CN102984435A - Image processing device and image processing method and image broadcast system - Google Patents
Image processing device and image processing method and image broadcast system Download PDFInfo
- Publication number
- CN102984435A CN102984435A CN2011102706684A CN201110270668A CN102984435A CN 102984435 A CN102984435 A CN 102984435A CN 2011102706684 A CN2011102706684 A CN 2011102706684A CN 201110270668 A CN201110270668 A CN 201110270668A CN 102984435 A CN102984435 A CN 102984435A
- Authority
- CN
- China
- Prior art keywords
- image
- block
- intermediate imagery
- picture
- interpolation
- 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
- Television Systems (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention provides an image processing device, an image processing method and an image broadcast system. The image processing device is used for generating a middle picture including a plurality of middle image blocks according to a front picture and a back picture. Each middle image block corresponds to a motion vector. The image processing device comprises a judgment module and a selection module. According to the front picture and the back picture, the judgment module judges whether an interpolation image generated by the motion vector reaches the accuracy requirement or not aiming at each middle image block. According to a standard whether each interpolation image reaches the accuracy requirement or not, the selection module determines to choose the interpolation image or another replacement image to represent the middle image block.
Description
Technical field
The present invention is relevant with image processing technique, and especially relevant with the technology that produces the interpolation image.
Background technology
In recent years, flourish along with various electronic products, the multimedia systems such as family's theater are day by day universal.In most multimedia systems, most important hardware unit just belongs to display device.How to promote the quality of image frame to attract the consumer, be the subject under discussion that the display device producer pays close attention to always.
In some applications, provide to the changing picture frequency of the dynamic image of display device and be lower than the frame updating frequency that display device itself is set.For example, the changing picture frequency of original dynamic image may be 60 hertz, and the frame updating frequency of display device may be 120 hertz.Under these circumstances, display device can produce 60 newly-increased pictures in addition according to the raw frames of 60 of per seconds usually, and making the picture that exports on the screen is 120 of per seconds.The simplest but the most coarse way is with twice of each raw frames repeat playing; The shortcoming of this mode is to make the beholder to feel that picture is smooth not.The rule of doing that another is common is to produce a newly-increased picture according to two the raw frames interpolations in front and back (interpolate), and plays interpolation image between two raw frames.
In recent years, the mobile estimating (motion estimation) and motion compensation (motion compensation) technology that develop of movement-based vector (motion vector) is widely used in the image compression software/hardware.Similarly, when needing to produce interpolation image, display device also can utilize two motion-vectors between raw frames to produce the interpolation image.Before producing the interpolation image according to motion-vector, why must judge first two motion-vectors between raw frames.On the practice, image processor can be divided into a plurality of blocks with every picture, estimates respectively the motion-vector of each block.Be of a size of the situation of 8 pixel *, 8 pixels as example take each block, suppose that the A1 block in the A picture is identical with B1 block plane coordinates in the B picture, image processor can the B1 block centered by, interior which size of comparing the scope of its peripheral 32 pixel *, 32 pixels is that the zone of 8 pixel *, 8 pixels is the most close with the A1 block, finds out whereby the motion-vector between A1 block and B1 block.
In theory, above-mentioned comparison scope is wider, and the probability of correctly judging motion-vector is higher.Yet, being subject to operation time, it is too large that image processor can't be set the comparison scope usually.If the movement of objects originally in the A1 block is exceedingly fast, the position in the B picture has exceeded this comparison scope, and image processor is when searching this comparison scope, in any case also can not correctly find out motion-vector between A1, B1 two blocks.Perhaps, if the object that is taken acute variation at short notice causes existing between A1, B1 two blocks the situation of strong discrepancy, image processor also can be judged its motion-vector by accident usually.
That is to say that inevitably, motion-vector may be misestimated sometimes, and then the interpolation result who makes the mistake.For example, the sky picture in the interpolation image may some be the shadow of the trees by being misplaced, and is incompatible with periphery.Human eyes are responsive especially for the lofty variation of this class, therefore are easy to just can find mistake.
In view of this, developed at present remedial measure with the interpolation failure correlation.Existing way is that the possible errors amount summation in judging whole picture is when being higher than a certain threshold value, just abandon with the interpolation image as newly-increased image, use raw frames (for example A picture or B picture) instead and as newly-increased image, be equal to the repeat playing raw frames.The shortcoming of this way is, even if only comprise several blocks that have gross error in whole the interpolation image, but may force image processor to abandon the correct block of other interpolations result, cause the whole fluency of picture to descend because of the repeat playing raw frames.
Summary of the invention
For the problems referred to above, the present invention propose a kind of can be in newly-increased picture the local motion-vector that adopts as the scheme of interpolation foundation.More particularly, be to determine whether adopting the interpolation image that produces according to motion-vector take block as unit according to image processor of the present invention, image treatment method and image playing system.If the employing motion-vector according to when some block may mistake occur, only with these image blocks replaces with non-employing motion-vector as the image of interpolation foundation according to image processor of the present invention, image treatment method and image playing system as interpolation.
Be different from prior art whole Zhang Xinzeng picture is replaced with raw frames, in the final newly-increased picture that produces of image processor of the present invention, it is the interpolation image that produces according to motion-vector that some image block may be arranged, and some then is the replacement image that is different from the interpolation image that produces according to motion-vector.Way according to the present invention is presented on beholder's probability in front except the interpolation result that can reduce mistake, and compared to prior art, the fluency during actual dynamic play image is also more satisfactory.
A specific embodiment according to the present invention is an image processor, in order to produce a midway film that comprises a plurality of intermediate imagery blocks according to a front picture and a rear picture.Each intermediate imagery block is separately corresponding to a motion-vector.This image processor comprises judge module and selects module.According to this front picture with should after picture, this judge module judges for each intermediate imagery block whether an interpolation image that produces according to this motion-vector reaches a correctness requirement.Reach the intermediate imagery block that this correctness requires for this interpolation image, this selection module is selected to represent this intermediate imagery block with this interpolation image.Do not reach the intermediate imagery block that this correctness requires for this interpolation image, the replacement image that this selection module selects to be different from this interpolation image represents this intermediate imagery block.
Another specific embodiment according to the present invention is a kind of image treatment method, in order to produce a midway film that comprises a plurality of intermediate imagery blocks according to a front picture and a rear picture.Each intermediate imagery block is separately corresponding to a motion-vector.The method comprises the following step: (a) according to should before picture with should after picture, judge for each intermediate imagery block whether an interpolation image that produces according to this motion-vector reaches a correctness requirement; And the intermediate imagery block that (b) reaches this correctness requirement for this interpolation image, select to represent this intermediate imagery block with this interpolation image; Do not reach the intermediate imagery block that this correctness requires for this interpolation image, select to represent this intermediate imagery block with a replacement image that is different from this interpolation image.
Another specific embodiment according to the present invention is a kind of image playing system, wherein comprises receiver module, judge module, selection module and display.Receiver module is in order to receive a front picture and a rear picture.One midway film comprises a plurality of intermediate imagery blocks.Each intermediate imagery block is separately corresponding to a motion-vector.According to this front picture with should after picture, this judge module judges for each intermediate imagery block whether an interpolation image that produces according to this motion-vector reaches a correctness requirement.Reach the intermediate imagery block that this correctness requires for this interpolation image, this selection module is selected to represent this intermediate imagery block with this interpolation image.Do not reach the intermediate imagery block that this correctness requires for this interpolation image, the replacement image that this selection module selects to be different from this interpolation image represents this intermediate imagery block.Display is in order to sequentially to present this front picture, this midway film and to be somebody's turn to do rear picture.
Description of drawings
For above-mentioned purpose of the present invention, feature and advantage can be become apparent, below in conjunction with accompanying drawing the specific embodiment of the present invention is elaborated, wherein:
Fig. 1 is the calcspar according to the image processor in the specific embodiment of the present invention.
Fig. 2 presents the enforcement example according to judge module of the present invention.
Fig. 3 is the relativeness example of front picture, rear picture and midway film.
Fig. 4～Fig. 7 further presents in the image processor of the present invention in order to produce the hardware blocks of replacement image.
Fig. 8 is the flow chart according to the image treatment method in the specific embodiment of the present invention.
Fig. 9 is the calcspar according to the image playing system in the specific embodiment of the present invention.
The main element symbol description:
100: image processor 10: memory
12: judge module 12A: the difference computing unit
12B: comparing unit 14: select module
16: mixing module 18: interpolation module
20: weighting block 22: the weight decision module
22A: the first arithmetic element 22B: the second arithmetic element
22C: the 3rd arithmetic element S81～S82: process step
30: receiver module 32: display
Embodiment
A specific embodiment according to the present invention is an image processor, in order to produce at least one midway film F according to a front picture F1 and a rear picture F2.In practical application, front picture F1 and rear picture F2 can be in the video streaming on time sequencing adjacent two pictures, but not as limit.As shown in Figure 1, the image processor in the present embodiment 100 comprises judge module 12 and selects module 14.On the practice, image processor 100 can be integrated in TV, DVD player or computer etc. to be possessed in the system of image projector function, but also individualism.Midway film F is regarded as comprising a plurality of intermediate imagery blocks, and for example a plurality of sizes are the image block of 8 pixel *, 8 pixels and non-overlapping copies.
Each intermediate imagery block is separately corresponding to a motion-vector, but this motion-vector is unknown number at the beginning.According to front picture F1 and rear picture F2, judge module 12 is responsible for judging for each intermediate imagery block whether an interpolation image that produces according to motion-vector reaches a correctness requirement.Fig. 2 is painted with one of judge module 12 and implements example.In this example, judge module 12 comprises difference computing unit 12A and comparing unit 12B.Difference computing unit 12A is in order to calculate the different summation of a minimum absolute difference (sum of absolute difference, SAD) of each intermediate imagery block according to front picture F1 and rear picture F2.The detailed function mode of difference computing unit 12A below is described.
See also Fig. 3.Suppose that size is the intermediate imagery block B of 8 pixel *, 8 pixels
FThe plane coordinates of center in midway film F be (100,100), it is the scope R of 32 pixel *, 32 pixels that its peripheral definition has a size, equally centered by coordinate (100,100).As shown in Figure 3, the also signable out-of-plane identical scope R in position in front picture F1 and rear picture F2.Difference computing unit 12A can seek intermediate imagery block B in scope R
FMinimum SAD.For example, front image block B
F1With rear image block B
F2Coordinate, size all with B
FIdentical, therefore motion-vector between the two can be considered as (0,0).Difference computing unit 12A can at first judge front image block B
F1With rear image block B
F2SAD.Easy speech, difference computing unit 12A at first finds out is SAD corresponding to motion-vector (0,0).Subsequently, difference computing unit 12A can continue to seek in scope R the SAD corresponding to other different motion-vectors again.
Must the explanation be, each image area block size is not limited with 8 pixel *, 8 pixels, the size of scope R also is not limited with 32 pixel *, 32 pixels.In addition, the benchmark when difference computing unit 12A calculates SAD can be two image blocks RGB gray value or YUV values separately, but is not limited with these two kinds of numerical value.
Situation take the size of scope R as 32 pixel *, 32 pixels is as example, and the scope of above-mentioned motion-vector is between (15 ,-15) to (16,16), and difference computing unit 12A can find out 1024 SAD altogether.Then, difference computing unit 12A can be by selecting a minimum SAD among these SAD.In this example, image block B
FXWith image block B
FYThe most similar, therefore the SAD that finds out according to these two image blocks is intermediate imagery block B
FMinimum SAD.In addition, image block B
FX, B
FYBetween the coordinate relativeness can be regarded as intermediate imagery block B
FCorresponding motion-vector MV.
As image block B
FX, B
FYBetween difference excessive, represent that this motion-vector MV is likely wrong; Represent image block B if select with the interpolation image that produces according to this motion-vector MV
F, the beholder probably discovers wherein wrong existence.Therefore, intermediate imagery block B
FMinimum SAD can be as judging whether the interpolation image that produces according to motion-vector MV reach the foundation that a correctness requires.Comparing unit 12B is responsible for relatively should minimum SAD and a threshold value.If intermediate imagery block B
FMinimum SAD greater than this threshold value, comparing unit 12B judges that namely the interpolation image that produces according to motion-vector MV can't reach this correctness requirement.Relatively, if intermediate imagery block B
FMinimum SAD less than or equal to this threshold value, comparing unit 12B judges that namely the interpolation image that produces according to motion-vector MV will be enough to reach this correctness requirement.On the practice, the designer can decide the size of this threshold value in its sole discretion, is not limited with particular value.
According to same mode, difference computing unit 12A can find out the minimum SAD of each intermediate imagery block.Comparing unit 12B also can judge whether the interpolation image that produces according to motion-vector can reach the correctness requirement for each intermediate imagery block.By above explanation as can be known, judge module 12 can judge whether the interpolation image that produces according to motion-vector can reach the correctness requirement as long as determine the minimum SAD of intermediate imagery block, not necessarily wants the actual size of finding out motion-vector.
Subsequently, reach the intermediate imagery block that correctness requires for the interpolation image energy that produces according to motion-vector, select module 14 namely to select to represent this intermediate imagery block with the interpolation image that produces according to motion-vector.Relatively, can't reach the intermediate imagery block that correctness requires for the interpolation image that produces according to motion-vector, select module 14 can select to represent this intermediate imagery block with a replacement image that is different from this interpolation image.With intermediate imagery block B shown in Figure 3
FBe example, represent image block B if judge module 12 is assert with the interpolation image that is produced according to motion-vector MV
FBe likely wrong, select module 14 just can not select to represent image block B with the interpolation image that produces according to motion-vector MV
FIn an embodiment, intermediate imagery block B
FReplacement image be front image block B
F1Or rear image block B
F2
Can be found out that by above explanation image processor 100 according to the present invention is to determine whether adopting the interpolation image that produces according to motion-vector take block as unit.If the employing motion-vector, is only replaced these image blocks according to image processor of the present invention, but not such as prior art whole midway film is replaced with raw frames according to when some block may mistake occur as interpolation.In the final midway film F that produces, it is the interpolation image that produces according to motion-vector that some image block may be arranged, and some then is the replacement image that is different from the interpolation image that produces according to motion-vector.Way according to the present invention is presented on beholder's probability in front except the interpolation result that can reduce mistake, and compared to prior art, the fluency during the actual play image is also more satisfactory.
As shown in Figure 4, in another embodiment, further comprise memory 10 and mixing module 16 according to image processor of the present invention.Memory 10 is responsible for receiving and temporary front picture F1 and rear picture F2, for motion-vector module 12 and mixing module 16 references.As mentioned above, each intermediate imagery block can be corresponding to the front image block among the front picture F1 and the rear image block among the rear picture F2.The mixing module 16 responsible ratios of complying with will be somebody's turn to do front image block and the image data mixing of being somebody's turn to do in the rear image block, to produce the corresponding replacement image of this intermediate imagery block.For example, this ratio can be 50%/50%, and the image data that also is about to front image block and rear image block is averaged.Image block or rear image block are as replacement image before the direct selection, and the way of present embodiment can further promote the image fluency.
As shown in Figure 5, in another embodiment, image processor according to the present invention comprises mixing module 16, interpolation module 18 and weighting block 20.With intermediate imagery block B
FBe example, mixing module 16 is in order to front image block B
F1With rear image block B
F2Image data mix, to produce a composite image.As discussed previously, difference computing unit 12A upon the look between image block B
FMinimum SAD the time, also can determine in the lump corresponding to intermediate imagery block B
FMotion-vector MV.As shown in Figure 5, this motion-vector MV is provided to interpolation module 18.It is intermediate imagery block B that interpolation module 18 is responsible for according to motion-vector MV, front picture F1 and rear picture F2
FProduce an interpolation image.
On the practice, it is relevant with the situation of the contiguous imagery zone of intermediate imagery block that above-mentioned the first weights W 1 and the second weights W 2 can be designed to.See also Fig. 6, Fig. 6 has illustrated the embodiment that further comprises a weight decision module 22 according to image processor of the present invention.In this embodiment, the function of selecting module 14 is to be realized by weight decision module 22.Weight decision module 22 is responsible for decision for the first weights W 1 and second weights W 2 of weighting block 20.Except the judged result of reference comparing unit 12B, weight decision module 22 also optionally makes these weights relevant with the contiguous imagery zone of intermediate imagery block.Lift intermediate imagery block B
FBe example, its contiguous imagery zone can be and intermediate imagery block B
FCentral point is identical, and size is the scope of 16 pixel *, 16 pixels or 32 pixel *, 32 pixels, but not as limit.
In Fig. 7, can see the detailed enforcement example of weight decision module 22.Weight decision module 22 in this example comprises the first arithmetic element 22A, the second arithmetic element 22B and the 3rd arithmetic element 22C, and intermediate imagery block B
FContiguous imagery zone comprise and greater than intermediate imagery block B
FIn this embodiment, except determining intermediate imagery block B
FMinimum SAD (following table is shown SAD
1) outside, difference computing unit 14A also assists to determine that (following table is shown SAD for the minimum SAD of this vicinity imagery zone
2).
In this example, the function of comparing unit 12B is to be realized by the first arithmetic element 22A.As shown in Figure 7, the first arithmetic element 22A comprises a subtracter, at first from the first antipode summation SAD
1Deduct the first compensation rate O
1The first compensation rate O
1Can be set to the threshold value that previous described comparing unit 12B adopts.If the first antipode summation SAD
1Less than or equal to the first compensation rate O
1, the Output rusults of this subtracter is set to 0.That is to say, if the first antipode summation SAD
1Enough little, be enough to make corresponding interpolation image to reach the correctness standard, the Output rusults of the first arithmetic element 22A will be 0; The first weights W 1 can be therefore lower, reduces accordingly the weight of composite image.
The first arithmetic element 22A also comprises a multiplier, in order to will deduct the first compensation rate O
1After the result that subtracts each other multiply by the first gain G
1, to produce one first operation values.The effect of this multiplier is optionally to adjust the size of subtracting each other the result, for example this is subtracted each other result standard (normalize).Similarly, the second arithmetic element 22B also comprises a subtracter and a multiplier, in order to from the second antipode summation SAD
2Deduct one second compensation rate O
2, will subtract each other again the result and multiply by the second gain G
2, to produce one second operation values.The 3rd arithmetic element 22C is responsible for this first operation values and this second operation values addition, to produce the first weights W 1.On the practice, the function of these subtracters also can turn by difference computing unit 12A and realizes.
In the framework that Fig. 7 shows, antipode summation SAD
1And/or SAD
2Larger, just larger corresponding to the first weights W 1 of average image.The SAD of contiguous imagery zone
2Can reflect intermediate imagery block B
FThe image variation tendency of periphery.Generally speaking, the SAD of adjacent domain
2Higher, the probability that is unsuitable for the interpolation image of selective basis motion-vector generation is just healed high.In fact, might occur in this vicinity imagery zone and (comprise intermediate imagery block B
F) occur significantly changing, but intermediate imagery block B
FThe SAD of itself
1Do not surpass threshold value (that is the first compensation rate O
1) situation.Change though select the less intermediate imagery block of size can react more subtly image, when above-mentioned situation occurs, but be easy to cause erroneous judgement.The shown framework of Fig. 7 can remedy this problem.
In other embodiment, can consider simultaneously the situation of the adjacent domain of two or more different sizes according to image processor of the present invention, give the different weight of imagery zone of a plurality of different range, again with all weighted results additions, to produce the first weights W 1.For example, in an embodiment, image processor considers that simultaneously size is the scope of 8 pixel *, 8 pixels, 16 pixel *, 16 pixels and 32 pixel *, 32 pixels, produce respectively SAD1, SAD2 and SAD3, after producing respectively first, second and the 3rd operation values, produced mutually the first weights W 1.The first weights W 1 is sent to weighting block 20 as the multiplier of interpolation module 18, and the second weights W 2 is the complement (that is 1-W1=W2) of the first weights W 1.Weighting block 20 multiply by the image data of this composite image the second weights W 2 and the image data in this interpolation image be multiply by the 1 rear addition of the first weights W, to produce target image block B
FReplacement image.
Another specific embodiment according to the present invention is a kind of image treatment method, and in order to produce a midway film that comprises a plurality of intermediate imagery blocks according to a front picture and a rear picture, Fig. 8 is its flow chart.Each intermediate imagery block is separately corresponding to a motion-vector.At first, step S81 is picture according to picture before being somebody's turn to do and after being somebody's turn to do, and whether reaches a correctness requirement for each intermediate imagery block judgement according to the interpolation image that this motion-vector produces.Step S82 selects to represent this intermediate imagery block with this interpolation image for reach the intermediate imagery block that this correctness requires for this interpolation image; Do not reach the intermediate imagery block that this correctness requires for this interpolation image, select to represent this intermediate imagery block with a replacement image that is different from this interpolation image.As discussed previously, the alternative picture that adopts among the step S82 has multiple possibility, is not limited with the image block in the raw frames.In addition, among the step S81 in order to judge whether the interpolation image reaches the mode that correctness requires, and can be the minimum SAD that calculates each intermediate imagery block, and its detailed execution mode please refer to previous explanation, repeats no more in this.
Another specific embodiment according to the present invention is a kind of image playing system, and Fig. 9 is its functional block diagram.Except before having come across the judge module 12 among Fig. 1 and having selected the module 14, this image playing system also comprises a receiver module 30 and a display 32.Receiver module 30 is will be in order to produce midway film before picture and rear picture in order to receive.32 of displays are sequentially to present this front picture, this midway film and be somebody's turn to do rear picture.On the practice, receiver module 30 can be the memory 10 of before introducing, but not as limit.In addition, this image playing system mode that can adopt Fig. 4, Fig. 5 or Fig. 7 to present produces replacement image.The detailed execution mode of each module and variation kenel can with reference to previous explanation, repeat no more among this embodiment.
As mentioned above, be to determine whether adopting the interpolation image that produces according to motion-vector take block as unit according to image processor of the present invention, image treatment method and image playing system.If the employing motion-vector according to when some block may mistake occur, only with these image blocks replaces with non-employing motion-vector as the image of interpolation foundation according to image processor of the present invention, image treatment method and image playing system as interpolation.Way according to the present invention is presented on beholder's probability in front except the interpolation result that can reduce mistake, and compared to prior art, the fluency during actual dynamic play image is also more satisfactory.
What must illustrate is that conception of the present invention is not defined in the situation that can only produce a midway film.Easy speech, also can be used for producing many midway films according to a front picture and a rear picture according to image processor of the present invention, image treatment method and image playing system.
Although the present invention discloses as above with preferred embodiment; so it is not to limit the present invention, any those skilled in the art, without departing from the spirit and scope of the present invention; when can doing a little modification and perfect, so protection scope of the present invention is when with being as the criterion that claims were defined.
Claims (18)
1. image processor, in order to produce a midway film that comprises a plurality of intermediate imagery blocks according to a front picture and a rear picture, each intermediate imagery block is separately corresponding to a motion-vector, and this image processor comprises:
One judge module, according to this front picture with should after picture, each intermediate imagery block is judged whether an interpolation image that produces according to this motion-vector reaches a correctness requirement; And
One selects module, and this interpolation image is reached the intermediate imagery block that this correctness requires, and this selection module is selected to represent that with this interpolation image this interpolation image reaches the intermediate imagery block that this correctness requires; This interpolation image is not reached the intermediate imagery block that this correctness requires, and the replacement image that this selections module selects to be different from this interpolation image represents that this interpolation image does not reach the intermediate imagery block of this correctness requirement.
2. image processor as claimed in claim 1 is characterized in that, this judge module comprises:
One difference computing unit is in order to calculate the different summation of a minimum absolute difference of each intermediate imagery block according to picture after being somebody's turn to do front picture and being somebody's turn to do; And
One comparing unit, for the intermediate imagery block of the different summation of its minimum absolute difference greater than a threshold value, this comparing unit judges that corresponding this interpolation image of this intermediate imagery block does not reach this correctness requirement.
3. image processor as claimed in claim 1, it is characterized in that, each intermediate imagery block in the described intermediate imagery block is corresponding to the rear image block in the front image block in this front picture and this rear picture, and corresponding this replacement image of this intermediate imagery block is for being somebody's turn to do front image block or being somebody's turn to do rear image block.
4. image processor as claimed in claim 1 is characterized in that, each the intermediate imagery block in the described intermediate imagery block is corresponding to the rear image block in the front image block in this front picture and this rear picture; This image processor further comprises:
One mixing module mixes in order to image block before will being somebody's turn to do according to a ratio and the image data that is somebody's turn to do in the rear image block, to produce corresponding this replacement image of this intermediate imagery block.
5. image processor as claimed in claim 1 is characterized in that, each the intermediate imagery block in the described intermediate imagery block is corresponding to the rear image block in the front image block in this front picture and this rear picture; This image processor further comprises:
One interpolation module in order to according to this motion-vector corresponding to this intermediate imagery block, produces this interpolation image corresponding to this intermediate imagery block;
One mixing module mixes in order to image block before will being somebody's turn to do according to a ratio and the image data that is somebody's turn to do in the rear image block, to produce a composite image; And
One weighting block multiply by one first weight and the image data of this interpolation image be multiply by addition after one second weight in order to the image data with this composite image, to produce corresponding this replacement image of this intermediate imagery block.
6. image processor as claimed in claim 5 is characterized in that, this intermediate imagery block has a contiguous imagery zone; This image processor further comprises:
One weight decision module in order to determine this first weight and this second weight, is characterized in that, this second weight is relevant with this contiguous imagery zone.
7. image processor as claimed in claim 6 is characterized in that, this vicinity imagery zone comprises and greater than this intermediate imagery block, this weight decision module comprises:
One difference computing unit is in order to the one first antipode summation that determines this intermediate imagery block and one second antipode summation of this vicinity imagery zone;
One first arithmetic element after this first antipode summation deducts one first compensation rate certainly, is subtracted each other the result with this and be multiply by one first gain, to produce one first operation values;
One second arithmetic element after this second antipode summation deducts one second compensation rate certainly, is subtracted each other the result with this and be multiply by one second gain, to produce one second operation values; And
One the 3rd arithmetic element in order to this first operation values and this second operation values addition, to produce this second weight, wherein is the complement relation between this first weight and this second weight.
8. image treatment method, in order to produce a midway film that comprises a plurality of intermediate imagery blocks according to a front picture and a rear picture, each intermediate imagery block is separately corresponding to a motion-vector, and the method comprises:
(a) according to should before picture with should after picture, judge for each intermediate imagery block whether an interpolation image that produces according to this motion-vector reaches a correctness requirement; And
(b) this interpolation image is reached the intermediate imagery block that this correctness requires, select to represent that with this interpolation image this interpolation image reaches the intermediate imagery block that this correctness requires; This interpolation image is not reached the intermediate imagery block that this correctness requires, select to represent that with a replacement image that is different from this interpolation image this interpolation image does not reach the intermediate imagery block that this correctness requires.
9. image treatment method as claimed in claim 8 is characterized in that, step (a) comprise:
Calculate the different summation of a minimum absolute difference of each intermediate imagery block according to the rear picture of this front picture and this; And
To the intermediate imagery block of the different summation of this minimum absolute difference greater than a threshold value, judge that corresponding this interpolation image of this intermediate imagery block does not reach this correctness requirement.
10. image treatment method as claimed in claim 8, it is characterized in that, each intermediate imagery block in the described intermediate imagery block is corresponding to the rear image block in the front image block in this front picture and this rear picture, and corresponding this replacement image of this intermediate imagery block is for being somebody's turn to do front image block or being somebody's turn to do rear image block.
11. image treatment method as claimed in claim 8 is characterized in that, each the intermediate imagery block in the described intermediate imagery block is corresponding to the rear image block in the front image block in this front picture and this rear picture; The method further comprises:
If this interpolation image of this intermediate imagery block does not reach this correctness requirement, will be somebody's turn to do front image block according to a ratio and mix with the image data that is somebody's turn to do in the rear image block, to produce corresponding this replacement image of this intermediate imagery block.
12. image treatment method as claimed in claim 8 is characterized in that, each the intermediate imagery block in the described intermediate imagery block is corresponding to the rear image block in the front image block in this front picture and this rear picture; If this interpolation image of this intermediate imagery block does not reach this correctness requirement, the method further comprises:
According to this motion-vector corresponding to this intermediate imagery block, produce this interpolation image corresponding to this intermediate imagery block;
To be somebody's turn to do front image block according to a ratio and mix with the image data that is somebody's turn to do in the rear image block, to produce a composite image; And
The image data of this composite image be multiply by one first weight and the image data of this interpolation image be multiply by addition after one second weight, to produce corresponding this replacement image of this intermediate imagery block.
13. image treatment method as claimed in claim 12 is characterized in that, this intermediate imagery block has a contiguous imagery zone, and this second weight is relevant with this contiguous imagery zone.
14. image treatment method as claimed in claim 13 is characterized in that, this vicinity imagery zone comprises and greater than this intermediate imagery block; The method further comprises:
Determine one first antipode summation of this intermediate imagery block and one second antipode summation of this vicinity imagery zone;
After this first antipode summation deducts one first compensation rate, this is subtracted each other the result multiply by one first gain, to produce one first operation values;
After this second antipode summation deducts one second compensation rate, this is subtracted each other the result multiply by one second gain, to produce one second operation values; And
With this first operation values and this second operation values addition, to produce this second weight.
15. an image playing system comprises:
One receiver module is in order to receive a front picture and a rear picture;
One judge module, processing comprises a midway film of a plurality of intermediate imagery blocks, each intermediate imagery block is separately corresponding to a motion-vector, according to this front picture with should after picture, this judge module judges to each intermediate imagery block whether an interpolation image that produces according to this motion-vector reaches a correctness requirement;
One selects module, and this interpolation image is reached the intermediate imagery block that this correctness requires, and this selection module is selected to represent that with this interpolation image this interpolation image reaches the intermediate imagery block that this correctness requires; This interpolation image is not reached the intermediate imagery block that this correctness requires, and the replacement image that this selections module selects to be different from this interpolation image represents that this interpolation image does not reach the intermediate imagery block of this correctness requirement; And
One display is in order to sequentially to present this front picture, this midway film and to be somebody's turn to do rear picture.
16. image playing system as claimed in claim 15 is characterized in that, this judge module comprises:
One difference computing unit is in order to calculate the different summation of a minimum absolute difference of each intermediate imagery block according to picture after being somebody's turn to do front picture and being somebody's turn to do; And
One comparing unit, for the intermediate imagery block of the different summation of its minimum absolute difference greater than a threshold value, this comparing unit judges that corresponding this interpolation image of this intermediate imagery block does not reach this correctness requirement.
17. image playing system as claimed in claim 15 is characterized in that, each the intermediate imagery block in the described intermediate imagery block is corresponding to the rear image block in the front image block in this front picture and this rear picture; This image playing system further comprises:
One interpolation module in order to according to this motion-vector corresponding to this intermediate imagery block, produces this interpolation image corresponding to this intermediate imagery block;
One mixing module mixes in order to image block before will being somebody's turn to do according to a ratio and the image data that is somebody's turn to do in the rear image block, to produce a composite image; And
One weighting block multiply by one first weight and the image data of this interpolation image be multiply by addition after one second weight in order to the image data with this composite image, to produce corresponding this replacement image of this intermediate imagery block.
18. image playing system as claimed in claim 17 is characterized in that, this intermediate imagery block has a contiguous imagery zone, and this vicinity imagery zone comprises and greater than this intermediate imagery block; This image playing system further comprises:
One weight decision module in order to determining this first weight and this second weight, and comprises:
One difference computing unit is in order to the one first antipode summation that determines this intermediate imagery block and one second antipode summation of this vicinity imagery zone;
One first arithmetic element after this first antipode summation deducts one first compensation rate certainly, is subtracted each other the result with this and be multiply by one first gain, to produce one first operation values;
One second arithmetic element after this second antipode summation deducts one second compensation rate certainly, is subtracted each other the result with this and be multiply by one second gain, to produce one second operation values; And
One the 3rd arithmetic element is in order to this first operation values and this second operation values addition, to produce this second weight.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110270668.4A CN102984435B (en) | 2011-09-02 | 2011-09-02 | Image processor, image treatment method and image playing system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110270668.4A CN102984435B (en) | 2011-09-02 | 2011-09-02 | Image processor, image treatment method and image playing system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102984435A true CN102984435A (en) | 2013-03-20 |
| CN102984435B CN102984435B (en) | 2016-02-17 |
Family
ID=47858144
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110270668.4A Expired - Fee Related CN102984435B (en) | 2011-09-02 | 2011-09-02 | Image processor, image treatment method and image playing system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102984435B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101616247A (en) * | 2008-06-26 | 2009-12-30 | 晨星软件研发(深圳)有限公司 | The method of determining moving vectors and device thereof |
| CN101631241A (en) * | 2008-07-14 | 2010-01-20 | 晨星软件研发(深圳)有限公司 | Image processing method and device |
| EP2157799A1 (en) * | 2008-08-18 | 2010-02-24 | Panasonic Corporation | Interpolation filter with local adaptation based on block edges in the reference frame |
| CN101753795A (en) * | 2008-12-16 | 2010-06-23 | 晨星软件研发(深圳)有限公司 | Image processing method for determining motion vector of interpolation image and related device |
-
2011
- 2011-09-02 CN CN201110270668.4A patent/CN102984435B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101616247A (en) * | 2008-06-26 | 2009-12-30 | 晨星软件研发(深圳)有限公司 | The method of determining moving vectors and device thereof |
| CN101631241A (en) * | 2008-07-14 | 2010-01-20 | 晨星软件研发(深圳)有限公司 | Image processing method and device |
| EP2157799A1 (en) * | 2008-08-18 | 2010-02-24 | Panasonic Corporation | Interpolation filter with local adaptation based on block edges in the reference frame |
| CN101753795A (en) * | 2008-12-16 | 2010-06-23 | 晨星软件研发(深圳)有限公司 | Image processing method for determining motion vector of interpolation image and related device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102984435B (en) | 2016-02-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110545413B (en) | Method and apparatus for performing tone mapping of high dynamic range video | |
| KR101135454B1 (en) | Temporal interpolation of a pixel on basis of occlusion detection | |
| CN100452832C (en) | Motion estimation employing line and column vectors | |
| US7936950B2 (en) | Apparatus for creating interpolation frame | |
| CN102542529A (en) | Image processor and image processing method | |
| JP2006504175A (en) | Image processing apparatus using fallback | |
| WO2004066624A1 (en) | Background motion vector detection | |
| CN102281390A (en) | Image processing apparatus and image processing method | |
| JP2006190306A (en) | Method and device for processing mosaic of image | |
| US20160132282A1 (en) | Display apparatus and display methods thereof | |
| US20090296818A1 (en) | Method and system for creating an interpolated image | |
| CN113691758A (en) | Frame insertion method and device, equipment and medium | |
| US8385430B2 (en) | Video signal processing apparatus and video signal processing method | |
| CN104202555A (en) | Method and device for deinterlacing | |
| TWI493978B (en) | Image processing apparatus, image processing method and image display system | |
| US7593059B2 (en) | Image conversion device and method | |
| US8878999B2 (en) | Detection of graphics added to a video signal | |
| CN104580978A (en) | Video detecting and processing method and video detecting and processing device | |
| CN102984435B (en) | Image processor, image treatment method and image playing system | |
| TWI404408B (en) | Image processing apparatus and image processing method | |
| JP2005519379A (en) | Noise filtering in images | |
| US9373039B2 (en) | Detection of graphics added to a video signal | |
| US10587840B1 (en) | Image processing method capable of deinterlacing the interlacing fields | |
| CN101572816B (en) | Video processing method and film mode detection method | |
| US8941780B2 (en) | Mechanism for facilitating dynamic phase detection with high jitter tolerance for images of media streams |
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 |
Granted publication date: 20160217 Termination date: 20190902 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |