CN105744201B - The de-interlaced treating method and apparatus of TV signal - Google Patents

The de-interlaced treating method and apparatus of TV signal Download PDF

Info

Publication number
CN105744201B
CN105744201B CN201511025765.1A CN201511025765A CN105744201B CN 105744201 B CN105744201 B CN 105744201B CN 201511025765 A CN201511025765 A CN 201511025765A CN 105744201 B CN105744201 B CN 105744201B
Authority
CN
China
Prior art keywords
pixel
value
interpolation
window
gradient zone
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
Application number
CN201511025765.1A
Other languages
Chinese (zh)
Other versions
CN105744201A (en
Inventor
周鸿�
韦虎
罗小伟
王森
陈敏杰
朱俊
薛端
林福辉
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Spreadtrum Communications Shanghai Co Ltd
Original Assignee
Spreadtrum Communications Shanghai Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Spreadtrum Communications Shanghai Co Ltd filed Critical Spreadtrum Communications Shanghai Co Ltd
Priority to CN201511025765.1A priority Critical patent/CN105744201B/en
Publication of CN105744201A publication Critical patent/CN105744201A/en
Priority to US15/234,356 priority patent/US9819900B2/en
Application granted granted Critical
Publication of CN105744201B publication Critical patent/CN105744201B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00Television systems
    • H04N7/01Conversion of standards, e.g. involving analogue television standards or digital television standards processed at pixel level
    • H04N7/0117Conversion of standards, e.g. involving analogue television standards or digital television standards processed at pixel level involving conversion of the spatial resolution of the incoming video signal
    • H04N7/012Conversion between an interlaced and a progressive signal
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00Television systems
    • H04N7/01Conversion of standards, e.g. involving analogue television standards or digital television standards processed at pixel level
    • H04N7/0135Conversion of standards, e.g. involving analogue television standards or digital television standards processed at pixel level involving interpolation processes
    • H04N7/0147Conversion of standards, e.g. involving analogue television standards or digital television standards processed at pixel level involving interpolation processes the interpolation using an indication of film mode or an indication of a specific pattern, e.g. 3:2 pull-down pattern

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Graphics (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Television Systems (AREA)

Abstract

The present invention discloses a kind of de-interlaced treating method and apparatus of TV signal.Method includes: to judge whether interpolation pixel is in strong still-mode;If so, obtaining the pixel value of the first pixel corresponding with the interpolation pixel in previous field, the pixel value of the interpolation pixel is supplemented, the pixel value of the interpolation pixel is equal to the pixel value of first pixel;If not, dividing the gradient zone of multiple directions centered on the interpolation pixel, the directionality of the interpolation pixel is determined by the gradient zone, and along the direction of the interpolation pixel, the pixel value of the interpolation pixel is supplemented by interpolation.Technical solution provided by the invention can effectively improve the accuracy of interpolation point pixel value.

Description

The de-interlaced treating method and apparatus of TV signal
This application claims in submission on December 31st, 2014 China State Intellectual Property Office, application No. is 201410854341.5, the Chinese patent application of entitled " the de-interlaced treating method and apparatus of TV signal " it is preferential Power, entire contents are hereby incorporated by reference in the application.
Technical field
The present invention relates to signal processing technology field more particularly to the de-interlaced treating method and apparatus of TV signal.
Background technique
Interlacing scan is the common form of TV signal, in the TV signal sequential interlace alternate according to top field bottom Afterwards, then by the top field data of adjacent (adjacent) one half data of missing and missing one half data bottom field data gather into a frame then at For the video flowing of an interlacing, but there are interlacing effect phenomenons for such video flowing.Therefore, TV signal de interlacing is processing The inevitable requirement of TV signal.
In current existing technology, known pixels point is generally basede on to infer the directionality of missing pixel point, then uses institute Known point on the direction of deduction carrys out the pixel value that interpolation obtains missing pixel point.
The scheme of the prior art has the following deficiencies:
(1) when hardware resource is limited, for example, in the vertical direction only have two rows data (1 row linebuffer) when, Almost all of algorithm all cannot be guaranteed that walking direction does not have vicious possibility, even and if certain details when data are more than enough still Good interpolation cannot be carried out by walking direction, because some details do not have very specific directionality (such as certain of text A little corners).Image in true the preceding paragraph video may have quite a few in absolute rest state, or be in Relative static conditions (global motion).
(2) be on the judgement direction under moving scene de interlacing (deinterlace) algorithm difficult point, especially direction Property present low angle when.
Summary of the invention
In view of this, the present invention provides a kind of de-interlaced treating method and apparatus of TV signal, to effectively improve The accuracy of interpolation point pixel value.
To achieve the above object, the invention provides the following technical scheme:
A kind of de-interlaced processing method of TV signal, comprising:
Judge whether interpolation pixel is in strong still-mode;
If so, the pixel value of the first pixel corresponding with the interpolation pixel in previous field is obtained, supplement The pixel value of the interpolation pixel, the pixel value of the interpolation pixel are equal to the pixel value of first pixel;
If not, the gradient zone of multiple directions is divided centered on the interpolation pixel, it is true by the gradient zone The directionality of the fixed interpolation pixel supplements the interpolation picture by interpolation along the direction of the interpolation pixel The pixel value of vegetarian refreshments.
It is preferably, described to judge whether interpolation pixel is in strong still-mode, comprising:
Centered on the interpolation pixel, first window is chosen, includes known at least two rows in the first window The pixel of pixel value successively judges pixel value between each pixel of every row and the pixel of former frame corresponding position The sum of absolute value, if less than preset threshold respectively voluntarily;
With centered on the second pixel, the second window is chosen, includes at least a line known pixels in second window The pixel of value successively judges the absolute of pixel value between each pixel of every row and the pixel of former frame corresponding position The sum of value, if less than preset threshold respectively voluntarily;It is on the scene that second pixel is located at the first window institute in same frame Opposite field, the position of second pixel and the interpolation pixel in their own field is corresponding.
It is preferably, described to judge whether interpolation pixel is in strong still-mode, further includes:
Centered on the interpolation pixel, third window is chosen, includes known at least two rows in the third window The pixel of pixel value successively judges pixel value between each pixel of every row and the pixel of former frame corresponding position The sum of absolute value, if less than preset threshold respectively voluntarily;The line number of pixel and the first window in the third window The line number of interior pixel is equal, columns is greater than the columns of pixel in the first window;
With centered on second pixel, the 4th window is chosen, includes at least known to a line in the 4th window The pixel of pixel value successively judges pixel value between each pixel of every row and the pixel of former frame corresponding position The sum of absolute value, if less than preset threshold respectively voluntarily.
Preferably, the gradient zone for dividing multiple directions, comprising:
Divide the gradient zone in 3 directions.
Preferably, the directionality that the interpolation pixel is determined by the gradient zone, comprising:
First differential value after comparing each direction gradient band normalization, taking the smallest direction of differential value is first direction, Secondary small direction is second direction, if first differential value and second direction gradient zone normalizing after the normalization of first direction gradient zone The absolute value of the difference of first differential value after change is greater than default value, it is determined that the first direction is the interpolation pixel The direction of point.
Preferably, the directionality that the interpolation pixel is determined by the gradient zone, further includes:
If the first differential value after the normalization of first direction gradient zone and the single order after the normalization of second direction gradient zone are micro- The absolute value of the difference of score value is less than default value, then chooses the ladder of the symmetry direction in each direction in the first and second directions Band is spent, the absolute value of the difference of described two directions and the first differential value of its symmetry direction gradient zone is respectively compared, takes difference The big direction of absolute value is the final interpolation direction of interpolation pixel, if the first differential value of first direction and its symmetry direction Difference absolute value be greater than second direction and its symmetry direction first differential value difference absolute value, then take first direction For interpolation direction, otherwise taking second direction is interpolation direction.
Preferably, before the directionality that the interpolation pixel is determined by the gradient zone, further includes:
Down-sampled processing is made to the gradient zone in the multiple direction, the gradient zone for the multiple directions being simplified.
A kind of de-interlaced processing unit of TV signal, comprising:
Judgment module, for judging whether interpolation pixel is in strong still-mode;
First processing module, for when the interpolation pixel be in strong still-mode, in acquisition previous field with institute The pixel value for stating corresponding first pixel of interpolation pixel, supplements the pixel value of the interpolation pixel, it is described to The pixel value of interpolating pixel point is equal to the pixel value of first pixel;
Second processing module is used for when the interpolation pixel is in motor pattern, with the interpolation pixel Centered on, the gradient zone of multiple directions is divided, the directionality of the interpolation pixel is determined by the gradient zone, along described The direction of interpolation pixel supplements the pixel value of the interpolation pixel by interpolation.
Preferably, the judgment module includes:
First judging unit is used for centered on the interpolation pixel, selection first window, in the first window Pixel including at least two row known pixel values, successively judges the pixel of every row each pixel and former frame corresponding position Between pixel value the sum of absolute value, if less than preset threshold respectively voluntarily;With centered on the second pixel, choose Second window, the interior pixel including at least a line known pixel values of second window, successively judges each pixel of every row The sum of the absolute value of pixel value between the pixel of former frame corresponding position, if less than preset threshold respectively voluntarily; Second pixel is located at the opposite field that the first window is on the scene in same frame, second pixel and described to be inserted It is corresponding to be worth position of the pixel in their own field.
Preferably, the judgment module further include:
Second judgment unit is used for centered on the interpolation pixel, selection third window, in the third window Pixel including at least two row known pixel values, successively judges the pixel of every row each pixel and former frame corresponding position Between pixel value the sum of absolute value, if less than preset threshold respectively voluntarily;The row of pixel in the third window Count the columns that, columns equal with the line number of pixel in the first window is greater than pixel in the first window;With with institute Centered on stating the second pixel, the 4th window is chosen, includes the pixel of at least a line known pixel values in the 4th window, Successively judge the sum of the absolute value of pixel value between each pixel of every row and the pixel of former frame corresponding position, if Less than preset threshold respectively voluntarily.
Preferably, the Second processing module includes:
Gradient zone division unit is used for when the interpolation pixel is in motor pattern, with the interpolation pixel Centered on point, the gradient zone of multiple directions is divided;
Direction calculating unit, for determining the directionality of the interpolation pixel by the gradient zone;
Interpolating unit, for supplementing the interpolation pixel by interpolation along the direction of the interpolation pixel Pixel value.
Preferably, the Second processing module further include:
Down-sampled processing unit makees down-sampled processing for the gradient zone to the multiple direction, and what is be simplified is multiple The gradient zone in direction.
Preferably, the directionality of the interpolation pixel is determined described in Second processing module by the gradient zone, is wrapped It includes:
First differential value after comparing each direction gradient band normalization, taking the smallest direction of differential value is first direction, Secondary small direction is second direction, if first differential value and second direction gradient zone normalizing after the normalization of first direction gradient zone The absolute value of the difference of first differential value after change is greater than default value, it is determined that the first direction is the interpolation pixel The direction of point.
Preferably, the directionality of the interpolation pixel is determined described in Second processing module by the gradient zone, also Include:
If the first differential value after the normalization of first direction gradient zone and the single order after the normalization of second direction gradient zone are micro- The absolute value of the difference of score value is less than default value, then chooses the ladder of the symmetry direction in each direction in the first and second directions Band is spent, the absolute value of the difference of described two directions and the first differential value of its symmetry direction gradient zone is respectively compared, takes difference The big direction of absolute value is the final interpolation direction of interpolation pixel, if the first differential value of first direction and its symmetry direction Difference absolute value be greater than second direction and its symmetry direction first differential value difference absolute value, then take first direction For interpolation direction, otherwise taking second direction is interpolation direction.
A kind of de-interlaced processing method of TV signal characterized by comprising
Judge whether interpolation pixel is in strong still-mode;
If so, the pixel value of pixel corresponding with the interpolation pixel in latter field is obtained, described in supplement The pixel value of interpolation pixel, the pixel value of the interpolation pixel are equal in latter field and the interpolation pixel phase The pixel value of corresponding pixel;Either, the picture of pixel corresponding with the interpolation pixel in previous field is obtained The pixel value of pixel corresponding with the interpolation pixel, supplements the interpolation pixel in element value and latter field The pixel value of the pixel value of point, the interpolation pixel is equal to pixel corresponding with the interpolation pixel in previous field Point pixel value, with the average value of both pixel values of pixel corresponding with the interpolation pixel in latter field;
If not, the gradient zone of multiple directions is divided centered on the interpolation pixel, it is true by the gradient zone The directionality of the fixed interpolation pixel supplements the interpolation picture by interpolation along the direction of the interpolation pixel The pixel value of vegetarian refreshments.
A kind of de-interlaced processing unit of TV signal characterized by comprising
Judgment module, for judging whether interpolation pixel is in strong still-mode;
First processing module, for when the interpolation pixel is in strong still-mode, obtain in latter field with institute The pixel value for stating the corresponding pixel of interpolation pixel supplements the pixel value of the interpolation pixel, the interpolation The pixel value of pixel is equal to the pixel value of pixel corresponding with the interpolation pixel in latter field;Either, it obtains Take in previous field in the pixel value of pixel corresponding with the interpolation pixel and latter field with the interpolation picture The pixel value of the corresponding pixel of vegetarian refreshments supplements the pixel value of the interpolation pixel, the picture of the interpolation pixel Element value be equal to previous field in pixel corresponding with the interpolation pixel pixel value, in latter field with it is described to be inserted It is worth the average value of both pixel values of the corresponding pixel of pixel;
Second processing module is used for when the interpolation pixel is in motor pattern, with the interpolation pixel Centered on, the gradient zone of multiple directions is divided, the directionality of the interpolation pixel is determined by the gradient zone, along described The direction of interpolation pixel supplements the pixel value of the interpolation pixel by interpolation.
It can be seen via above technical scheme that compared with prior art, the present invention provides a kind of TV signal de interlacings Treating method and apparatus.Using technical solution provided by the invention, first determine whether interpolation pixel is in strong static Mode;If interpolation pixel is in strong still-mode, obtain corresponding with the interpolation pixel the in previous field The pixel value of one pixel, supplements the pixel value of the interpolation pixel, and the pixel value of the interpolation pixel is equal to institute The pixel value of the first pixel is stated, so that the interpolation pixel be made to obtain accurate pixel value;If interpolation pixel In non-strong still-mode, i.e. motor pattern, then centered on the interpolation pixel, the gradient zone of multiple directions is divided, The precision level that can be improved detection interpolation pixel directionality, then determines the interpolation pixel by the gradient zone The directionality of point supplements the pixel value of the interpolation pixel by interpolation finally along the direction of the interpolation pixel, Based on more accurate interpolation pixel directionality, the pixel value of the interpolation pixel supplemented is also more accurate. Therefore, using technical solution provided by the invention, the accuracy of interpolation point pixel value can be effectively improved.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this The embodiment of invention for those of ordinary skill in the art without creative efforts, can also basis The attached drawing of offer obtains other attached drawings.
Fig. 1 is a kind of flow chart of the de-interlaced processing method of TV signal provided in an embodiment of the present invention;
Fig. 2 is the constraint scene figure that video flowing when strong still-mode is judged provided by the embodiment of the present invention;
Fig. 3 (a) is a kind of schematic diagram of gradient zone provided in an embodiment of the present invention;
Fig. 3 (b) is the schematic diagram of another gradient zone provided in an embodiment of the present invention;
Fig. 4 is the schematic diagram of the gradient zone after one kind provided in an embodiment of the present invention is down-sampled;
Fig. 5 is a kind of structure chart of the de-interlaced processing unit of TV signal provided in an embodiment of the present invention;
Fig. 6 (a) is schematic diagram before the processing provided in an embodiment of the present invention for strong still-mode;
Fig. 6 (b) is schematic diagram after the processing provided in an embodiment of the present invention for strong still-mode;
Fig. 7 (a) is schematic diagram before the processing provided in an embodiment of the present invention for moving scene mode;
Fig. 7 (b) is schematic diagram after the processing provided in an embodiment of the present invention for moving scene mode;
Fig. 8 is to judge the constraint scene figure of video flowing when strong still-mode (with Fig. 2 pairs provided by the embodiment of the present invention Than).
Specific embodiment
With reference to background technology part analysis it is found that at present in existing technology, be generally basede on known pixels point to infer Then the directionality of missing pixel point obtains the pixel value of missing pixel point with the known point on the direction inferred come interpolation.
The scheme of the prior art has the following deficiencies:
(1) when hardware resource is limited, for example the data for only having two rows in the vertical direction (push up field and bottom only have respectively Known to the pixel value of one-row pixels point) when, almost all of algorithm all cannot be guaranteed that walking direction does not have vicious possibility, and certain Even if a little details still cannot carry out good interpolation by walking direction when data are more than enough, because some details do not have very Specific directionality (such as certain corners of text).It is place that image in true the preceding paragraph video, which may have quite a few, In absolute rest state, or it is in relative static conditions (global motion).
(2) be on the judgement direction under moving scene de interlacing (deinterlace) algorithm difficult point, especially direction Property present low angle when.
For above-mentioned (1), the invention proposes the de interlacing algorithms of the detection based on strong static scene.
For above-mentioned (2), the invention proposes band-like gradient calculation method and its down-sampled simplification method, can with compared with Low computation complexity and Logical cost exchanges preferable performance for.
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other Embodiment shall fall within the protection scope of the present invention.
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, with reference to the accompanying drawing and specific real Applying mode, the present invention is described in further detail.
Embodiment one
Referring to Fig. 1, Fig. 1 is a kind of process of the de-interlaced processing method of TV signal provided in an embodiment of the present invention Figure.As shown in Figure 1, this method comprises:
Step S101, judges whether interpolation pixel is in strong still-mode;If so, step S102 is executed, if It is no, execute step S103;
It include at least two in the first window specifically, choosing first window centered on the interpolation pixel The pixel of row known pixel values successively judges pixel value between each pixel of every row and the pixel of former frame corresponding position The sum of absolute value of difference, if less than preset threshold respectively voluntarily;With centered on the second pixel, the second window is chosen, Include the pixel of at least a line known pixel values in second window, successively judges each pixel of every row and former frame pair Answer the sum of the absolute value of pixel value between the pixel of position, if less than preset threshold respectively voluntarily;Second picture Vegetarian refreshments is located at the opposite field that the first window is on the scene in same frame, and second pixel and the interpolation pixel exist Position in their own field is corresponding, and (in this example, the first pixel and the second pixel are same pixels, as Bmid0 in src_col).The preset threshold is related with noise level, is the multiple of noise level.
For example, referring to Fig. 2, Fig. 2 is the constraint for judging video flowing when strong still-mode provided by the embodiment of the present invention Scene figure.As shown in Fig. 2, src_col and src are front court and back court in same frame (being denoted as the second frame), pre_src_col with Pre_src is front court and back court in former frame (being denoted as first frame).Wherein, pre_src_col is the bottom in first frame, Pre_src is the top in first frame, and src_col is the bottom in the second frame, and src is the top in the second frame.
, both can be according to the sequential scan at bottom behind first top field when scanning, it can also be according to the sequence for pushing up field behind first bottom Scanning.After first to push up field for the sequence at bottom, then the sequence scanned in Fig. 2 (and Fig. 8) is successively pre_src --- Pre_src_col --- src --- src_col.In another embodiment, if first to push up the sequence at bottom behind field Scanning, the then sequence scanned in Fig. 2 (and Fig. 8) --- pre_src --- src_col that are successively pre_src_col --- src.
Optionally, first window A is the window of 3X3 (3 rows 3 column) size, and central point, i.e. interpolation point are denoted as Acenter;With First window A is corresponding, and the second window B is also that (but operation result and the second window B are used for the window of 3X3 (3 rows 3 column) size The operation result of the window of 1X3 size is identical), central point is denoted as Bcenter(as mid0);In Fig. 2 (and Fig. 8), dot " " and difference "×" indicate pixel, wherein the pixel value of pixel represented by dot " " is it is known that and difference "×" institute table Show that the pixel value of pixel is unknown.First window A and the second window B is strong constraint.
It is to be understood that in the present embodiment by taking the columns of window is 3 and 7 (respectively strong constraints and weak constraint) as an example, But the columns of window is not limited by 3 and 7, and in specific implementation, the columns of window can adjust according to the actual situation.One In a preferred embodiment, the columns of window may be greater than or the odd number (bilateral symmetry) equal to 3, uses strong constraint at the same time In the case where weak constraint, the columns of strong constraint window is less than the columns of weak constraint window (for example, the columns of strong constraint window is 3, the columns of weak constraint window is 7, and for another example, the columns of strong constraint window is 5,11) columns of weak constraint window is.
The pixel adjacent with interpolation pixel is contained in first window A and the second window B.Wherein, first window A In Aup row and Adown row and interpolation pixel be expert at (i.e. Amid row) it is adjacent in position, in the second window B Bmid row (and pixel of the corresponding position in the bottom (i.e. pre_src_col) of former frame of the Bmid row in the second window) with Interpolation pixel be expert at (i.e. Amid row) it is adjacent in time (as previously mentioned, when scanning, first to push up the sequence at bottom behind field For, then the sequence scanned in Fig. 2 (and Fig. 8) --- pre_src_col --- src --- src_ that is successively pre_src Col, it is seen that src_col and pre_src_col adjacent in time with src).
It is understood that although the second window B in attached drawing 2 contains 3 rows, due in Bup row and Bdown row Pixel be it is unknown, therefore, judging interpolation pixel AcenterDuring whether being in strong still-mode, the second window The pixel of actual participation operation is the pixel in Bmid row in mouth B.It please also refer to Fig. 2 and Fig. 8, in the above-described example, The second window B in src_col uses the window (as shown in Figure 2) of 3X3 (3 rows 3 column), and the window using 1X3 (1 row 3 column) (as shown in Figure 8), the operation result obtained in actual operation are identical.
It is read incorporated by reference to Fig. 2 and following formula, for strong constraint (i.e. the in first window A and src_col in src Two window B):
Wherein, up0 AcenterSurface pixel Aup0Pixel value, up_m1 be pixel Aup0Adjacent first left The pixel value of pixel, up_p1 are pixel Aup0The pixel value of adjacent first pixel in the right;It is corresponding, pre_up0 For in former frame with pixel Aup0Corresponding pixel (i.e. the first pixel) Apre_up0Pixel value, pre_up_m1 is picture Vegetarian refreshments Apre_up0The pixel value of adjacent first left pixel, pre_up_p1 are pixel Apre_up0First, adjacent the right picture The pixel value of vegetarian refreshments;NL_up is pixel Aup0Noise level of the row, Coffe1 are constant.
Down0 is AcenterUnderface pixel Adown0Pixel value, down_m1 be pixel Adown0The adjacent left side first The pixel value of a pixel, down_p1 are pixel Adown0The pixel value of adjacent first pixel in the right;It is corresponding, Pre_down0 be former frame in pixel Adown0Corresponding pixel (i.e. the first pixel) Apre_down0Pixel value, Pre_down_m1 is pixel Apre_down0The pixel value of adjacent first left pixel, pre_down_p1 are pixel Apre_down0The pixel value of adjacent first pixel in the right.NL_down is pixel Adown0Noise level of the row.
Please also refer to Fig. 2 and Fig. 8, due to the pixel in Bup row and Bdown row be it is unknown, judging interpolation Pixel AcenterDuring whether being in strong still-mode, the pixel of actual participation operation is Bmid row in the second window B In pixel.
Wherein, the second pixel BcenterPositioned at center (as mid0 namely the B of the second windowmid0), mid_m1 For pixel Bmid0The pixel value of adjacent first left pixel, mid_p1 are pixel Bmid0First, adjacent the right pixel The pixel value of point;It is corresponding, pre_mid0 be former frame in pixel Bmid0Corresponding pixel Bpre_mid0(as Pre_mid_0 pixel value), pre_mid_m1 are pixel Bpre_mid0The pixel value of adjacent first left pixel, pre_ Mid_p1 is pixel Bpre_mid0The pixel value of adjacent first pixel in the right.NL_mid is pixel Bmid0It is of the row to make an uproar Sound is horizontal.
Optionally, technical solution provided by the invention further includes weak constraint to increase the accuracy of judgement.I.e. with institute Centered on stating interpolation pixel, third window is chosen, includes the pixel of at least two row known pixel values in the third window Point successively judges the sum of the absolute value of pixel value between each pixel of every row and the pixel of former frame corresponding position, Whether each preset threshold voluntarily is less than;The row of the line number of pixel and pixel in the first window in the third window Count the columns that equal, columns is greater than pixel in the first window;With centered on second pixel, the 4th window is chosen Mouthful, include the pixel of at least a line known pixel values in the 4th window, successively judge each pixel of every row with it is previous The sum of the absolute value of pixel value between the pixel of frame corresponding position, if less than preset threshold respectively voluntarily.Similarly may be used Know, the line number of pixel is equal with the line number of pixel in second window in the 4th window, columns is greater than described the The columns of pixel in two windows.
As shown in Fig. 2, third window C and the 4th window D is weak constraint.Third window C is the window of 3X7 (3 rows 7 column) size Mouthful, the 4th window D is also that (but operation result and the 4th window D are using the window of 1X7 size for the window of 3X7 (3 rows 7 column) size Operation result it is identical).It is read incorporated by reference to Fig. 2 and following formula, for weak constraint (i.e. third window C and src_ in src The 4th window D in col):
Similarly, Fig. 2 and Fig. 8 be please also refer to, due to the pixel in Bup row and Bdown row be it is unknown, judge to Interpolating pixel point AcenterDuring whether being in strong still-mode, the pixel of actual participation operation is in the 4th window D Pixel in Dmid row.
Wherein, the meaning of specific parameters please refers to above-mentioned strong constraint part, and details are not described herein again, and Coffe2 is normal Number.The Coffe 2 of weak constraint is generally less than directed to for the Coffe1 of strong constraint.
Since noise level is different under different luminance levels, so the noise level at different location may Different (i.e. each row can use different NL).
It should be noted that embodiments of the present invention offer judges whether interpolation pixel is in strong still-mode Method be only a kind of optional that is, by the way of window strong constraint, or in such a way that strong constraint and weak constraint combine Embodiment, technical solution provided by the invention are not limited to aforesaid way.
In specific implementation, the quantity of window can be one or more, such as can be two windows (one of them is Strong constraint, the other is weak constraint);The size of window can adjust according to the actual situation, and columns usually may be greater than equal to 3 Odd number;When the center of window is interpolation pixel (pixel of window center is unknown), the line number of the window be greater than Or the odd number (pixel including at least two row known pixel values) equal to 3, when known to the pixel of window center, the window Line number be about or equal to 1 odd number (pixel including at least a line known pixel values).
It, " should be to when judging that obtained result is according to each window in the case where the quantity of window is multiple Interpolating pixel point is in strong stationary state " when, this thinks that the interpolation pixel is in strong stationary state;When there are one or more The judgement of a window obtained the result is that " the interpolation pixel is not at strong stationary state ", then it is assumed that the interpolation pixel It is not at strong stationary state.
Step S102 obtains the pixel value of the first pixel corresponding with the interpolation pixel in previous field, mends The pixel value of the interpolation pixel is filled, the pixel value of the interpolation pixel is equal to the pixel of first pixel Value;
Specifically, obtaining the pixel of the first pixel corresponding with the interpolation pixel in same frame previous field Value, the pixel value of the interpolation pixel are equal to the pixel value of first pixel, supplement the interpolation pixel Pixel value.
In other embodiments, also in available latter field pixel corresponding with the interpolation pixel picture Element value, to supplement the pixel value of the interpolation pixel, the pixel value of the interpolation pixel is equal in latter field and institute State the pixel value of the corresponding pixel of interpolation pixel;Either, also in available previous field with the interpolation picture The pixel of pixel corresponding with the interpolation pixel in the pixel value of the corresponding pixel of vegetarian refreshments and latter field Value, to supplement the pixel value of the interpolation pixel, the pixel value of the interpolation pixel be equal in previous field with it is described The pixel value of the corresponding pixel of interpolation pixel, with pixel corresponding with the interpolation pixel in latter field Both pixel values average value.
Step S103 divides the gradient zone of multiple directions centered on the interpolation pixel, passes through the gradient zone The directionality for determining the interpolation pixel supplements the interpolation by interpolation along the direction of the interpolation pixel The pixel value of pixel.
Specifically, optional, the gradient zone for dividing multiple directions, comprising: divide the gradient zone in 3 directions.
For example, please referring to Fig. 3 (a) and Fig. 3 (b), Fig. 3 (a) is a kind of signal of gradient zone provided in an embodiment of the present invention Figure, Fig. 3 (b) are the schematic diagram of another gradient zone provided in an embodiment of the present invention.Fig. 3 (a) and Fig. 3 (b) are combined, then institute There are 3, including vertical direction, up-1 pixel (pixel where up row -1) and down1 picture in the direction of the gradient zone of division Rectilinear direction where vegetarian refreshments (pixel where down row 1)AndThese three Direction.As shown in Fig. 3 (a) and Fig. 3 (b), central point Ccenter.It is vertical when only considering, When direction, need band-like gradient algorithm such as Fig. 3 (a) of 5 lines shown, by analysis, such band-like energy It is enough to guarantee the detection even the direction of only 1 pixel wide.Wherein every line indicates the absolute value of direction first differential, That is the absolute value for the two neighboring pixel pixel value that every line is passed through.
Specifically, determining the directionality of the interpolation pixel by the gradient zone, comprising:
First differential value after comparing each direction gradient band normalization, taking the smallest direction of differential value is first direction, Secondary small direction is second direction.If first differential value and second direction gradient zone normalizing after the normalization of first direction gradient zone The absolute value of the difference of first differential value after change is greater than default value, it is determined that the first direction is the interpolation pixel The direction of point.
Specifically, such as vertical direction, the first differential value after normalization are as follows:
[(up-1)-(down-1)|+|(up0)-(down0)|+|(up1)-(down1)]/3;
AndFirst differential value after the gradient zone normalization in the two directions Calculation method it is identical as vertical direction.
If the first differential value after the normalization of first direction gradient zone and the single order after the normalization of second direction gradient zone are micro- The absolute value of the difference of score value is less than default value, then chooses the ladder of the symmetry direction in each direction in the first and second directions Degree band (symmetrically refers to symmetrical about vertical direction), is respectively compared the first differential in described two directions Yu its symmetry direction gradient zone The absolute value of the difference of value, the direction for taking absolute difference big are the final interpolation direction of interpolation pixel.If first direction It is greater than second direction and the first differential value of its symmetry direction with the absolute value of the difference of the first differential value of its symmetry direction The absolute value of difference, then taking first direction is interpolation direction, and otherwise taking second direction is interpolation direction.
For example,WithIt is all smaller than other directions, andWithGap is little, butWithGap Δ 1 compare WithGap Δ 2 it is much smaller, then direction is at this time
After determining the directionality of the interpolation pixel (such as direction is), along the direction into Row interpolation: Intp_value=(up-1_value+down1_value)/2;
Wherein, Intp_valueIndicate the pre-selection pixel value of the interpolation pixel, up-1_valuePicture where indicating up-1 The pixel value of vegetarian refreshments, down1_valueThe pixel value of pixel where indicating down1.
The pixel value of the interpolation pixel finally supplemented optionally passes through the pre- of the interpolation pixel Select pixel value Intp_valueIt is obtained with the data fusion of previous field, the condition of fusion is equally to consider movement and static opposite pass System, such as in conjunction with up0_value-pre_up0_valueAnd down0_value–pre_down0_valueAnd (up0_value+down0_value)/ 2–mid0_valueThree differences carry out final distribution coefficient and merge (up0_value+down0_value)/2 and Intp_valueIt finally obtains scarce The data of mistake.Alternatively, the pre-selection pixel value Intp for passing through the interpolation pixel_valueAnd the replacement under strong still-mode The pixel value for the interpolation pixel for obtaining finally being supplemented is added after value distribution weight.
It should be noted that in the step S103, it is described that the interpolation pixel is determined by the gradient zone Before directionality, further includes:
Down-sampled processing is made to the gradient zone in the multiple direction, the gradient zone for the multiple directions being simplified.
Specifically, down-sampled method is not that uniquely, the direction that can retain after direction is down-sampled after becoming more is more, it is The correctness in direction is found in guarantee, and the present invention points out the critical key point for wanting band-like gradient as far as possible to include other directions, and It must include the critical key point in this direction.
So-called critical key point refers to: for indicating the scanned point in direction.Such as up-1 and down1 are directionsCritical key point.
As shown in figure 4, Fig. 4 is the schematic diagram of the gradient zone after one kind provided in an embodiment of the present invention is down-sampled.Specifically, Fig. 4 is the schematic diagram of the gradient zone to Fig. 3 (a) after down-sampled.To the schematic diagram and Fig. 3 of gradient zone of the Fig. 3 (b) after down-sampled (b) identical.Down-sampled processing is used to gradient zone, the complexity of calculating can be simplified.
It can be seen via above technical scheme that the embodiment of the invention provides a kind of de-interlaced processing methods of TV signal And device.Using technical solution provided in an embodiment of the present invention, first determine whether interpolation pixel is in strong still-mode; If interpolation pixel is in strong still-mode, the first pixel corresponding with the interpolation pixel in previous field is obtained The pixel value of point, supplements the pixel value of the interpolation pixel, and the pixel value of the interpolation pixel is equal to described first The pixel value of pixel, so that the interpolation pixel be made to obtain accurate pixel value;If interpolation pixel is in non- Strong still-mode, i.e. motor pattern divide the gradient zone of multiple directions, Neng Gouti then centered on the interpolation pixel The precision level of high detection interpolation pixel directionality, then determines the side of the interpolation pixel by the gradient zone Tropism supplements the pixel value of the interpolation pixel by interpolation finally along the direction of the interpolation pixel, based on more Add accurate interpolation pixel directionality, the pixel value of the interpolation pixel supplemented is also more accurate.Therefore, it answers With technical solution provided by the invention, the accuracy of interpolation point pixel value can be effectively improved.
Embodiment two
In order to more comprehensively illustrate technical solution of the present invention, the invention also discloses a kind of TV signal is de-interlaced Processing unit.
Referring to Fig. 5, Fig. 5 is a kind of structure of the de-interlaced processing unit of TV signal provided in an embodiment of the present invention Figure.As shown in figure 5, the device includes:
Judgment module 501, for judging whether interpolation pixel is in strong still-mode;
First processing module 502, for when the interpolation pixel be in strong still-mode, obtain in previous field and The pixel value of corresponding first pixel of the interpolation pixel, supplements the pixel value of the interpolation pixel, described The pixel value of interpolation pixel is equal to the pixel value of first pixel;
Second processing module 503 is used for when the interpolation pixel is in motor pattern, with the interpolation pixel Centered on point, the gradient zone of multiple directions is divided, the directionality of the interpolation pixel is determined by the gradient zone, along institute The direction for stating interpolation pixel supplements the pixel value of the interpolation pixel by interpolation.
In other embodiments, the first processing module also in available latter field with the interpolation pixel phase The pixel value of corresponding pixel, to supplement the pixel value of the interpolation pixel, the pixel value of the interpolation pixel Equal to the pixel value of pixel corresponding with the interpolation pixel in latter field;Either, also available previous field In pixel corresponding with the interpolation pixel pixel value and latter field in it is opposite with the interpolation pixel The pixel value for the pixel answered, to supplement the pixel value of the interpolation pixel, the pixel value etc. of the interpolation pixel In previous field the pixel value of pixel corresponding with the interpolation pixel, with latter field in the interpolation pixel The average value of both pixel values of the corresponding pixel of point.
Specifically, the judgment module includes:
First judging unit is used for centered on the interpolation pixel, selection first window, in the first window Pixel including at least two row known pixel values, successively judges the pixel of every row each pixel and former frame corresponding position Between pixel value the sum of absolute value, if less than preset threshold respectively voluntarily;With centered on the second pixel, choose Second window, the interior pixel including at least a line known pixel values of second window, successively judges each pixel of every row The sum of the absolute value of pixel value between the pixel of former frame corresponding position, if less than preset threshold respectively voluntarily; Second pixel is located at the opposite field that the first window is on the scene in same frame, second pixel and described to be inserted It is corresponding to be worth position of the pixel in their own field.
Further, the judgment module further include:
Second judgment unit is used for centered on the interpolation pixel, selection third window, in the third window Pixel including at least two row known pixel values, successively judges the pixel of every row each pixel and former frame corresponding position Between pixel value the sum of absolute value, if less than preset threshold respectively voluntarily;The row of pixel in the third window Count the columns that, columns equal with the line number of pixel in the first window is greater than pixel in the first window;With with institute Centered on stating the second pixel, the 4th window is chosen, includes the pixel of at least a line known pixel values in the 4th window, Successively judge the sum of the absolute value of pixel value between each pixel of every row and the pixel of former frame corresponding position, if Less than preset threshold respectively voluntarily.
In specific implementation, window strong constraint can be used, or in such a way that strong constraint and weak constraint combine, i.e., The quantity of window can be one or more, for example, can be two windows (one of them is strong constraint, the other is it is weak about Beam);The size of window can adjust according to the actual situation, and columns usually may be greater than the odd number equal to 3;When the center of window When for interpolation pixel (pixel of window center is unknown), the line number of the window be odd number more than or equal to 3 (including extremely The pixel of few two row known pixel values), when known to the pixel of window center, the line number of the window is about or equal to 1 Odd number (pixel including at least a line known pixel values).
It, " should be to when judging that obtained result is according to each window in the case where the quantity of window is multiple Interpolating pixel point is in strong stationary state " when, this thinks that the interpolation pixel is in strong stationary state;When there are one or more The judgement of a window obtained the result is that " the interpolation pixel is not at strong stationary state ", then it is assumed that the interpolation pixel It is not at strong stationary state.
Specifically, the Second processing module includes:
Gradient zone division unit is used for when the interpolation pixel is in motor pattern, with the interpolation pixel Centered on point, the gradient zone of multiple directions is divided;
Direction calculating unit, for determining the directionality of the interpolation pixel by the gradient zone;
Interpolating unit, for supplementing the interpolation pixel by interpolation along the direction of the interpolation pixel Pixel value.
Further, the Second processing module further include:
Down-sampled processing unit makees down-sampled processing for the gradient zone to the multiple direction, and what is be simplified is multiple The gradient zone in direction.
Further, the directionality of the interpolation pixel is determined described in Second processing module by the gradient zone, Include:
First differential value after comparing each direction gradient band normalization, taking the smallest direction of differential value is first direction, Secondary small direction is second direction, if first differential value and second direction gradient zone normalizing after the normalization of first direction gradient zone The absolute value of the difference of first differential value after change is greater than default value, it is determined that the first direction is the interpolation pixel The direction of point.
Further, the directionality of the interpolation pixel is determined described in Second processing module by the gradient zone, Further include:
If the first differential value after the normalization of first direction gradient zone and the single order after the normalization of second direction gradient zone are micro- The absolute value of the difference of score value is less than default value, then chooses the ladder of the symmetry direction in each direction in the first and second directions Band is spent, the absolute value of the difference of described two directions and the first differential value of its symmetry direction gradient zone is respectively compared, takes difference The big direction of absolute value is the final interpolation direction of interpolation pixel, if the first differential value of first direction and its symmetry direction Difference absolute value be greater than second direction and its symmetry direction first differential value difference absolute value, then take first direction For interpolation direction, otherwise taking second direction is interpolation direction.
Part effect picture is also disclosed in the embodiment of the present invention.Fig. 6 (a) and Fig. 6 (b) is provided in an embodiment of the present invention for strong The comparison diagram before and after the processing of still-mode, wherein Fig. 6 (a) is untreated, and Fig. 6 (b) is the present invention at strong still-mode Effect after reason.Fig. 7 (a) and Fig. 7 (b) is the comparison provided in an embodiment of the present invention before and after the processing for moving scene mode Figure, wherein Fig. 7 (a) is untreated, and Fig. 7 (b) is the present invention for the effect after moving scene mode treatment.Specifically, figure 7 (b) handle Fig. 7 (a) using band-like gradient algorithm.
Finally, it is to be noted that, herein, relational terms such as first and second and the like be used merely to by One entity or operation are distinguished with another entity or operation, without necessarily requiring or implying these entities or operation Between there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant meaning Covering non-exclusive inclusion, so that the process, method, article or equipment for including a series of elements not only includes that A little elements, but also including other elements that are not explicitly listed, or further include for this process, method, article or The intrinsic element of equipment.In the absence of more restrictions, the element limited by sentence "including a ...", is not arranged Except there is also other identical elements in the process, method, article or apparatus that includes the element.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with other The difference of embodiment, the same or similar parts in each embodiment may refer to each other.For device disclosed in embodiment For, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, related place is said referring to method part It is bright.
The step of method described in conjunction with the examples disclosed in this document or algorithm, can directly be held with hardware, processor The combination of capable software module or the two is implemented.Software module can be placed in random access memory (RAM), memory, read-only deposit Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology In any other form of storage medium well known in field.
The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, as defined herein General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, of the invention It is not intended to be limited to the embodiments shown herein, and is to fit to and the principles and novel features disclosed herein phase one The widest scope of cause.

Claims (14)

1. a kind of de-interlaced processing method of TV signal characterized by comprising
Judge whether interpolation pixel is in strong still-mode;
If so, the pixel value of the first pixel corresponding with the interpolation pixel in previous field is obtained, described in supplement The pixel value of interpolation pixel, the pixel value of the interpolation pixel are equal to the pixel value of first pixel;
If not, dividing the gradient zone of multiple directions centered on the interpolation pixel, institute is determined by the gradient zone The directionality for stating interpolation pixel supplements the interpolation pixel by interpolation along the direction of the interpolation pixel Pixel value;
It is wherein, described to judge whether interpolation pixel is in strong still-mode, comprising:
Centered on the interpolation pixel, first window is chosen, includes at least two row known pixels in the first window The pixel of value successively judges the absolute of pixel value between each pixel of every row and the pixel of former frame corresponding position The sum of value, if less than preset threshold respectively voluntarily;
With centered on the second pixel, the second window is chosen, includes at least a line known pixel values in second window Pixel, successively judge pixel value between each pixel of every row and the pixel of former frame corresponding position absolute value it With, if less than preset threshold respectively voluntarily;Second pixel is located at the phase that the first window is on the scene in same frame Adjacent field, the position of second pixel and the interpolation pixel in their own field are corresponding.
2. judging whether interpolation pixel is in strong stationary mold the method according to claim 1, wherein described Formula, further includes:
Centered on the interpolation pixel, third window is chosen, includes at least two row known pixels in the third window The pixel of value successively judges the absolute of pixel value between each pixel of every row and the pixel of former frame corresponding position The sum of value, if less than preset threshold respectively voluntarily;The line number of pixel and picture in the first window in the third window The line number of vegetarian refreshments is equal, columns is greater than the columns of pixel in the first window;
With centered on second pixel, the 4th window is chosen, includes at least a line known pixels in the 4th window The pixel of value successively judges the absolute of pixel value between each pixel of every row and the pixel of former frame corresponding position The sum of value, if less than preset threshold respectively voluntarily.
3. the method according to claim 1, wherein the gradient zone for dividing multiple directions, comprising:
Divide the gradient zone in 3 directions.
4. the method according to claim 1, wherein described determine the interpolation pixel by the gradient zone The directionality of point, comprising:
First differential value after comparing each direction gradient band normalization, taking the smallest direction of differential value is first direction, secondary small Direction be second direction, if first direction gradient zone normalization after first differential value and second direction gradient zone normalization after The absolute value of difference of first differential value be greater than default value, it is determined that the first direction is the interpolation pixel Direction.
5. according to the method described in claim 4, it is characterized in that, described determine the interpolation pixel by the gradient zone The directionality of point, further includes:
If the first differential value after first differential value and the normalization of second direction gradient zone after the normalization of first direction gradient zone Difference absolute value be less than default value, then choose the gradient of the symmetry direction in each direction in the first and second directions Band is respectively compared the absolute value of the difference of described two directions and the first differential value of its symmetry direction gradient zone, takes difference exhausted To the final interpolation direction that big direction is interpolation pixel is worth, if first direction and the first differential value of its symmetry direction The absolute value of difference is greater than the absolute value of the difference of the first differential value of second direction and its symmetry direction, then takes the first direction to be Interpolation direction, otherwise taking second direction is interpolation direction.
6. the method according to claim 1, wherein described determine the interpolation pixel by the gradient zone Before the directionality of point, further includes:
Down-sampled processing is made to the gradient zone in the multiple direction, the gradient zone for the multiple directions being simplified.
7. a kind of de-interlaced processing unit of TV signal characterized by comprising
Judgment module, for judging whether interpolation pixel is in strong still-mode;
First processing module, for when the interpolation pixel be in strong still-mode, in acquisition previous field with it is described to The pixel value of corresponding first pixel of interpolating pixel point, supplements the pixel value of the interpolation pixel, the interpolation The pixel value of pixel is equal to the pixel value of first pixel;
Second processing module with the interpolation pixel is for when the interpolation pixel is in motor pattern The heart divides the gradient zone of multiple directions, the directionality of the interpolation pixel is determined by the gradient zone, along described to be inserted It is worth the direction of pixel, the pixel value of the interpolation pixel is supplemented by interpolation;
Wherein, the judgment module includes:
First judging unit is used for centered on the interpolation pixel, is chosen first window, is included in the first window At least pixel of two row known pixel values successively judges between each pixel of every row and the pixel of former frame corresponding position The sum of absolute value of pixel value, if less than preset threshold respectively voluntarily;With centered on the second pixel, second is chosen Window, the interior pixel including at least a line known pixel values of second window, successively judges each pixel of every row with before The sum of the absolute value of pixel value between the pixel of one frame corresponding position, if less than preset threshold respectively voluntarily;It is described Second pixel is located at the opposite field that the first window is on the scene in same frame, second pixel and the interpolation picture Position of the vegetarian refreshments in their own field is corresponding.
8. device according to claim 7, which is characterized in that the judgment module further include:
Second judgment unit is used for centered on the interpolation pixel, is chosen third window, is included in the third window At least pixel of two row known pixel values successively judges between each pixel of every row and the pixel of former frame corresponding position The sum of absolute value of pixel value, if less than preset threshold respectively voluntarily;In the third window line number of pixel with The columns that the line number of pixel is equal in the first window, columns is greater than pixel in the first window;With with described Centered on two pixels, the 4th window is chosen, includes the pixel of at least a line known pixel values in the 4th window, successively Judge the sum of the absolute value of pixel value between each pixel of every row and the pixel of former frame corresponding position, if be less than Respectively preset threshold voluntarily.
9. device according to claim 7, which is characterized in that the Second processing module includes:
Gradient zone division unit, for being with the interpolation pixel when the interpolation pixel is in motor pattern Center divides the gradient zone of multiple directions;
Direction calculating unit, for determining the directionality of the interpolation pixel by the gradient zone;
Interpolating unit, for supplementing the pixel of the interpolation pixel by interpolation along the direction of the interpolation pixel Value.
10. device according to claim 9, which is characterized in that the Second processing module further include:
Down-sampled processing unit makees down-sampled processing for the gradient zone to the multiple direction, the multiple directions being simplified Gradient zone.
11. device according to claim 7, which is characterized in that determined described in Second processing module by the gradient zone The directionality of the interpolation pixel, comprising:
First differential value after comparing each direction gradient band normalization, taking the smallest direction of differential value is first direction, secondary small Direction be second direction, if first direction gradient zone normalization after first differential value and second direction gradient zone normalization after The absolute value of difference of first differential value be greater than default value, it is determined that the first direction is the interpolation pixel Direction.
12. device according to claim 11, which is characterized in that determined described in Second processing module by the gradient zone The directionality of the interpolation pixel, further includes:
If the first differential value after first differential value and the normalization of second direction gradient zone after the normalization of first direction gradient zone Difference absolute value be less than default value, then choose the gradient of the symmetry direction in each direction in the first and second directions Band is respectively compared the absolute value of the difference of described two directions and the first differential value of its symmetry direction gradient zone, takes difference exhausted To the final interpolation direction that big direction is interpolation pixel is worth, if first direction and the first differential value of its symmetry direction The absolute value of difference is greater than the absolute value of the difference of the first differential value of second direction and its symmetry direction, then takes the first direction to be Interpolation direction, otherwise taking second direction is interpolation direction.
13. a kind of de-interlaced processing method of TV signal characterized by comprising
Judge whether interpolation pixel is in strong still-mode;
If so, obtaining the pixel value of pixel corresponding with the interpolation pixel in latter field, supplement described to be inserted It is worth the pixel value of pixel, the pixel value of the interpolation pixel is equal to corresponding with the interpolation pixel in latter field Pixel pixel value;Either, obtain previous field in pixel corresponding with the interpolation pixel pixel value, And in latter field pixel corresponding with the interpolation pixel pixel value, supplement the picture of the interpolation pixel Element value, the pixel value of the interpolation pixel are equal to the picture of pixel corresponding with the interpolation pixel in previous field Element value, with the average value of both pixel values of pixel corresponding with the interpolation pixel in latter field;
If not, dividing the gradient zone of multiple directions centered on the interpolation pixel, institute is determined by the gradient zone The directionality for stating interpolation pixel supplements the interpolation pixel by interpolation along the direction of the interpolation pixel Pixel value;
It is wherein, described to judge whether interpolation pixel is in strong still-mode, comprising:
Centered on the interpolation pixel, first window is chosen, includes at least two row known pixels in the first window The pixel of value successively judges the absolute of pixel value between each pixel of every row and the pixel of former frame corresponding position The sum of value, if less than preset threshold respectively voluntarily;
With centered on the second pixel, the second window is chosen, includes at least a line known pixel values in second window Pixel, successively judge pixel value between each pixel of every row and the pixel of former frame corresponding position absolute value it With, if less than preset threshold respectively voluntarily;Second pixel is located at the phase that the first window is on the scene in same frame Adjacent field, the position of second pixel and the interpolation pixel in their own field are corresponding.
14. a kind of de-interlaced processing unit of TV signal characterized by comprising
Judgment module, for judging whether interpolation pixel is in strong still-mode;
First processing module, for when the interpolation pixel is in strong still-mode, obtain in latter field with it is described to The pixel value of the corresponding pixel of interpolating pixel point supplements the pixel value of the interpolation pixel, the interpolation pixel The pixel value of point is equal to the pixel value of pixel corresponding with the interpolation pixel in latter field;Either, before acquisition In one in the pixel value of pixel corresponding with the interpolation pixel and latter field with the interpolation pixel The pixel value of corresponding pixel supplements the pixel value of the interpolation pixel, the pixel value of the interpolation pixel Equal to pixel corresponding with the interpolation pixel in previous field pixel value, with latter field in the interpolation picture The average value of both pixel values of the corresponding pixel of vegetarian refreshments;
Second processing module with the interpolation pixel is for when the interpolation pixel is in motor pattern The heart divides the gradient zone of multiple directions, the directionality of the interpolation pixel is determined by the gradient zone, along described to be inserted It is worth the direction of pixel, the pixel value of the interpolation pixel is supplemented by interpolation;
It is wherein, described to judge whether interpolation pixel is in strong still-mode, comprising:
Centered on the interpolation pixel, first window is chosen, includes at least two row known pixels in the first window The pixel of value successively judges the absolute of pixel value between each pixel of every row and the pixel of former frame corresponding position The sum of value, if less than preset threshold respectively voluntarily;
With centered on the second pixel, the second window is chosen, includes at least a line known pixel values in second window Pixel, successively judge pixel value between each pixel of every row and the pixel of former frame corresponding position absolute value it With, if less than preset threshold respectively voluntarily;Second pixel is located at the phase that the first window is on the scene in same frame Adjacent field, the position of second pixel and the interpolation pixel in their own field are corresponding.
CN201511025765.1A 2014-12-31 2015-12-30 The de-interlaced treating method and apparatus of TV signal Active CN105744201B (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201511025765.1A CN105744201B (en) 2014-12-31 2015-12-30 The de-interlaced treating method and apparatus of TV signal
US15/234,356 US9819900B2 (en) 2015-12-30 2016-08-11 Method and apparatus for de-interlacing television signal

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
CN201410854341 2014-12-31
CN2014108543415 2014-12-31
CN201511025765.1A CN105744201B (en) 2014-12-31 2015-12-30 The de-interlaced treating method and apparatus of TV signal

Publications (2)

Publication Number Publication Date
CN105744201A CN105744201A (en) 2016-07-06
CN105744201B true CN105744201B (en) 2019-01-18

Family

ID=56296120

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201511025765.1A Active CN105744201B (en) 2014-12-31 2015-12-30 The de-interlaced treating method and apparatus of TV signal

Country Status (1)

Country Link
CN (1) CN105744201B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101600061A (en) * 2009-07-09 2009-12-09 杭州士兰微电子股份有限公司 De-interlaced method of video motion-adaptive and device
CN101640783A (en) * 2008-07-30 2010-02-03 展讯通信(上海)有限公司 De-interlacing method and de-interlacing device for interpolating pixel points
CN103369208A (en) * 2013-07-15 2013-10-23 青岛海信信芯科技有限公司 Self-adaptive de-interlacing method and device
CN104202555A (en) * 2014-09-29 2014-12-10 建荣集成电路科技(珠海)有限公司 Method and device for deinterlacing

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101127220B1 (en) * 2004-07-28 2012-04-12 세종대학교산학협력단 Apparatus for motion compensation-adaptive de-interlacing and method the same

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101640783A (en) * 2008-07-30 2010-02-03 展讯通信(上海)有限公司 De-interlacing method and de-interlacing device for interpolating pixel points
CN101600061A (en) * 2009-07-09 2009-12-09 杭州士兰微电子股份有限公司 De-interlaced method of video motion-adaptive and device
CN103369208A (en) * 2013-07-15 2013-10-23 青岛海信信芯科技有限公司 Self-adaptive de-interlacing method and device
CN104202555A (en) * 2014-09-29 2014-12-10 建荣集成电路科技(珠海)有限公司 Method and device for deinterlacing

Also Published As

Publication number Publication date
CN105744201A (en) 2016-07-06

Similar Documents

Publication Publication Date Title
CN102036091B (en) Method of and apparatus for processing image data for display by a multiple-view display device
US7701508B2 (en) Accurate motion detection for the combination of motion adaptive and motion compensation de-interlacing applications
EP1313310A2 (en) Method of low latency interlace to progressive video format conversion
CN1574951A (en) Apparatus and method for deinterlace video signal
JP2006203884A (en) Method for edge-based pixel location, and interpolation
CN104219532B (en) The method and apparatus for determining interpolation frame between the method in wisp region, frame of video
CN101860660B (en) Frame rate conversion apparatus and frame rate conversion method
KR100506086B1 (en) Apparatus and Method for enhancing the quality of reproducing image
KR100563023B1 (en) Method and system for edge-adaptive interpolation for interlace-to-progressive conversion
US6812936B2 (en) Method of and unit for displaying an image in sub-fields
CN110072141A (en) A kind of media processing method, device, equipment and storage medium
US8675080B2 (en) Motion estimation in imaging systems
CN1231055C (en) Digital detection method for transmitting movie scene by TV
CN105744201B (en) The de-interlaced treating method and apparatus of TV signal
CN110418081A (en) High dynamic range images full resolution method for reconstructing, device and electronic equipment
CN107306346B (en) Image processing method and device, player, electronic equipment
EP1267575B1 (en) A method for converting the scanning format of images, a system and computer program product therefor
US9819900B2 (en) Method and apparatus for de-interlacing television signal
US8189874B2 (en) Method for motion detection of horizontal line
CN102780900B (en) Image display method of multi-person multi-view stereoscopic display
KR20100044977A (en) Method for video interpolation and computer recordable medium storing the method
CN108712642B (en) Automatic selection method for adding position of three-dimensional subtitle suitable for three-dimensional video
CN106131528A (en) 3D video format recognition methods and system
SG188546A1 (en) A method, an apparatus and a computer program product for deinterlacing an image having a plurality of pixels
US7893994B2 (en) Motion adaptive directional deinterlacing in a video decoder

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant