xp/256); Otherwise Δ new=Δ * (1-yp); Wherein Xp is positive overshoot width control coefrficient, Xp ∈ [0,255].
13. the sharpening system of digital video image as claimed in claim 10 is characterized in that, described negative sawtooth control unit comprises:
The module of calculating variable hd: hd=max (| P
X-1-P
X+1|, | P
X-P
X+1|, | P
X-1-P
X|); P wherein
X-1Be current pixel P
XLast pixel, P
X+1Be current pixel P
XA back pixel;
Calculate the module of variable yp: yp=hd* (jaggyKn+1.0)/256, wherein jaggyKn is negative sawtooth control coefrficient, jaggyKn ∈ [0,255];
Judge module is if yp＞1.0 make yp=1.0.
14. the sharpening system of digital video image as claimed in claim 13 is characterized in that, described negative overshoot width control unit comprises:
Judge that yp is whether less than the module of (1-Xn/256)/2; With the assignment module;
If yp＜(1-Xn/256)/2, Δ new=Δ * (yp+Xn/256); Otherwise Δ new=Δ * (1-yp); Wherein Xn is negative overshoot width control coefrficient, Xn ∈ [0,255].
15. the sharpening system of digital video image as claimed in claim 10 is characterized in that, described sharpening filter is the linear sharpening filters in N rank.
16. the sharpening system as the described digital video image of arbitrary claim in the claim 10 to 15 is characterized in that, the overshoot amplitude control unit of the current pixel of renewal during greater than the maximum Gmax of pixel in the neighborhood of current pixel comprises:
Recomputate the current pixel P of renewal
XThe module of new: P
XNew=P
XNew-(P
XNew-Gmax) * Ap/256; Wherein Ap is positive excess modulation control coefrficient, Ap ∈ [0,255].
17. the sharpening system as the described digital video image of arbitrary claim in the claim 10 to 15 is characterized in that, the overshoot amplitude control unit of the current pixel of renewal during less than the minimum value Gmin of pixel in the neighborhood of current pixel comprises:
Recomputate the current pixel P of renewal
XThe module of new: P
XNew=P
XNew+ (Gmin-P
XNew) * An/256; Wherein An is negative overshoot amplitude control coefrficient, An ∈ [0,255].
18. the sharpening system of digital video image as claimed in claim 10 is characterized in that, described system comprises that also one is positioned at described adder low pass filter before, and current pixel is carried out low-pass filtering.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010023153XA CN102137221A (en) | 2010-01-22 | 2010-01-22 | Method and system for sharpening digital video image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010023153XA CN102137221A (en) | 2010-01-22 | 2010-01-22 | Method and system for sharpening digital video image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102137221A true CN102137221A (en) | 2011-07-27 |
Family
ID=44296866
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201010023153XA Pending CN102137221A (en) | 2010-01-22 | 2010-01-22 | Method and system for sharpening digital video image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102137221A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102752480A (en) * | 2011-12-01 | 2012-10-24 | 新奥特（北京）视频技术有限公司 | Method for achieving sharpening special effects for image |
| CN107430762A (en) * | 2015-02-25 | 2017-12-01 | 高通股份有限公司 | Digital zooming method and system |
| CN109698892A (en) * | 2018-12-29 | 2019-04-30 | 深圳开阳电子股份有限公司 | A kind of video image sharpening method and image processing equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030161545A1 (en) * | 2002-02-27 | 2003-08-28 | Eastman Kodak Company | Method for sharpening a digital image with signal to noise estimation |
| US20040047514A1 (en) * | 2002-09-05 | 2004-03-11 | Eastman Kodak Company | Method for sharpening a digital image |
| CN1545328A (en) * | 2003-11-10 | 2004-11-10 | 华亚微电子（上海）有限公司 | System and method for promoting marginal definition of video image |
| CN1635779A (en) * | 2003-12-30 | 2005-07-06 | 华亚微电子（上海）有限公司 | Adjusting method for brightness edge in image and adjusting system therefor |
| US20080101717A1 (en) * | 2006-11-01 | 2008-05-01 | Quanta Computer Inc. | Image edge enhancing apparatus and method |
-
2010
- 2010-01-22 CN CN201010023153XA patent/CN102137221A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030161545A1 (en) * | 2002-02-27 | 2003-08-28 | Eastman Kodak Company | Method for sharpening a digital image with signal to noise estimation |
| US20040047514A1 (en) * | 2002-09-05 | 2004-03-11 | Eastman Kodak Company | Method for sharpening a digital image |
| CN1545328A (en) * | 2003-11-10 | 2004-11-10 | 华亚微电子（上海）有限公司 | System and method for promoting marginal definition of video image |
| CN1635779A (en) * | 2003-12-30 | 2005-07-06 | 华亚微电子（上海）有限公司 | Adjusting method for brightness edge in image and adjusting system therefor |
| US20080101717A1 (en) * | 2006-11-01 | 2008-05-01 | Quanta Computer Inc. | Image edge enhancing apparatus and method |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102752480A (en) * | 2011-12-01 | 2012-10-24 | 新奥特（北京）视频技术有限公司 | Method for achieving sharpening special effects for image |
| CN107430762A (en) * | 2015-02-25 | 2017-12-01 | 高通股份有限公司 | Digital zooming method and system |
| CN109698892A (en) * | 2018-12-29 | 2019-04-30 | 深圳开阳电子股份有限公司 | A kind of video image sharpening method and image processing equipment |
| CN109698892B (en) * | 2018-12-29 | 2021-06-11 | 深圳开阳电子股份有限公司 | Video image sharpening method and image processing equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110148095B (en) | Underwater image enhancement method and enhancement device | |
| CN103514583B (en) | Image sharpening method and equipment | |
| CN108229279A (en) | Face image processing process, device and electronic equipment | |
| CN104076928B (en) | A kind of method for adjusting text importing image | |
| CN105046658B (en) | A kind of low-light (level) image processing method and device | |
| CN103617638A (en) | Image processing method and device | |
| CN104935902B (en) | Image color Enhancement Method, device and electronic equipment | |
| CN109345490B (en) | Method and system for enhancing real-time video image quality of mobile playing terminal | |
| CN108596828B (en) | Image floodlight processing method and device, electronic equipment and storage medium | |
| CN102082954A (en) | Image processing apparatus, image processing method and program | |
| CN109767408B (en) | Image processing method, image processing device, storage medium and computer equipment | |
| EP1368960A2 (en) | Digital image appearance enhancement and compressibility improvement method and system | |
| CN108665408A (en) | Method for regulating skin color, device and electronic equipment | |
| CN112200736B (en) | Image processing method based on reinforcement learning and model training method and device | |
| US9934560B2 (en) | User sliders for simplified adjustment of images | |
| US8488899B2 (en) | Image processing apparatus, method and recording medium | |
| US11727543B2 (en) | Systems and methods for content-aware enhancement of images | |
| CN107610675A (en) | A kind of image processing method and device based on dynamic level | |
| CN104091307A (en) | Frog day image rapid restoration method based on feedback mean value filtering | |
| CN102137221A (en) | Method and system for sharpening digital video image | |
| US9305338B1 (en) | Image detail enhancement and edge sharpening without overshooting | |
| CN111583103A (en) | Face image processing method and device, electronic equipment and computer storage medium | |
| KR20030076452A (en) | Image processing apparatus and image outputting apparatus | |
| KR101514152B1 (en) | Method and apparatus for improving image quality using singular value decomposition | |
| CN110895789A (en) | Face beautifying method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C05 | Deemed withdrawal (patent law before 1993) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20110727 |