CN101521761B - Method and device for deinterlacing video by utilizing horizontal shift prediction and compensation - Google Patents
Method and device for deinterlacing video by utilizing horizontal shift prediction and compensation Download PDFInfo
- Publication number
- CN101521761B CN101521761B CN200810081294XA CN200810081294A CN101521761B CN 101521761 B CN101521761 B CN 101521761B CN 200810081294X A CN200810081294X A CN 200810081294XA CN 200810081294 A CN200810081294 A CN 200810081294A CN 101521761 B CN101521761 B CN 101521761B
- Authority
- CN
- China
- Prior art keywords
- pixel
- horizontal displacement
- displacement vector
- unit
- video data
- 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.)
- Active
Links
Images
Landscapes
- Television Systems (AREA)
Abstract
The invention discloses a deinterlacing method and a deinterlacing device. The deinterlacing method is used for deinterlacing video data comprising a first, a second, a third, a fourth and a fifth scenes. An output frame is generated when the video data is output at the third scene. The method comprises the following steps: detecting a horizontal displacement vector through a horizontal displacement prediction according to the first scene, the third scene, and at least one group of the third, the fifth, the second and the fourth scenes; selecting a first pixel from the second scene and a second pixel from the fourth scene based on the horizontal displacement vector if the horizontal displacement vector is detected, and interpolating a specific pixel value corresponding to a target pixel in the output frame according to the first pixel and the second pixel.
Description
Technical field
The invention provides a kind of image uninterleaving method and apparatus, refer to a kind of method and apparatus that utilizes horizontal displacement to estimate the image uninterleavingization of (horizontal motion estimation) and horizontal displacement compensation (horizontal motioncompensation) especially.
Background technology
Traditional interleaved type scanning (interlaced scanning) technology is that staggered the demonstration done in two fields forming this frame.These two fields are made up of strange field and even place.And formula scans (progressive scan) technology in proper order, claim noninterlaced scanning (non-interlaced) again, then be earlier a frame to be merged in two fields, scan all scan lines in this frame in regular turn with the horizontal frequency that doubles again, the quality of picture is promoted.
Image data in order to show alternating expression with formula scan mode in proper order just need utilize deinterleave (de-interlacing) method to mend out a new scan line in original continuous two slis.Prior art generally utilizes the processing method of displacement adaptability (motion-adaptive) to carry out the deinterleave operation.
Yet, when utilizing the displacement adaptability to carry out the deinterleave processing, can't interpolation go out high frequency and mobile details mostly, and, handles video data if passing through image coding and decoding (video encode/decode) again before carrying out the deinterleave operation, its film color phenomenon can allow HFS distortion and be difficult to interpolation and go out original image more, and then makes that the image quality of the image after the deinterleave processing is not good.
Summary of the invention
Therefore main purpose of the present invention is to provide a kind of the detecting with the horizontal displacement vector between a kind of and according to this horizontal displacement vector that one object pixel of one output frame is carried out the method and apparatus of the image uninterleavingization of horizontal displacement compensation, to promote image quality and to reduce the related hardware cost.
According to embodiments of the invention, it discloses a kind of de-interlacing methods, is used for deinterleave one video data.This video data comprises one first, second, third, fourth and the 5th, and this video data produces an output frame when the 3rd output.This method comprises the following steps: at least wherein one group according to second and the 4th of this first and the 3rd, the 3rd and the 5th and this, carries out horizontal displacement and estimates to detect a horizontal displacement vector; And if detect this horizontal displacement vector, then in this second, select one first pixel and in the 4th, select one second pixel, and come interpolation to go out a particular pixel values an object pixel in should output frame according to this first pixel and this second pixel according to this horizontal displacement vector.
According to embodiments of the invention, it discloses a kind of deinterlacing device in addition, is used for deinterleave one video data.This video data comprises one first, second, third, fourth and the 5th, and this video data produces an output frame when the 3rd output.This device comprises that a horizontal displacement estimates a unit and a horizontal displacement compensating unit.This horizontal displacement is estimated the unit and is carried out horizontal displacement according at least wherein one group of second and the 4th of this first and the 3rd, the 3rd and the 5th and this and estimate to detect a horizontal displacement vector.This horizontal displacement compensating unit is coupled to this horizontal displacement and estimates the unit, wherein estimate the unit and detect this horizontal displacement vector as if this horizontal displacement, then in this second, select one first pixel and in the 4th, select one second pixel, and come interpolation to go out a particular pixel values an object pixel in should output frame according to this first pixel and this second pixel according to this horizontal displacement vector.
Description of drawings
Fig. 1 includes a video data of five opposite fields and the schematic diagram of a corresponding output frame.
Fig. 2 is the schematic diagram of the deinterlacing device of one embodiment of the invention.
Fig. 3 carries out the schematic diagram of horizontal displacement compensation for horizontal displacement compensating unit of the present invention.
The flow chart that Fig. 4 carries out the deinterleave computing for deinterlacing device shown in Figure 2.
Fig. 5 is the schematic diagram of the deinterlacing device of another embodiment of the present invention.
The flow chart that Fig. 6 carries out the deinterleave computing for deinterlacing device shown in Figure 5.
[primary clustering symbol description]
200、500? | |
210、 510? | The unit is estimated in |
220、520? | The horizontal |
222、522? | Selected |
224、524? | |
530? | |
540? | The decision unit | ? | ? |
Embodiment
See also Fig. 1.Fig. 1 includes a video data 100 of five opposite fields (field) and the schematic diagram of a corresponding output frame (frame) 160.Output frame 160 behind the deinterleave is corresponding to time T, and five opposite fields 110,120,130,140 and 150 in the video data 100 correspond respectively to time T-2, T-1, T, T+1 and T+2.In Fig. 1, scan line 112,132,152 and 162 is respectively the N-1 bar scan line in the field 110,130,150 and 160; Scan line 123,143 and 163 is respectively the N bar scan line in the field 120,140 and 160; Scan line 114,134,154 and 164 then is respectively the N+1 bar scan line in the field 110,130,150 and 160, and wherein N-1 bar scan line and N+1 bar scan line belong to the another kind of field of N bar scan line genus with a kind of.In addition, the de-interlacing methods in embodiments of the invention is to produce output frame 160 by pixel-wise (pixel-by-pixel basis).
See also Fig. 2, Fig. 2 is the schematic diagram of the deinterlacing device 200 of one embodiment of the invention.Deinterlacing device 200 includes a horizontal displacement and estimates a unit 210 and a horizontal displacement compensating unit 220, and horizontal displacement compensating unit 220 then includes a selected cell 222 and a computing unit 224.When video data inputs to after horizontal displacement in the deinterlacing device 200 estimates unit 210, unit 210 is estimated in horizontal displacement can carry out the action that horizontal displacement is estimated to this video data.In the present embodiment, suppose that deinterlacing device 200 carries out the deinterleave processing corresponds to this output frame 160 with generation this object pixel 10 for an object pixel 10 of field 130.Therefore, the action that unit 210 can carry out the horizontal displacement judgement for the adjacent domain of object pixel 10 is estimated in horizontal displacement of the present invention.On real the work, unit 210 is estimated in horizontal displacement can be the center by this object pixel 10, set one and estimate scope (window), and in this estimates scope, the displacement detection (motion detection) of interframe is carried out to find out a horizontal displacement vector (motion vector) in field 130.As shown in Figure 1, in the present embodiment, suppose to estimate scope 101 and be set at a 2*5 pixel size, then horizontal displacement estimate unit 210 can compare according to estimating scope 101 130 with 150 or with the change in displacement of field 110, or foundation is estimated the change in displacement that scope 101 is compared contiguous field 120 with field 140, the pixel 15 of field 130 for example shown in Figure 1 and 16 pixel value move to the pixel 13 and 14 of field 150, or the pixel value of the pixel 12 in the field 120 moves to the pixel value of the pixel 11 of field 140.Therefore horizontal displacement is estimated unit 210 and is judged that the neighbouring part at object pixel 10 has horizontal displacement generation and horizontal displacement vector to be+2 in the present embodiment.
See also Fig. 3, Fig. 3 carries out the schematic diagram of horizontal displacement compensation for horizontal displacement compensating unit 220 of the present invention.Estimate after unit 210 inputs to the horizontal displacement vector that is judged selected cell 222 in the horizontal displacement compensating unit 220 in horizontal displacement.Selected cell 222 is chosen according to the horizontal displacement vector and is desired to be used for the reference pixel that interpolation goes out object pixel 10 and come out, as shown in Figure 3, in the present embodiment, because horizontal displacement vector be+2, so selected cell 222 selections are corresponding to the pixel 12 of the previous field 120 of object pixel 10 in the field 130 and be used as being used for interpolation corresponding to the pixel 11 of the back field 140 of object pixel 10 in the field 130 and go out the reference pixel of object pixel 10.Please note, its horizontal displacement vector is an even number value in the present embodiment, therefore each pixel was used as reference pixel before and after horizontal displacement compensating unit 220 can be selected, yet if the horizontal displacement vector is an odd number value (for example+3), horizontal displacement compensating unit 220 then of the present invention also can select each two pixel of front and back of corresponding this horizontal displacement vector to be used as reference pixel, for example the pixel 12 of the previous field 120 of object pixel 10 in the horizontal displacement vector can be selected corresponding to field 130 for selected cell 222 under+3 the situation, 20 and corresponding to the pixel 11 of the back field 140 of object pixel 10 in the field 130,19 are used as being used for interpolation goes out the reference pixel of object pixel 10.Next, computing unit 224 in the horizontal displacement compensating unit 220 is again according to the reference pixel of target 10 (for example 120 pixel 12 and 140 pixel 11), utilizes the pixel value of average reference pixel to come interpolation to go out the pixel value of object pixel 10 of output frame 160 to finish the action of deinterleaveization.
See also Fig. 4, the flow chart that Fig. 4 carries out the deinterleave computing for deinterlacing device 200 shown in Figure 2.Note that correlation step is not necessarily carried out continuously in accordance with this ordering in the flow chart, and other step also may be inserted wherein.The step that deinterlacing device 200 carries out the deinterleave computing can be summarized as follows:
Step 400: horizontal displacement is estimated unit 210 and is estimated 101 pairs of these object pixels 10 of scope according to one and carry out the action that horizontal displacement is estimated.
Step 402: judge that the adjacent domain of this object pixel 10 has the horizontal displacement action to take place if unit 210 is estimated in horizontal displacement, then execution in step 404, otherwise, finish this flow process.
Step 404: horizontal displacement is estimated unit 210 and is determined a horizontal displacement vector according to this horizontal displacement.
Step 406: selected cell 222 is according to a plurality of reference pixels 11,12 of this horizontal displacement vector decision with respect to this object pixel 10.
Step 408: the pixel value of computing unit 224 average reference pixels 11,12 is as the pixel value of object pixel 10.
See also Fig. 5, Fig. 5 is the schematic diagram of the deinterlacing device 500 of another embodiment of the present invention.In embodiments of the present invention, deinterlacing device 500 includes horizontal displacement and estimates unit 510, a horizontal displacement compensating unit 520, a displacement adaptability circuit 530 and a decision unit 540.520 of horizontal displacement compensating units include a selected cell 522 and a computing unit 524.Note that deinterlacing device shown in Figure 5 500 and assembly of the same name in the deinterlacing device 200 shown in Figure 2 have identical functions and operation, so do not give unnecessary details in addition in this.As mentioned above, because known displacement adaptability circuit 530 utilizes displacement adaptability de-interlacing methods (motion adaptive de-interlacing) to carry out deinterleave when action, mostly can't interpolation go out high frequency and mobile details thereby cause the distortion of image.Therefore, can utilize horizontal displacement of the present invention to estimate the deficiency that unit 510 and horizontal displacement compensating unit 520 are supplied displacement adaptability circuit 530.In the present embodiment, when horizontal displacement is estimated 510 pairs one object pixels in unit and is carried out horizontal displacement and estimate, if this object pixel 10 has horizontal displacement to take place, the pixel value that then determines unit 540 to be calculated according to horizontal displacement compensating unit 520 is used as the pixel value of object pixel 10, and does not choose the pixel value that displacement adaptability circuit 530 is calculated; On the other hand, if this object pixel 10 is estimated unit 510 by horizontal displacement and is judged when not having horizontal displacement to take place, then determine unit 540 just to choose the pixel value that displacement adaptability circuit 530 calculated and be used as the pixel value of object pixel 10, and do not remove to choose the pixel value that horizontal displacement compensating unit 520 is calculated.Therefore, whether deinterlacing device 500 of the present invention can take place according to the horizontal displacement of object pixel, decide the desire difference to mend the deinterleave mode of this object pixel 10, therefore strengthened the accuracy and the susceptibility of deinterleave process, and then significantly promoted the quality of picture.
See also Fig. 6, the flow chart that Fig. 6 carries out the deinterleave computing for deinterlacing device 500 shown in Figure 5.Note that correlation step is not necessarily carried out continuously in accordance with this ordering in the flow chart, and other step also may be inserted wherein.The step that deinterlacing device 500 carries out the deinterleave computing can be summarized as follows:
Step 600: horizontal displacement is estimated 510 pairs of these object pixels 10 in unit and is carried out the action that horizontal displacement is estimated.
Step 602: horizontal displacement compensating unit 520 is estimated the horizontal displacement vector that unit 510 determined according to horizontal displacement and is produced a particular pixel values.
Step 604: 530 foundations of displacement adaptability circuit, one displacement adaptability deinterleave is to produce a candidate pixel value.
Step 606:, then determine unit 540 to use this particular pixel values to set this object pixel if detect this horizontal displacement vector; And if do not detect this horizontal displacement vector, then this decision unit 540 uses this candidate pixel value to set this object pixel.
Compared to prior art, deinterlacing device of the present invention carries out image uninterleaving when operation utilizing horizontal displacement to estimate and compensate with horizontal displacement, utilized one to estimate scope and detect moving horizontally of target zone, therefore can significantly reduce and search required hardware cost on a large scale.On the other hand, deinterlacing device of the present invention also can utilize horizontal displacement to estimate the deficiency of supplying displacement adaptability deinterleave mode with horizontal displacement, not only strengthen the accuracy and the susceptibility of deinterleave process, and and then significantly promoted the quality of whole picture.
The above only is preferred embodiment of the present invention, and all equalizations of being done according to the present patent application claim change and modify, and all should belong to covering scope of the present invention.
Claims (8)
1. de-interlacing methods is used for deinterleave one video data, and this video data comprises one first, second, third, fourth and the 5th, and this video data produces an output frame when the 3rd output, and this method comprises the following steps:
According to this first and the 3rd, the 3rd and the 5th and this one of them group of second and the 4th, carry out horizontal displacement and estimate to detect a horizontal displacement vector; And
If detect this horizontal displacement vector, then in this second, select one first pixel and in the 4th, select one second pixel, and come interpolation to go out a particular pixel values an object pixel in should output frame according to this first pixel and this second pixel according to this horizontal displacement vector.
2. the method for claim 1, the step of wherein detecting this horizontal displacement vector comprises:
Set a processes pixel scope according to this object pixel; And
Carrying out horizontal displacement according to this processes pixel scope estimates to detect this horizontal displacement vector.
3. the method for claim 1 wherein comprises in order to the step that produces this object pixel:
If this horizontal displacement vector that detects equals 2N, then:
The pixel of choosing in this second at a distance of N pixel of this object pixel according to this horizontal displacement vector is used as this first pixel;
The pixel of choosing in the 4th at a distance of N pixel of this object pixel according to this horizontal displacement vector is used as this second pixel; And
The pixel value of average this first pixel and the pixel value of this second pixel are used as this particular pixel values; And
If this horizontal displacement vector that detects equals 2N+1, then:
According to this horizontal displacement vector choose in this second pixel at a distance of N pixel of this object pixel be used as this first pixel and apart the pixel of N+1 pixel of this object pixel be used as one the 3rd pixel, wherein this first pixel is adjacent with the 3rd pixel;
According to this horizontal displacement vector choose in the 4th pixel at a distance of N pixel of this object pixel be used as this second pixel and apart the pixel of N+1 pixel of this object pixel be used as one the 4th pixel, wherein this second pixel is adjacent with the 4th pixel; And
The pixel value of this first pixel, this second pixel, the 3rd pixel and the 4th pixel is average, be used as this particular pixel values.
4. the method for claim 1, it comprises in addition:
This video data is carried out a displacement adaptability deinterleave to produce the candidate pixel value to this object pixel in should output frame, wherein when not detecting this horizontal displacement vector, use this candidate pixel value to set this object pixel.
5. deinterlacing device is used for deinterleave one video data, and this video data comprises one first, second, third, fourth and the 5th, and this video data produces an output frame when the 3rd output, and this device comprises:
The unit is estimated in one horizontal displacement, and at least wherein one group according to second and the 4th of this first and the 3rd, the 3rd and the 5th and this carries out horizontal displacement and estimates to detect a horizontal displacement vector; And
One horizontal displacement compensating unit, be coupled to this horizontal displacement and estimate the unit, if this horizontal displacement is estimated the unit and is detected this horizontal displacement vector, then in this second, select one first pixel and in the 4th, select one second pixel, and come interpolation to go out a particular pixel values an object pixel in should output frame according to this first pixel and this second pixel according to this horizontal displacement vector.
6. device as claimed in claim 5, wherein this horizontal displacement is estimated the unit and is set a processes pixel scope according to this object pixel; And carrying out horizontal displacement according to this processes pixel scope estimates to detect this horizontal displacement vector.
7. device as claimed in claim 5, wherein this horizontal displacement compensating unit comprises:
One selected cell, be used for according to this horizontal displacement vector choose in this second pixel at a distance of N pixel of this object pixel be used as this first pixel and apart the pixel of N+1 pixel of this object pixel be used as one the 3rd pixel, wherein this first pixel is adjacent with the 3rd pixel, and according to this horizontal displacement vector choose in this output frame pixel at a distance of N pixel of this object pixel be used as this second pixel and apart the pixel of N+1 pixel of this object pixel be used as one the 4th pixel, wherein this second pixel is adjacent with the 4th pixel; And
One computing unit, be coupled to this selected cell, when this horizontal displacement vector equals 2N, the pixel value of average this first pixel and second pixel is used as this particular pixel values, when this horizontal displacement vector equals 2N+1, on average the pixel value of this first pixel, this second pixel, the 3rd pixel and the 4th pixel is used as this particular pixel values.
8. device as claimed in claim 5, it comprises in addition:
One displacement adaptability circuit is used for this video data is carried out a displacement adaptability deinterleave to produce the candidate pixel value to this object pixel in should output frame; And
One decision unit, be coupled to this displacement adaptability circuit, unit and this horizontal displacement compensating unit are estimated in this horizontal displacement, if detect this horizontal displacement vector, then this decision unit uses this particular pixel values to set this object pixel, and if do not detect this horizontal displacement vector, then this decision unit uses this candidate pixel value to set this object pixel.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN200810081294XA CN101521761B (en) | 2008-02-26 | 2008-02-26 | Method and device for deinterlacing video by utilizing horizontal shift prediction and compensation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN200810081294XA CN101521761B (en) | 2008-02-26 | 2008-02-26 | Method and device for deinterlacing video by utilizing horizontal shift prediction and compensation |
Publications (2)
Publication Number | Publication Date |
---|---|
CN101521761A CN101521761A (en) | 2009-09-02 |
CN101521761B true CN101521761B (en) | 2011-03-30 |
Family
ID=41082108
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN200810081294XA Active CN101521761B (en) | 2008-02-26 | 2008-02-26 | Method and device for deinterlacing video by utilizing horizontal shift prediction and compensation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101521761B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116016831B (en) * | 2022-12-13 | 2023-12-05 | 湖南快乐阳光互动娱乐传媒有限公司 | Low-time-delay image de-interlacing method and device |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6512550B1 (en) * | 1999-06-29 | 2003-01-28 | Lsi Logic Corporation | Motion compensated de-interlacing |
US6577345B1 (en) * | 1999-07-29 | 2003-06-10 | Lg Electronics Inc. | Deinterlacing method and apparatus based on motion-compensated interpolation and edge-directional interpolation |
CN1674669A (en) * | 2004-03-26 | 2005-09-28 | 瑞昱半导体股份有限公司 | Image dealternating method and apparatus utilizing displacement vector compensation |
CN1741603A (en) * | 2004-08-26 | 2006-03-01 | 瑞昱半导体股份有限公司 | Method for cancelling interlacing and related apparatus |
CN1774079A (en) * | 2004-11-10 | 2006-05-17 | 瑞昱半导体股份有限公司 | Method for suppressing error image colour |
-
2008
- 2008-02-26 CN CN200810081294XA patent/CN101521761B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6512550B1 (en) * | 1999-06-29 | 2003-01-28 | Lsi Logic Corporation | Motion compensated de-interlacing |
US6577345B1 (en) * | 1999-07-29 | 2003-06-10 | Lg Electronics Inc. | Deinterlacing method and apparatus based on motion-compensated interpolation and edge-directional interpolation |
CN1674669A (en) * | 2004-03-26 | 2005-09-28 | 瑞昱半导体股份有限公司 | Image dealternating method and apparatus utilizing displacement vector compensation |
CN1741603A (en) * | 2004-08-26 | 2006-03-01 | 瑞昱半导体股份有限公司 | Method for cancelling interlacing and related apparatus |
CN1774079A (en) * | 2004-11-10 | 2006-05-17 | 瑞昱半导体股份有限公司 | Method for suppressing error image colour |
Also Published As
Publication number | Publication date |
---|---|
CN101521761A (en) | 2009-09-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Kang et al. | Motion compensated frame rate up-conversion using extended bilateral motion estimation | |
CN101222604B (en) | Operation mobile estimation value and method for estimating mobile vector of image | |
US8144778B2 (en) | Motion compensated frame rate conversion system and method | |
CN102946505B (en) | Self-adaptive motion detection method based on image block statistics | |
CN101106685A (en) | An interlining removal method and device based on motion detection | |
CN101971209A (en) | Sparse geometry for super resolution video processing | |
CN103369208A (en) | Self-adaptive de-interlacing method and device | |
Tai et al. | A multi-pass true motion estimation scheme with motion vector propagation for frame rate up-conversion applications | |
Heinrich et al. | Optimization of hierarchical 3DRS motion estimators for picture rate conversion | |
US7405765B2 (en) | Method and apparatus for de-interlacing video data | |
CN102447870A (en) | Detection method for static objects and motion compensation device | |
CN100401763C (en) | Motion correction device and method | |
US7446818B2 (en) | Apparatus and related method for film mode detection using motion estimation | |
CN102364933A (en) | Motion-classification-based adaptive de-interlacing method | |
CN101521761B (en) | Method and device for deinterlacing video by utilizing horizontal shift prediction and compensation | |
CN101340539A (en) | Deinterlacing video processing method and system by moving vector and image edge detection | |
CN201222771Y (en) | High speed edge self-adapting de-interlaced interpolation device | |
CN102170549A (en) | Edge correlation image intra-field de-interlacing algorithm of edge pre-judgment | |
CN1199449C (en) | Method for removing intersection by adopting error protection motion compensation and its equipment | |
CN107124617B (en) | Method and system for generating random vector in motion estimation motion compensation | |
CN105915835B (en) | One kind can hard-wired frame frequency lifting method | |
Kang et al. | Design and implementation of median filter based adaptive motion vector smoothing for motion compensated frame rate up-conversion | |
US8922711B2 (en) | Method and apparatus for de-interlacing video data | |
CN105611214A (en) | Method for de-interlacing through intra-field linear interpolation based on multidirectional detection | |
CN102497492B (en) | Detection method for subtitle moving in screen |
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 |