CN102196232A - Method for restoring missing lines in interlaced video image - Google Patents
Method for restoring missing lines in interlaced video image Download PDFInfo
- Publication number
- CN102196232A CN102196232A CN2010101206484A CN201010120648A CN102196232A CN 102196232 A CN102196232 A CN 102196232A CN 2010101206484 A CN2010101206484 A CN 2010101206484A CN 201010120648 A CN201010120648 A CN 201010120648A CN 102196232 A CN102196232 A CN 102196232A
- Authority
- CN
- China
- Prior art keywords
- pixel
- optimum match
- value
- row
- low
- 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
- 238000000034 method Methods 0.000 title claims abstract description 51
- 238000001514 detection method Methods 0.000 claims abstract description 84
- 230000008034 disappearance Effects 0.000 claims description 84
- 230000015654 memory Effects 0.000 claims description 16
- 238000011084 recovery Methods 0.000 claims description 5
- 239000004744 fabric Substances 0.000 claims description 4
- 239000000203 mixture Substances 0.000 claims description 4
- GNFTZDOKVXKIBK-UHFFFAOYSA-N 3-(2-methoxyethoxy)benzohydrazide Chemical compound COCCOC1=CC=CC(C(=O)NN)=C1 GNFTZDOKVXKIBK-UHFFFAOYSA-N 0.000 claims description 2
- FGUUSXIOTUKUDN-IBGZPJMESA-N C1(=CC=CC=C1)N1C2=C(NC([C@H](C1)NC=1OC(=NN=1)C1=CC=CC=C1)=O)C=CC=C2 Chemical compound C1(=CC=CC=C1)N1C2=C(NC([C@H](C1)NC=1OC(=NN=1)C1=CC=CC=C1)=O)C=CC=C2 FGUUSXIOTUKUDN-IBGZPJMESA-N 0.000 claims description 2
- 230000000750 progressive effect Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 5
- 230000008878 coupling Effects 0.000 description 4
- 238000010168 coupling process Methods 0.000 description 4
- 238000005859 coupling reaction Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 231100000691 up-and-down procedure Toxicity 0.000 description 1
Images
Abstract
The invention discloses a method for restoring missing lines in an interlaced video image. The method comprises the following steps: setting a small-angle interpolation window; setting a matching detection window; calculating an optimum matching detection value; acquiring a corresponding pixel H in the former missing line along a direction corresponding to the optimum matching detection value; judging whether the absolute value of difference of the optimum matching direction information and the optimum matching direction information of the corresponding pixel H in the former missing line is less than or equal to a threshold set by a user; and if so, calculating and restoring the current lack pixel by using the interpolated value in the optimum matching direction. According to the method, the effectiveness of the matching detection of the current missing line is judged by using the matching detection result of the former missing line, thereby greatly reducing the erroneous judgment of the matching detection and improving the accuracy of the matching detection, and further applying a small-angle interpolation method to the restoration of the missing lines in the interlaced video image successfully so as to achieve more accurate restoration of the missing lines in the image.
Description
The technology neighborhood
The invention belongs to the digital video signal processing neighborhood, specifically, relate to the restoration methods of the disappearance row in a kind of interlaced video image.
Background technology
For the interlaced digital video signal, traditional Motion Adaptive de-interlacing method normally: use the information of 3D to carry out interpolation for static image, and use the information of 2D to carry out interpolation for the image of motion.Here, the meaning of 2D is meant the information in the interlaced picture, an internal information on the spot.The problem of discussion of the present invention be exactly how the method by the 2D field interpolation produce disappearance row in the interlaced video image.
As shown in Figure 1, for an interlaced picture, wherein have only or half image information very capable or the idol row, promptly strange field picture only contains the odd-numbered scan lines of image, and even field picture only contains the even-line interlace row of image.Strange replaces transmission with even field picture and has just formed usually said interlaced video image stream.In traditional Motion Adaptive de-interlacing method, when picture material changes, we will use the method for 2D field interpolation to produce disappearance row in the interlaced video image, this is also referred to as 2D usually and separates the interlacing operation, promptly for the strange field picture of a width of cloth, utilize the information in this field picture to produce the even number line pixel of disappearance, and, then utilize the interior information of this field picture to produce the odd-numbered line pixel of disappearance for width of cloth idol field picture.
As shown in Figure 1, under normal conditions, the operation that 2D separates interlacing is: concerning strange field picture, utilize strange row 1 and strange row 3 to come interpolation to produce the idol row a of disappearance, utilize strange row 3 and strange row 5 to come interpolation to produce the idol row b of disappearance, or the like; The dual field image utilizes even row 2 and even row 4 to come interpolation to produce the strange capable c of disappearance, utilizes even row 4 and even row 6 to come interpolation to produce the strange capable d that lacks, or the like.For image topmost and nethermost disappearance row, we can simply carry out special processing, as repeat to lack capable next line or lastrow represents that disappearance is capable.This special processing is not watched so can not influence owing to occur in the uppermost edge of image and edge bottom.
It is to utilize the lastrow of disappearance row and the pixel average of next line to be used as lacking capable pixel value that modal 2D separates the interlacing operation, and for example, the disappearance idol row a among Fig. 1 can obtain like this:
A=(the very capable strange row 3 of 1+)/2
That is to say that the value of a pixel among the disappearance row a equals the strange mean value of the pixel value in the row 3 in pixel value and this pixel below in this strange row 1 in pixel top.The 2D that the straightforward procedure of this " average up and down " can be handled big portion separates the interlacing problem, but unable to do what one wishes for the situation at " low-angle edge ".
Can be clearly seen that from Fig. 2 when having low-angle edge in the image, adopt average up and down method of simple interpolation can not recover original progressive image, its result can make the low-angle edge in the former progressive image present stepped.In this case, if can detect the trend at low-angle edge and carry out interpolation along the direction at edge, " ● " pixel of disappearance row is that direction along the edge is obtained by " ■ " picture element interpolation as shown in FIG., and the 2D result that separates interlacing just can recover the edge of former progressive image fully so.The trend at low-angle edge is normally determined by the technology of window coupling.In the drawings, we see two detection windows that can find coupling in the lastrow of disappearance row and next line, and these two position of window have just determined the trend at edge, and then have also just determined corresponding " ● " locations of pixels of disappearance row.Be envisioned that at the more little edge of angle, the matching detection window just needs wide more, also just expend resources of chip more.In addition, owing to always have noise in the video image,, and be meant on the whole pixel " the most close " in two detection windows so here " coupling " is not meant in two matching detection windows that the pixel of correspondence is all identical.
The key of low-angle interpolation method is accurately to detect the trend at low-angle edge.Because interlaced picture only contains half information of former progressive image, can become very difficult so accurately judge the trend at low-angle edge under many circumstances, and the situation of erroneous judgement can often take place also.Figure 3 shows that example, former progressive image has shown four English alphabets " BETH ".If the spaced rows between them is the delegation's disappearance row in the interlaced picture as shown in the figure, then we will utilize the lastrow and the next line of disappearance row to carry out matching detection and low-angle interpolation.From lacking the pixel distribution of capable lastrow and next line, matching detection window 3 and matching detection window 4 be than the matching detection window on other direction all " close ", and the presentation of results that is to say matching detection exists a low-angle edge along the direction of matching detection window 3 and matching detection window 4.Yet we can know the pixel that does not have any dark color in the corresponding disappearance row in former progressive image, also just do not have a low-angle edge, so top matching detection result is wrong in fact.If the matching detection result according to such mistake carries out low-angle interpolation, some dark pixels that will occur should not existing in the disappearance row that obtains so.
The situation that causes the matching detection erroneous judgement easily among similar Fig. 3 can often run in practice.So can method that utilize the low-angle interpolation to realize that 2D separates interlacing depend on the occurrence probability that reduce erroneous judgement greatly.
Summary of the invention
The object of the present invention is to provide the restoration methods of the disappearance row in a kind of interlaced video image, thereby cause the matching detection erroneous judgement can't the most closely recover the technical problem of original progressive image easily to solve present low-angle interpolation method.
In order to achieve the above object, technical scheme of the present invention is as follows:
The restoration methods of the disappearance row in a kind of interlaced video image, comprise the steps: a) to set the step of three low-angle interpolation windows, comprise with current missing pixel being the low-angle interpolation window of 2N+1 the pixel composition at center, with this current missing pixel top pixel is the top low-angle interpolation window of 2N+1 the pixel composition at center, with the below low-angle interpolation window that 2N+1 the pixel that is the center with this current missing pixel lower pixel formed, wherein N is a positive integer; B) be that 2N+1 the pixel at center and 2N+1 pixel that current missing pixel lower pixel is the center are the step of 4N+2 matching detection window of center setting with this current missing pixel top pixel respectively, all matching detection windows are made up of the odd number of pixels that is no less than 2N+3, and wherein N is a positive integer; C) with pass current missing pixel and connect described top low-angle interpolation window respectively and below low-angle interpolation window in the pairing direction of 2N+1 bar straight line of pixel be matching direction calculate 2N+1 matching detection value respectively and with wherein minimum matching detection value as the optimum Match detected value; D) determine and the step of storage optimum Match directional information; E) obtain corresponding pixel H in the last disappearance row along the pairing direction of this optimum Match detected value; F) judge that whether the absolute value of difference of optimum Match directional information of pixel H of the correspondence in this an optimum Match directional information and the last disappearance row is smaller or equal to user's preset threshold; If then adopt the interpolation calculation on this optimum Match direction to recover this current missing pixel; If not, then up and down average computation recover this current missing pixel; If because of the width of matching detection window can't form the current missing pixel that the matching detection window in this step is handled inadequately, then the optimum Match direction is designated as vertical direction and interpolation calculation and recovers this current missing pixel; G) a) handling current missing pixel to step f) according to above-mentioned steps disposes up to current disappearance row with the next pixel in the delegation; H) a) handle next according to above-mentioned steps and lack the step of going to step g).
Adopt said method, judge the matching detection validity of current disappearance row by the matching detection result who utilizes a last disappearance row, can greatly reduce the generation of matching detection erroneous judgement situation, improve the accuracy of matching detection, thereby make the low-angle interpolation method can be applied to smoothly in the recovery of the disappearance row in the interlaced video image, the recovery of the disappearance row of image can be more accurate at this moment.
Description of drawings
Fig. 1 is the strange field picture of interlaced video stream and the schematic diagram of even field picture;
Fig. 2 is the comparison diagram of " average up and down " interpolation and low-angle interpolation method;
Fig. 3 is the schematic diagram of matching detection erroneous judgement behind the employing low-angle interpolation method;
Fig. 4 is the low-angle interpolation window schematic diagram in the low-angle interpolation method of the present invention;
Fig. 5 is the validity check schematic diagram of the matching detection result in the low-angle interpolation method of the present invention;
Fig. 6 is the process chart of the restoration methods of the disappearance row in the interlaced video image of the present invention.
Embodiment
According to Fig. 4 to Fig. 6, provide preferred embodiment of the present invention, and described in detail below, enable to understand better function of the present invention, characteristics.
Low-angle interpolation method among the present invention and system have utilized the matching detection result of a last disappearance row to check the matching detection result of current disappearance row, and judge whether the matching detection result of current disappearance row is effective.Carry out low-angle interpolation to being judged as effective matching detection result, and carry out traditional processing of the method for average up and down for being judged as invalid matching detection result.
The first step is a matching detection.At first we will limit the minimum value of angle in the low-angle interpolation operation, and this is to realize by the width of regulation low-angle interpolation window.As shown in Figure 4, low-angle interpolation window is a window that width is a 2N+1 pixel.Respectively defining a low-angle interpolation window in the lastrow of disappearance row and the next line, is the center with top pixel " A " and the lower pixel " B " that lacks current interpolating pixel " X " in the row respectively.Low-angle interpolation window in the lastrow of disappearance row is designated as pup, and wherein the pixel of 2N+1 from left to right is designated as pup (1), pup (2) ..., pup (2N+1), and wherein center pixel A is pup (N+1).Similarly, the low-angle interpolation window in the next line of disappearance row is designated as pdn, and wherein the pixel of 2N+1 from left to right is designated as pdn (1), pdn (2) ..., pdn (2N+1), and wherein center pixel B is pdn (N+1).When pixel p up (1) is matched pixel p dn (2N+1), or when pixel p dn (1) was matched pixel p up (2N+1), the angle of coupling trend reached minimum, and this minimum angles can be calculated like this:
θmin＝arctan[3/(2N+1)]
For instance, if the width of low-angle interpolation window is 9 pixels, so treatable minimum interpolation angle should be
θmin＝arctan[3/9]≈18.435degrees
That is to say,, so just choose width and be 9 low-angle interpolation window if the user wants the minimum angles of low-angle interpolation correspondence to be about 18～19 degree.In practice, along with the angle of low-angle edge correspondence less than θ min and reduce gradually, the effect of low-angle interpolation is variation gradually also.Obviously, more little θ min also means big more low-angle interpolation window, thereby needs the resource of many more Digital Video Processing chips.
Behind the width of having determined low-angle interpolation window, the width M that we define the matching detection window is the odd number that is not less than 2N+3.In matching detection, we have the individual matching detection window of 2* (2N+1), respectively to lack the pup (1) in the capable lastrow, pup (2) ..., the pdn (1) in the next line of pup (2N+1) and disappearance row, pdn (2) ..., pdn (2N+1) is a window center.
May improve the accuracy of matching detection to a certain extent though be noted that big more matching detection window here, thereby also can cause more computing to increase the cost of chip simultaneously.And very big matching detection window might comprise that too much this is because very big matching detection window can comprise the pixel far apart from the interpolating pixel position with the very weak information of interpolating pixel correlation----, so the accuracy of detection also can be affected.So in the design of the low-angle interpolation circuit of reality, it is excessive that the width of matching detection window should not be chosen, recommendation of the present invention is M=2N+3.Because the width of matching detection window is M, so beginning (M+1)/2 pixel and ending (M+1)/2 pixel of delegation's disappearance row can can't be handled inadequately because of window width, for near the pixel these image boundaries, we can simply stipulate to carry out average treatment, i.e. X=(A+B)/2 up and down.Also can be understood as, if, then the optimum Match direction is designated as vertical direction because of the width of matching detection window can't form the current missing pixel that the matching detection window in this step is handled inadequately herein.
Narration for convenience, M the pixel that we further define in the matching detection window is:
| Matching detection window center pixel | The from left to right continuous N pixel that the matching detection window is contained |
| pup(1) | MWup_1 (1), MWup_1 (2) ..., MWup_1 (M) wherein MWup_1 ((M+1)/2) is pup (1) |
| pup(2) | MWup_2 (1), MWup_2 (2) ..., MWup_2 (M) wherein MWup_2 ((M+1)/2) is pup (2) |
| ... | ... |
| pup(2N+1) | MWup_2N+1 (1), MWup_2N+1 (2) ..., MWup_2N+1 (M) wherein MWup_2N+1 ((M+1)/2) is pup (2N+1) |
| pdn(1) | MWdn_1 (1), MWdn_1 (2) ..., MWdn_1 (M) wherein MWdn_1 ((M+1)/2) is pdn (1) |
| pdn(2) | MWdn_2 (1), MWdn_2 (2) ..., MWdn_2 (M) wherein MWdn_2 ((M+1)/2) is pdn (2) |
| ... | ... |
| pdn(2N+1) | MWdn_2N+1 (1), MWdn_2N+1 (2) ..., MWdn_2N+1 (M) wherein MWdn_2N+1 ((M+1)/2) is pdn (2N+1) |
In matching detection, we do following hypothesis:
The condition that current interpolating pixel X in the disappearance row need carry out the low-angle interpolation is
1. there is a low-angle edge in low-angle interpolation window pup (the 1)～pup (2N+1) in the lastrow of disappearance row between low-angle interpolation window pdn (the 1)～pdn (2N+1) that lacks in the next line of going;
2. this low-angle edge passes the current interpolating pixel X in the disappearance row;
3. this low-angle edge is straight line at pup (1)～pup (2N+1) to the part between pdn (1)～pdn (2N+1).
Like this, what we will calculate is following 2N+1 matching detection value, and these matching detection values are 2N+1 values that the brightness value by pixel in the corresponding matching detection window as above calculates:
| The matching detection value | Matching direction | Matching detection value computing formula |
| D(1) | pup(1)～pdn(2N+1) | M D(1)＝∑|MWup_1(i)-MWdn_2N+1(i)| i＝1 |
| D(2) | pup(2)～pdn(2N) | M D(2)＝∑|MWup_2(i)-MWdn_2N(i)| i＝1 |
| ... | ... | ... |
| D(2N+1) | pup(2N+1)～pdn(1) | M D(2N+1)＝∑|MWup_2N+1(i)-MWdn_1(i)| i＝1 |
From the above 2N+1 that the obtains matching detection value, we select the numerical value minimum:
Dmin＝min[D(1)，D(2)，...，D(2N+1)]
And what think the Dmin representative is exactly the optimum matching point that obtains in the matching detection, and its corresponding matching direction is exactly the low-angle edge direction that we will carry out the low-angle interpolation.
Second step was matching detection result's a validity check.If upcheck, we will lack the interpolation arithmetic of current interpolating pixel X in the row along the pairing matching direction of Dmin, if i.e. Dmin=D (k), wherein k ∈ 1,2 ..., 2N+1}, so
X＝[pup(k)+pdn(2N+2-k)]/2
If check can't pass, our pixel value that will use traditional method of average up and down to calculate X gets so, promptly
X＝(A+B)/2
Matching detection result's validity check is to utilize the last one matching detection result who lacks row to check the matching detection result of current disappearance row.As shown in Figure 5, if Dmin=D (k), we know that the direction of optimum Match is to match pdn (2N+2-k) from pup (k) by top narration so, and line therebetween also necessarily passes the current interpolating pixel X in the current disappearance row.Pixel p dn (2N+2-k) is prolonged the position that intersects at pixel H with a last disappearance row to the line of pixel p up (k), and its optimum Match direction optimum Match direction corresponding with current interpolating pixel X that H is calculated during as current pixel compares.
Here we will make a hypothesis: the curvature by the edge in the image of low-angle interpolation processing can not be too big, and promptly the degree of crook at edge is less.For the bigger place of ratio of curvature on the edge, matching detection result's validity check can not pass through, and the interpolation of curvature larger part will be used traditional method of average up and down on the edge.Based on this hypothesis, we can think that the trend of trend when passing current disappearance row in same edge when row disappearance on passing is more or less the same.Thus, be designated as k_H if we will go up the optimum Match direction at H point place in the disappearance row, we think that the numerical value of k_H and k should be more or less the same so if edge traverses to the X point of current disappearance row from the H point of a last disappearance row:
| k-k_H|≤APT, wherein APT 〉=0 is integer numerical value for curvature is provided with register, and its default value is 1, is set up on their own by the user, and big more APT represents from a last disappearance row big more to the permission curvature at the existing low-angle of current disappearance edge in the ranks.Since the span of k be k ∈ 1,2 ..., 2N+1}, so | the span of k-k_H| is to 2N, so the setting of APT also should be between 0 to 2N from 0.
As above Biao Shi condition is the validity check condition of the matching detection result among the present invention.If this condition is set up, then matching detection result's validity check passes through, we can utilize the result [pup (k)+pdn (2N+2-k)]/2 of low-angle interpolation to determine the pixel value of current interpolating pixel X in the current disappearance row, otherwise matching detection result's validity check does not pass through, the pixel value of method of average decision X got about we will utilize, i.e. X=(A+B)/2.
From top method introduction as can be seen, this method requires the matching detection result with each disappearance row low-angle interpolation, and promptly the information stores of optimum Match direction is got up, to be used by the matching detection result's of next disappearance row validity check.The optimum Match directional information is the numerical value of k.
Because the direction of optimum Match is to match pdn (2N+2-k) from pup (k), obviously also can unify 2N+2-k is made as the optimum Match directional information, carries out similar validity check then.Promptly in whole restoration methods, or the optimum Match directional information all is made as the numerical value of k, or the optimum Match directional information all is made as the numerical value of 2N+2-k, both are that exclusiveness is selected.
The span of k is 1～2N+1, when using 2N+2-K to store the optimum Match directional information, because the value of k is 1～2N+1, so the scope of 2N+2-k also is 1～2N+1, so the bit wide that needs is the same, no matter so any in above-mentioned two kinds of situations, the bit wide of optimum Match direction memory should be more than or equal to log2 (2N+1).For example, if the width of low-angle interpolation window is 2N+1=9, the bit wide of its required optimum Match direction memory should be more than or equal to log2 (9)=3.17, so bit wide is at least 4bit so.When carrying out the processing of current disappearance row, we will read the optimum Match directional information of a last disappearance row from optimum Match direction memory, and the optimum Match directional information of current disappearance row itself also will store and thinks that next disappearance exercises usefulness.For this reason, we can prepare two optimum Match direction memories and use by turns, also can utilize a fritter temporary storage to realize the read-write and the overlapping operation of new and old optimum Match directional information as buffer.Using by turns of memory is well-known technology in the design of Digital Video Processing chip, and we do not add detailed description at this.Before the low-angle interpolation processing of each width of cloth interlaced picture began, all optimum Match direction memories were wanted earlier zero clearing.
What will illustrate in addition a bit is, the low-angle interpolation method among the present invention only is applied to the brightness value of image pixel, and this is because human eye is to many than color edge sensitivity of the flatness of luminance edges.But the present invention does not limit the separating in the interlacing of chrominance signal that the low-angle interpolation method among the present invention is applied to image.
Fig. 6 is the process chart of the restoration methods of the disappearance row in the interlaced video image of the present invention.As shown in the figure, this restoration methods comprises the steps:
The input interlaced video flows in the row cache memory;
Judge whether to be new interlaced video field? if, then with the zero clearing of optimum Match directional information memory; Then read the picturedeep certificate if not from the row cache memory, set low-angle interpolation window width according to mode recited above, matching detection window width and curvature are provided with the value of register, matching detection produces the matching detection value and determines the optimum Match detected value, generation optimum Match detected value pairing optimum Match directional information k also stores in the described optimum Match directional information memory, then from optimum Match direction memory, read the optimum Match directional information k_H of a last disappearance row mid point H, judge | k-k_H|≤APT? recover this current missing pixel if then adopt the interpolation calculation on this optimum Match direction; If not, then the optimum Match direction is designated as vertical direction and this current missing pixel of calculating recovery; If because of the width of matching detection window can't form the current missing pixel that the matching detection window in this step is handled inadequately, then the optimum Match direction is designated as vertical direction and calculates this current missing pixel of recovery;
Do you judge that current disappearance row disposes? if not, then continue to handle next pixel in the current disappearance row, if then handle next disappearance row.
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.For example, the optimum Match directional information is the slope value of optimum Match direction, during the interlacing that the low-angle interpolation method among the present invention also can be used for the chrominance signal of image recovers.Should be appreciated that these modifications or conversion do not break away from protection scope of the present invention.
Claims (12)
1. the restoration methods of the disappearance row in the interlaced video image comprises the steps:
A) step of three low-angle interpolation windows of setting, comprise with current missing pixel being the low-angle interpolation window of 2N+1 the pixel composition at center, with this current missing pixel top pixel is the top low-angle interpolation window of 2N+1 the pixel composition at center, with the below low-angle interpolation window that 2N+1 the pixel that is the center with this current missing pixel lower pixel formed, wherein N is a positive integer;
B) be that 2N+1 the pixel at center and 2N+1 pixel that current missing pixel lower pixel is the center are the step of 4N+2 matching detection window of center setting with this current missing pixel top pixel respectively, all matching detection windows are made up of the odd number of pixels that is no less than 2N+3, and wherein N is a positive integer;
C) with pass current missing pixel and connect described top low-angle interpolation window respectively and below low-angle interpolation window in the pairing direction of 2N+1 bar straight line of pixel be matching direction calculate 2N+1 matching detection value respectively and with wherein minimum matching detection value as the optimum Match detected value;
D) determine and the step of storage optimum Match directional information;
E) obtain corresponding pixel H in the last disappearance row along the pairing direction of this optimum Match detected value;
F) judge that whether the absolute value of difference of optimum Match directional information of pixel H of the correspondence in this an optimum Match directional information and the last disappearance row is smaller or equal to user's preset threshold; If then adopt the interpolation calculation on this optimum Match direction to recover this current missing pixel; If not, then up and down average computation recover this current missing pixel; If because of the width of matching detection window can't form the current missing pixel that the matching detection window in this step is handled inadequately, then the optimum Match direction is designated as vertical direction and interpolation calculation and recovers this current missing pixel;
G) a) handling current missing pixel to step f) according to above-mentioned steps disposes up to current disappearance row with the next pixel in the delegation;
H) a) handle next according to above-mentioned steps and lack the step of going to step g).
2. the restoration methods of the disappearance row in the interlaced video image as claimed in claim 1, it is characterized in that, the step of described definition optimum Match directional information is specially: if the pixel in the pairing direction of the optimum Match detected value top low-angle interpolation window pointed is a k pixel in this top low-angle interpolation window, then defining the optimum Match directional information is k; Described user's preset threshold is that curvature is provided with the value that the user sets among the register APT, and this value is in 0 to 2N.
3. the restoration methods of the disappearance row in the interlaced video image as claimed in claim 1, it is characterized in that, the step of described definition optimum Match directional information is specially: if the pixel in the pairing direction of the optimum Match detected value below low-angle interpolation window pointed is the individual pixel of k ' in this below low-angle interpolation window, then defining the optimum Match directional information is k '; Described user's preset threshold is that curvature is provided with the value that the user sets among the register APT, and this value is in 0 to 2N.
4. as the restoration methods of the disappearance row in the described interlaced video image of arbitrary claim in the claim 1 to 3, it is characterized in that, the concrete grammar of described interpolation calculation is, the pixel value of described current missing pixel is the mean value of the pixel value sum of pixel in the pixel value of pixel and the below low-angle interpolation window in the low-angle interpolation window of determined top on the optimum Match direction.
5. the restoration methods of the disappearance row in the interlaced video image as claimed in claim 4, it is characterized in that, the optimum Match directional information of a missing pixel in the described corresponding disappearance row is stored in the optimum Match direction memory, and the bit wide of the memory cell of corresponding each missing pixel of this optimum Match direction memory is more than or equal to log2 (2N+1).
6. the restoration methods of the disappearance row in the interlaced video image as claimed in claim 5, it is characterized in that, described optimum Match direction memory is two, is used for storing in turn the optimum Match directional information of current disappearance row and the optimum Match directional information of a last disappearance row.
7. the restoration methods of the disappearance row in the interlaced video image as claimed in claim 5, it is characterized in that, utilize a fritter temporary storage as buffer realize current disappearance row the optimum Match directional information and the read-write and the overlapping operation of optimum Match directional information of a last disappearance row.
8. the restoration methods of the disappearance row in the interlaced video image as claimed in claim 5 is characterized in that, before the recovery of each width of cloth interlaced picture is handled, and all optimum Match direction memory elder generation zero clearings.
9. the restoration methods of the disappearance row in the interlaced video image as claimed in claim 4 is characterized in that described pixel value is a brightness value.
10. the restoration methods of the disappearance row in the interlaced video image as claimed in claim 1 is characterized in that, the step of described definition optimum Match directional information is specially: the optimum Match directional information is the slope value of optimum Match direction.
11. the restoration methods of the disappearance row in the interlaced video image as claimed in claim 10, it is characterized in that, the concrete grammar of described calculating is, the pixel value of described current missing pixel is the mean value of the pixel value sum of pixel in the pixel value of pixel and the below low-angle interpolation window in the low-angle interpolation window of determined top on the optimum Match direction.
12. the restoration methods of the disappearance row in the interlaced video image as claimed in claim 11 is characterized in that described pixel value is a brightness value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010120648.4A CN102196232B (en) | 2010-03-09 | 2010-03-09 | Method for restoring missing lines in interlaced video image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010120648.4A CN102196232B (en) | 2010-03-09 | 2010-03-09 | Method for restoring missing lines in interlaced video image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102196232A true CN102196232A (en) | 2011-09-21 |
| CN102196232B CN102196232B (en) | 2015-01-28 |
Family
ID=44603526
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201010120648.4A Expired - Fee Related CN102196232B (en) | 2010-03-09 | 2010-03-09 | Method for restoring missing lines in interlaced video image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102196232B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117095134A (en) * | 2023-10-18 | 2023-11-21 | 中科星图深海科技有限公司 | Three-dimensional marine environment data interpolation processing method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1512774A (en) * | 2002-12-13 | 2004-07-14 | ���ܿ���ϵͳ����˾ | High grade edge self adaption interpolation method and system for interleave to inching change |
| CN1529500A (en) * | 2003-10-16 | 2004-09-15 | 上海交通大学 | Three-dimensional video format conversion method based on motion adaption and marginal protection |
| CN101014086A (en) * | 2007-01-31 | 2007-08-08 | 天津大学 | De-interlacing apparatus using motion detection and adaptive weighted filter |
| CN101437137A (en) * | 2008-12-19 | 2009-05-20 | 四川虹微技术有限公司 | Field interpolation method |
| US20090256957A1 (en) * | 2008-04-11 | 2009-10-15 | Canon Kabushiki Kaisha | Moving image processing apparatus, control method thereof, and program |
-
2010
- 2010-03-09 CN CN201010120648.4A patent/CN102196232B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1512774A (en) * | 2002-12-13 | 2004-07-14 | ���ܿ���ϵͳ����˾ | High grade edge self adaption interpolation method and system for interleave to inching change |
| CN1529500A (en) * | 2003-10-16 | 2004-09-15 | 上海交通大学 | Three-dimensional video format conversion method based on motion adaption and marginal protection |
| CN101014086A (en) * | 2007-01-31 | 2007-08-08 | 天津大学 | De-interlacing apparatus using motion detection and adaptive weighted filter |
| US20090256957A1 (en) * | 2008-04-11 | 2009-10-15 | Canon Kabushiki Kaisha | Moving image processing apparatus, control method thereof, and program |
| CN101437137A (en) * | 2008-12-19 | 2009-05-20 | 四川虹微技术有限公司 | Field interpolation method |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117095134A (en) * | 2023-10-18 | 2023-11-21 | 中科星图深海科技有限公司 | Three-dimensional marine environment data interpolation processing method |
| CN117095134B (en) * | 2023-10-18 | 2023-12-22 | 中科星图深海科技有限公司 | Three-dimensional marine environment data interpolation processing method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102196232B (en) | 2015-01-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8199252B2 (en) | Image-processing method and device | |
| US6810156B1 (en) | Image interpolation device | |
| WO2010021965A1 (en) | Signature derivation for images | |
| US7667776B2 (en) | Video display device, video encoder, noise level estimation module and methods for use therewith | |
| US20040130619A1 (en) | Method for 3:2 pull-down film source detection | |
| US8081256B2 (en) | Method and system for edge directed deinterlacing in video image processing | |
| CN103856781A (en) | Self-adaptation threshold value video streaming multi-texture-direction error concealment method | |
| CN103119939A (en) | Techniques for identifying block artifacts | |
| US8538070B2 (en) | Motion detecting method and apparatus thereof | |
| CN102509311B (en) | Motion detection method and device | |
| CN1224274C (en) | Error concealment method and device | |
| US7447383B2 (en) | Directional interpolation method using frequency information and related device | |
| US7548663B2 (en) | Intra-field interpolation method and apparatus | |
| EP1862970B1 (en) | Image motion vector detecting device | |
| CN102196232A (en) | Method for restoring missing lines in interlaced video image | |
| CN102395019B (en) | Deinterlacing method | |
| US7233363B2 (en) | De-interlacing method, apparatus, video decoder and reproducing apparatus thereof | |
| US20050094877A1 (en) | Method and apparatus for detecting the location and luminance transition range of slant image edges | |
| CN104620279A (en) | Method and apparatus to detect artificial edges in images | |
| CN105611214A (en) | Method for de-interlacing through intra-field linear interpolation based on multidirectional detection | |
| WO2014025295A1 (en) | 2d/3d image format detection | |
| JP2002330354A (en) | Device, method and program for detecting and correcting defective pixel, and video signal processor | |
| US20100034520A1 (en) | Apparatus and method for analyzing pictures for video compression with content-adaptive resolution | |
| CN100583960C (en) | Deciding method and system for video de-interleaving solution strategy | |
| CN103763501A (en) | Self-adaptive video de-interlacing algorithm and device thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C56 | Change in the name or address of the patentee |
Owner name: FUJITSU SEMICONDUCTOR PACIFIC ASIA LTD. Free format text: FORMER NAME: FUJITSU MICROELECTRONICS (SHANGHAI) CO., LTD. |
|
| CP01 | Change in the name or title of a patent holder |
Address after: 200002, room 3102, the Bund center, 222 East Yanan Road, Shanghai, Huangpu District Patentee after: Fujitsu semiconductor (Shanghai) Co., Ltd. Address before: 200002, room 3102, the Bund center, 222 East Yanan Road, Shanghai, Huangpu District Patentee before: Fujitsu Microelectronics (Shanghai) Co.,Ltd. |
|
| C56 | Change in the name or address of the patentee | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 200002, room 3102, the Bund center, 222 East Yanan Road, Shanghai, Huangpu District Patentee after: Fujitsu electronic component (Shanghai) Co., Ltd. Address before: 200002, room 3102, the Bund center, 222 East Yanan Road, Shanghai, Huangpu District Patentee before: Fujitsu semiconductor (Shanghai) Co., Ltd. |
|
| C56 | Change in the name or address of the patentee | ||
| CP02 | Change in the address of a patent holder |
Address after: 201204 Shanghai Pudong New Area City 1155 Fangdian road Pudong Kerry 30 layer Patentee after: Fujitsu electronic component (Shanghai) Co., Ltd. Address before: 200002, room 3102, the Bund center, 222 East Yanan Road, Shanghai, Huangpu District Patentee before: Fujitsu electronic component (Shanghai) Co., Ltd. |
|
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20150128 Termination date: 20170309 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |