CN103826134B

CN103826134B - Image intra prediction method and device

Info

Publication number: CN103826134B
Application number: CN201410111041.8A
Authority: CN
Inventors: 朱策; 张玉花; 林永兵; 王昕�
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2014-03-21
Filing date: 2014-03-21
Publication date: 2017-08-18
Anticipated expiration: 2034-03-21
Also published as: CN103826134A

Abstract

The embodiment of the present invention provides a kind of image intra prediction method and device, and this method includes：It is the first subregion and the second subregion by present encoding depth dividing elements；According to the distribution of pixel in first subregion and the second subregion, determine that the first pixel determines the CPV predicted values that the second pixel is second subregion as the CPV predicted values of first subregion, in reference pixel in reference pixel；Infra-frame prediction is carried out to the present encoding depth unit according to the CPV predicted values of first subregion and the CPV predicted values of second subregion.In the embodiment of the present invention, the CPV predicted values of two subregions are directly determined, selection CPV complexity is reduced, and do not interfere with coding result.

Description

Image intra prediction method and device

Technical field

The present invention relates to image coding technique, more particularly to a kind of image intra prediction method and device.

Background technology

In video coding process, there is certain spatial coherence in a two field picture between adjacent coding unit, for Current coded unit, it is possible to use reference pixel value is closed on around it and is predicted coding, is called infra-frame prediction.Video of new generation Coding standard, i.e. efficient video coding（High Efficiency Video Coding, abbreviation HEVC）Frame in standard is pre- Survey method is to be used for approximate smooth region coding, but easily produces coded faulty for very sharp border, these coding flaws Defect can be showed in synthetic mesophase viewpoint.

In order to preferably realize the prediction to depth map sharp borders, the efficient video coding based on 3 D stereo video （3Dimensions-HEVC, abbreviation 3D-HEVC）Standard has been increased newly for deep on the basis of HEVC standard intra prediction mode Spend the depth model predictive coding pattern of Image Coding（Depth modelling mode, abbreviation DMM）.DMM predictive coding moulds Formula is by the block that present encoding depth unit approximate model is the equal size with two subregions, in block models in each subregion All sample values utilize subregion constant value（Constant Partition Value, abbreviation CPV）Represent.For given Subregion, optimal situation be the original depth-map sampling point signal mean approximation for using respective regions as CPV value, however, by The sampling point information of original depth-map can not be obtained in decoding end, thus CPV differences are sought in the use of 3D-HEVC standards（ΔCPV）Side Formula is further compensate for the CPV value of each subregion, and this is accomplished by one CPV value of extra prediction.

In the prior art mainly by seeking the average value of two reference pixels, the modes such as operation being compared to reference pixel To obtain CPV value, but these modes operate it is complex.

The content of the invention

The embodiment of the present invention provides a kind of image intra prediction method and device, for solving to determine CPV in the prior art The problem of Value Operations are complicated.

First aspect of the embodiment of the present invention provides a kind of image intra prediction method, including：

It is the first subregion and the second subregion by present encoding depth dividing elements；

According to the distribution of pixel in first subregion and the second subregion, determine that the first pixel is made in reference pixel Determine that the second pixel is second subregion for the subregion constant value C PV predicted values of first subregion, in reference pixel CPV predicted values；

It is deep to the present encoding according to the CPV predicted values of first subregion and the CPV predicted values of second subregion Spend unit and carry out infra-frame prediction；

Wherein, the present encoding depth unit includes N × N number of pixel；The reference pixel includes reference pixel row With reference pixel row；The reference pixel row is not belonging to the present encoding depth unit, and the reference pixel row is and institute The adjacent one-row pixels point of present encoding depth the 1st row pixel of unit is stated, the 1st of the reference pixel row arranges and described current 1st column alignment of coding depth unit；The reference pixel row are not belonging to the present encoding depth unit, and the reference image Element row are a row pixels adjacent with the 1st row pixel of the present encoding depth unit, and the 1st of reference pixel row the Row aligns with the 1st row of the present encoding depth unit；The intersecting pixel of reference pixel row and the reference pixel row The starting point that point arranges for the reference pixel row and the reference pixel.

With reference in a first aspect, in first aspect in the first possible embodiment, in the present encoding depth unit The pixel of 1st row the 1st row and the pixel of the 1st row Nth column belong to first subregion, and the pixel category of the 1st row Nth row In second subregion；

It is described determined in reference pixel the first pixel as first subregion subregion constant value C PV predicted values, The CPV predicted values that the second pixel is second subregion are determined in reference pixel, including：

The pixel for determining N/2 row in the reference pixel row is the CPV predicted values of first subregion, it is determined that described The pixel of N+1 rows is the CPV predicted values of second subregion in reference pixel row.

With reference in a first aspect, in second of possible embodiment of first aspect, in the present encoding depth unit The pixel of 1st row Nth column belongs to second subregion, and the 1st arranges the pixel of the 1st row and the pixel category of the 1st row Nth row In first subregion；

The pixel for determining N/2 rows in the reference pixel row is the CPV predicted values of first subregion, it is determined that described The pixel that N+1 is arranged in reference pixel row is the CPV predicted values of second subregion.

With reference in a first aspect, in first aspect in the third possible embodiment, in the present encoding depth unit The pixel of 1st row Nth column and the pixel of the 1st row Nth row belong to first subregion or belong to second subregion；

It is described to determine the first pixel as the subregion constant value C PV predicted value bags of first subregion in reference pixel Include：

The pixel for determining the 1st row in the reference pixel row is the CPV predicted values of first subregion；Or,

Determine that the reference pixel row and the intersecting pixel of reference pixel row are predicted for the CPV of first subregion Value；

Or,

The pixel for determining the 1st row in the reference pixel row is the CPV predicted values of first subregion.

With reference to first aspect or first aspect the third possible embodiment, implement the 4th kind of first aspect is possible In mode, the pixel of the 1st row Nth column and the pixel of the 1st row Nth row belong to described in the present encoding depth unit The pixel of Nth row Nth column belongs to the second subregion in first subregion, and the present encoding depth unit；

It is described to determine that the CPV predicted values that the second pixel is second subregion include in reference pixel：

The pixel for determining 2N row in the reference pixel row is the CPV predicted values of second subregion；Or,

The pixel for determining 2N rows in the reference pixel row is the CPV predicted values of second subregion.

With reference to first aspect or first aspect the third possible embodiment, implement the 5th kind of first aspect is possible In mode, the pixel of the 1st row Nth column and the pixel of the 1st row Nth row belong to described in the present encoding depth unit The pixel of Nth row Nth column belongs to the second subregion in first subregion, and the present encoding depth unit；

Compare the size of the first absolute value and the second absolute value, wherein, first absolute value is the reference pixel row In 2N arrange pixel with the 1st row pixel difference absolute value, second absolute value be the reference pixel row In 2N rows pixel and the 1st row pixel difference absolute value；

If first absolute value is more than second absolute value, it is determined that the pixel that 2N is arranged in the reference pixel row Point is the CPV value of second subregion；

If first absolute value is less than or equal to second absolute value, it is determined that 2N rows in the reference pixel row Pixel is the CPV value of above-mentioned second subregion.

With reference to first aspect or first aspect the third possible embodiment, implement the 6th kind of first aspect is possible In mode, the pixel of the 1st row Nth column and the pixel of the 1st row Nth row belong to described in the present encoding depth unit The pixel of Nth row Nth column belongs to the second subregion in first subregion, and the present encoding depth unit；

Compare the size of the 3rd absolute value and the 4th absolute value, wherein, the 3rd absolute value is the reference pixel row In the 2N pixel and the reference pixel row that arrange and the reference pixel arrange the pixel intersected difference absolute value, institute The pixel of 2N rows during the 4th absolute value arranges for the reference pixel is stated to arrange with the reference pixel row and the reference pixel The absolute value of the difference of intersecting pixel；

If the 3rd absolute value is more than the 4th absolute value, it is determined that the pixel that 2N is arranged in the reference pixel row Point is the CPV value of second subregion；

If the 3rd absolute value is less than or equal to the 4th absolute value, it is determined that 2N rows in the reference pixel row Pixel is the CPV value of above-mentioned second subregion.

With reference to first aspect or first aspect the third possible embodiment, implement the 7th kind of first aspect is possible In mode, the pixel of the 1st row Nth column, the pixel and Nth row N of the 1st row Nth row in the present encoding depth unit The pixel of row belongs to the pixel that the 1st row the 1st is arranged in second subregion, and the present encoding depth unit and belongs to institute State the first subregion；

The CPV predicted values for determining that the second pixel is second subregion in reference pixel, including：

The pixel for determining N+1 row in the reference pixel row is the CPV predicted values of second subregion；

Or,

The pixel for determining N+1 rows in the reference pixel row is the CPV predicted values of second subregion.

With reference to first aspect or first aspect the third possible embodiment, implement the 8th kind of first aspect is possible In mode, the pixel of the 1st row Nth column, the pixel and Nth row N of the 1st row Nth row in the present encoding depth unit The pixel of row belongs to the pixel that the 1st row the 1st is arranged in second subregion, and the present encoding depth unit and belongs to institute State the first subregion；

Compare the size of the 5th absolute value and the 6th absolute value, wherein, the 5th absolute value is the reference pixel row In the absolute values of the difference of pixel and the 1st row pixel that arrange of N+1, the 6th absolute value is during the reference pixel is arranged The absolute value of the difference of N+1 rows pixel and the 1st row pixel；

If the 5th absolute value is more than the 6th absolute value, it is determined that the picture that N+1 is arranged in the reference pixel row Vegetarian refreshments is the CPV predicted values of second subregion；

If the 5th absolute value is less than or equal to the 6th absolute value, it is determined that N+1 rows in the reference pixel row Pixel is the CPV predicted values of second subregion.

With reference to first aspect or first aspect the third possible embodiment, implement the 9th kind of first aspect is possible In mode, the pixel of the 1st row Nth column, the pixel and Nth row N of the 1st row Nth row in the present encoding depth unit The pixel of row belongs to the pixel that the 1st row the 1st is arranged in second subregion, and the present encoding depth unit and belongs to institute State the first subregion；

Compare the size of the 7th absolute value and the 8th absolute value, the 7th absolute value is N+ in the reference pixel row Pixel and the reference pixel row and the reference pixel of 1 row arrange the absolute value of the difference of the pixel intersected, and described the Eight absolute values are that the pixel of N+1 rows intersects with the reference pixel row and reference pixel row during the reference pixel is arranged Pixel difference absolute value；

If the 7th absolute value is more than the 8th absolute value, it is determined that the picture that N+1 is arranged in the reference pixel row Vegetarian refreshments is the CPV predicted values of second subregion；

If the 7th absolute value is less than or equal to the 8th absolute value, it is determined that the picture of N+1 rows in the reference pixel row Vegetarian refreshments is the CPV predicted values of second subregion.

Second aspect of the embodiment of the present invention provides a kind of image infra-frame prediction device, including：

Zoning unit, for being the first subregion and the second subregion by present encoding depth dividing elements；

Determining unit, for the distribution according to pixel in first subregion and the second subregion, in reference pixel really Fixed first pixel as first subregion subregion constant value C PV predicted values, the second pixel is determined in reference pixel For the CPV predicted values of second subregion；

Coding unit, for the CPV predicted values according to first subregion and the CPV predicted values of second subregion to institute State present encoding depth unit and carry out infra-frame prediction；

With reference to second aspect, in the first possible embodiment of second aspect, the determining unit, specifically for The pixel and the pixel of the 1st row Nth column that the 1st row the 1st is arranged in the present encoding depth unit belong to described first point Area, and the pixel of the 1st row Nth row is when belonging to second subregion, determines the pixel that N/2 is arranged in the reference pixel row For the CPV predicted values of first subregion, the pixel for determining N+1 rows in the reference pixel row is second subregion CPV predicted values.

With reference to second aspect, in second of possible embodiment of second aspect, the determining unit, specifically for The pixel of the 1st row Nth column belongs to second subregion in the present encoding depth unit, and the 1st pixel for arranging the 1st row When point and the pixel of the 1st row Nth row belong to first subregion, the pixel of N/2 rows in the reference pixel row is determined For the CPV predicted values of first subregion, determine that the pixel that N+1 is arranged in the reference pixel row is second subregion CPV predicted values.

With reference to second aspect, in the third possible embodiment of second aspect, the determining unit, specifically for The pixel of the 1st row Nth column and the pixel of the 1st row Nth row belong to described first in the present encoding depth unit Subregion or when belonging to second subregion, the pixel for determining the 1st row in the reference pixel row is first subregion CPV predicted values；Or, determine that the reference pixel row and the intersecting pixel of reference pixel row are first subregion CPV predicted values；Or, the pixel for determining the 1st row in the reference pixel row is the CPV predicted values of first subregion.

With reference to second aspect or second aspect the third possible embodiment, in the 4th kind of possible reality of second aspect Apply in mode, the determining unit, pixel and the 1st row specifically for the 1st row Nth column in the present encoding depth unit The pixel of Nth row belongs to the pixel category of Nth row Nth column in first subregion, and the present encoding depth unit When the second subregion, the pixel for determining 2N row in the reference pixel row is the CPV predicted values of second subregion；Or Person, the pixel for determining 2N rows in the reference pixel row is the CPV predicted values of second subregion.

With reference to second aspect or second aspect the third possible embodiment, in the 5th kind of possible reality of second aspect Apply in mode, the determining unit, pixel and the 1st row specifically for the 1st row Nth column in the present encoding depth unit The pixel of Nth row belongs to the pixel category of Nth row Nth column in first subregion, and the present encoding depth unit When the second subregion, compare the size of the first absolute value and the second absolute value, wherein, first absolute value is the reference image 2N is arranged in plain row pixel and the absolute value of the difference of the pixel of the 1st row, second absolute value is the reference image The absolute value of the difference of the pixel of the pixel of 2N rows and the 1st row in element row；If first absolute value is more than described the Two absolute values, it is determined that the pixel that 2N is arranged in the reference pixel row is the CPV value of second subregion；If described first Absolute value is less than or equal to second absolute value, it is determined that the pixel of 2N rows is above-mentioned second point in the reference pixel row The CPV value in area.

With reference to second aspect or second aspect the third possible embodiment, in the 6th kind of possible reality of second aspect Apply in mode, the determining unit, specifically for the pixel and the 1st of the 1st row Nth column in the present encoding depth unit The pixel of row Nth row belongs to the pixel of Nth row Nth column in first subregion, and the present encoding depth unit When belonging to the second subregion, compare the size of the 3rd absolute value and the 4th absolute value, wherein, the 3rd absolute value is the reference The difference for the pixel that the pixel of 2N row intersects with the reference pixel row and reference pixel row is exhausted in pixel column To value, the 4th absolute value is the pixel of 2N rows and the reference pixel row and the reference during the reference pixel is arranged The absolute value of the difference of the intersecting pixel of pixel column；If the 3rd absolute value is more than the 4th absolute value, it is determined that institute State the CPV value that the pixel that 2N is arranged in reference pixel row is second subregion；If the 3rd absolute value is less than or equal to institute State the 4th absolute value, it is determined that the pixel of 2N rows is the CPV value of above-mentioned second subregion in the reference pixel row.

With reference to second aspect or second aspect the third possible embodiment, in the 7th kind of possible reality of second aspect Apply in mode, the determining unit, specifically for the pixel of the 1st row Nth column, the 1st in the present encoding depth unit The pixel of row Nth row and the pixel of Nth row Nth column belong to second subregion, and the present encoding depth unit In the pixel that arranges of the 1st row the 1st when belonging to first subregion, determine that the pixel that N+1 is arranged in the reference pixel row is The CPV predicted values of second subregion；Or, the pixel for determining N+1 rows in the reference pixel row is described second point The CPV predicted values in area.

With reference to second aspect or second aspect the third possible embodiment, in the 8th kind of possible reality of second aspect Apply in mode, the determining unit, specifically for the pixel of the 1st row Nth column, the 1st in the present encoding depth unit The pixel of row Nth row and the pixel of Nth row Nth column belong to second subregion, and the present encoding depth unit In the pixel that arranges of the 1st row the 1st when belonging to first subregion, compare the size of the 5th absolute value and the 6th absolute value, wherein, 5th absolute value is the pixels arranged of N+1 in the reference pixel row and the absolute value of the difference of the 1st row pixel, institute State the absolute value of the difference of N+1 rows pixel and the 1st row pixel during the 6th absolute value arranges for the reference pixel；If described 5th absolute value is more than the 6th absolute value, it is determined that the pixel that N+1 is arranged in the reference pixel row is described second The CPV predicted values of subregion；If the 5th absolute value is less than or equal to the 6th absolute value, it is determined that in the reference pixel row N+1 rows pixel is the CPV predicted values of second subregion.

With reference to second aspect or second aspect the third possible embodiment, in the 9th kind of possible reality of second aspect Apply in mode, the determining unit, specifically for the pixel of the 1st row Nth column, the 1st in the present encoding depth unit The pixel of row Nth row and the pixel of Nth row Nth column belong to second subregion, and the present encoding depth unit In the pixel that arranges of the 1st row the 1st when belonging to first subregion, compare the size of the 7th absolute value and the 8th absolute value, it is described 7th absolute value is the pixel and the reference pixel row and the reference pixel row phase of N+1 row in the reference pixel row The absolute value of the difference of the pixel of friendship, the pixel of N+1 rows and institute during the 8th absolute value arranges for the reference pixel State the absolute value of the difference of reference pixel row and the intersecting pixel of reference pixel row；If the 7th absolute value is more than institute State the 8th absolute value, it is determined that the pixel that N+1 is arranged in the reference pixel row is the CPV predicted values of second subregion； If the 7th absolute value is less than or equal to the 8th absolute value, it is determined that the pixel of N+1 rows is institute in the reference pixel row State the CPV predicted values of the second subregion.

It is the first subregion and the second subregion by present encoding depth dividing elements in the embodiment of the present invention, and according to first The distribution of pixel in subregion and the second subregion, determines that the first pixel is predicted as the CPV of the first subregion in reference pixel Value, determines the CPV predicted values that the second pixel is above-mentioned second subregion, and the CPV predicted values using the first subregion and second point The CPV predicted values in area are encoded to present encoding depth unit.The CPV predicted values of two subregions are so directly determined, are dropped Low complexity, and do not interfere with coding result.

Brief description of the drawings

In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing There is the accompanying drawing used required in technology description to be briefly described, it should be apparent that, drawings in the following description are this hairs Some bright embodiments, for those of ordinary skill in the art, without having to pay creative labor, can be with Other accompanying drawings are obtained according to these accompanying drawings.

The schematic flow sheet for the image intra prediction method embodiment one that Fig. 1 provides for the present invention；

The pixel schematic diagram for the image intra prediction method embodiment two that Fig. 2 provides for the present invention；

The pixel schematic diagram for the image intra prediction method embodiment three that Fig. 3 provides for the present invention；

The pixel schematic diagram for the image intra prediction method example IV that Fig. 4 provides for the present invention；

The pixel schematic diagram for the image intra prediction method embodiment five that Fig. 5 provides for the present invention；

The pixel schematic diagram for the image intra prediction method embodiment six that Fig. 6 provides for the present invention；

The pixel schematic diagram for the image intra prediction method embodiment seven that Fig. 7 provides for the present invention；

The pixel schematic diagram for the image intra prediction method embodiment eight that Fig. 8 provides for the present invention；

The structural representation for the image infra-frame prediction device embodiment one that Fig. 9 provides for the present invention；

The structural representation for the image infra-frame prediction device embodiment two that Figure 10 provides for the present invention.

Embodiment

To make the purpose, technical scheme and advantage of the embodiment of the present invention clearer, below in conjunction with the embodiment of the present invention In accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is A part of embodiment of the present invention, rather than whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art The every other embodiment obtained under the premise of creative work is not made, belongs to the scope of protection of the invention.

The schematic flow sheet for the image intra prediction method embodiment one that Fig. 1 provides for the present invention, Fig. 2 provides for the present invention Image intra prediction method embodiment two pixel schematic diagram, Fig. 3 is that the image intra prediction method that provides of the present invention is implemented The pixel schematic diagram of example three, the pixel schematic diagram of image intra prediction method example IV that Fig. 4 provides for the present invention, Fig. 5 is The pixel schematic diagram for the image intra prediction method embodiment five that the present invention is provided, as shown in figure 1, this method includes：

S101, by present encoding depth dividing elements be the first subregion and the second subregion.I.e. by present encoding depth unit It is divided into two subregions.

S102, the distribution according to pixel in above-mentioned first subregion and the second subregion, determine the first picture in reference pixel Vegetarian refreshments determines the CPV predicted values that the second pixel is the second subregion as the CPV predicted values of the first subregion, in reference pixel.

Which specifically, pixel can be included respectively according to the first subregion, the second subregion to determine the first subregion and the The CPV predicted values of two subregions.

S103, the CPV predicted values according to above-mentioned first subregion and the second subregion CPV predicted values are to present encoding depth list Member carries out infra-frame prediction.

It should be noted that above-mentioned present encoding depth unit includes N × N number of pixel, that is, there are N row N row pixels, such as Shown in Fig. 2-Fig. 5, illustrated by taking 8 × 8 present encoding depth unit as an example, wherein each square represents a pixel. In the schematic diagram of the embodiment of the present invention, white square belongs to the first subregion, and the square of twill shade belongs to the second subregion, mark The square that note has coordinate symbol is all reference pixel.

Above-mentioned reference pixel includes reference pixel row and reference pixel is arranged, wherein, reference pixel row is not belonging to present encoding Depth unit, and the reference pixel row is the pixel of a line adjacent with present encoding depth the 1st row pixel of unit, and on The 1st column alignment of the 1st row and above-mentioned present encoding depth unit of reference pixel row is stated, usually, in reference pixel row at most 2N+1 pixel can be included；Above-mentioned reference pixel row are not belonging to above-mentioned present encoding depth unit, and reference pixel row are The pixel of the adjacent row of the 1st row pixel of above-mentioned present encoding depth unit, and the 1st row that above-mentioned reference pixel is arranged is with working as The 1st row alignment of preceding coding depth unit, usually, can at most include 2N+1 pixel in reference pixel row.Need It is bright, that the intersecting pixel of above-mentioned reference pixel row and reference pixel row arranges for above-mentioned reference pixel row and reference pixel Point.By taking Fig. 1 as an example, wherein the square for indicating coordinate number is reference image vegetarian refreshments, p_-1,-1~p_15,-1Belong to reference pixel row, p_-1,-1 ~p_-1,15Belong to reference pixel row, wherein, p_-1,-1For reference pixel row the pixel intersected is arranged with reference pixel.For the ease of Illustrate, with p in the embodiment of the present invention_0,-1As the 1st row pixel of reference pixel row, with p_-1,0The 1st arranged as reference pixel Row pixel.

It is the first subregion and the second subregion by present encoding depth dividing elements in the present embodiment, and according to the first subregion With the distribution of pixel in the second subregion, the first pixel is determined in reference pixel as the CPV predicted values of the first subregion, really Fixed second pixel is the CPV predicted values of above-mentioned second subregion, and CPV predicted values using the first subregion and the second subregion CPV predicted values are encoded to present encoding depth unit.The CPV predicted values of two subregions are so directly determined, are reduced Complexity, and do not interfere with coding result.

Further, when the pixel and the pixel of the 1st row Nth row of the 1st row Nth column in above-mentioned present encoding depth unit It is above-mentioned to determine the first pixel as the first subregion in reference pixel when point belongs to the first subregion or belongs to the second subregion CPV predicted values, be specially：Determine that the first pixel is predicted as the CPV of above-mentioned first subregion in above-mentioned reference pixel row Value.

Alternatively, when the pixel and the pixel of the 1st row Nth row of the 1st row Nth column in above-mentioned present encoding depth unit It is above-mentioned to determine that the second pixel is the second subregion in reference pixel when belonging to the first subregion or belonging to the second subregion CPV predicted values, can be the CPV predicted values that the second pixel is determined in reference pixel row as the second subregion.But not with This is limited.

On the basis of above-described embodiment,（a）A kind of distribution situation of pixel is as follows：Above-mentioned present encoding depth unit In the pixel of the pixel of the 1st row Nth column, the pixel of the 1st row Nth row and the 1st row the 1st row belong to the first subregion, and When the pixel of Nth row Nth column belongs to the second subregion in present encoding depth unit, such as the pixel distribution shown in Fig. 2, on The CPV predicted values that the first pixel is determined in reference pixel as above-mentioned first subregion are stated, are specially：Determine above-mentioned reference image The pixel of the 1st row is the CPV predicted values of the first subregion in plain row, or, determine above-mentioned reference pixel row and reference pixel row Intersecting pixel is the CPV predicted values of above-mentioned first subregion.Reference picture 2, that is, determine p_0,-1CPV for above-mentioned first subregion is pre- Measured value, or, determine p_-1,-1For the CPV predicted values of the first subregion.

It is above-mentioned（a）In the case of, the above-mentioned pixel of determination second is the CPV predicted values of above-mentioned second subregion, is specially：1）Really The pixel that 2N is arranged in fixed above-mentioned reference pixel row is the CPV predicted values of above-mentioned second subregion.Reference picture 2, determines p_15,-1For The CPV predicted values of above-mentioned second subregion.Or, 2）Compare the size of the first absolute value and the second absolute value.Wherein first is absolute It is worth the pixel and the absolute value of the difference of the 1st pixel arranged arranged for 2N in above-mentioned reference pixel row, reference picture 2, i.e., the One absolute value is | p_15,-1-p_0,-1|；Second absolute value is the pixel and the pixel of the 1st row of 2N rows during above-mentioned reference pixel is arranged The absolute value of the difference of point, reference picture 2, i.e. the second absolute value be | p_-1,15-p_-1,0|.If it is absolute that the first absolute value is more than second Value, it is determined that the pixel that 2N is arranged in above-mentioned reference pixel row is the CPV value of above-mentioned second subregion, reference picture 2 determines p_15,-1For the CPV value of the second subregion；If the first absolute value is less than or equal to the second absolute value, it is determined that the in above-mentioned reference pixel row The pixel of 2N rows is the CPV value of above-mentioned second subregion, and reference picture 2 determines p_-1,15For the CPV value of the second subregion.Or, 3） Compare the size of the 3rd absolute value and the 4th absolute value.Wherein, the 3rd absolute value is 2N row pixels in above-mentioned reference pixel row Point arranges the absolute value of the difference of the pixel intersected with reference pixel row and reference pixel, and reference picture 2, i.e. the 3rd absolute value be | p_15,-1-p_-1,-1|；4th absolute value is the pixel and reference pixel row and reference pixel of 2N rows during above-mentioned reference pixel is arranged The absolute value of the difference of the intersecting pixel of row, reference picture 2, i.e. the 4th absolute value be | p_-1,15-p_-1,-1|.If the 3rd absolute value More than the 4th absolute value, it is determined that the pixel that 2N is arranged in above-mentioned reference pixel row is the CPV value of above-mentioned second subregion, reference Fig. 2, that is, determine p_15,-1For the CPV value of the second subregion.If the 3rd absolute value is less than or equal to the 4th absolute value, it is determined that above-mentioned reference The pixel of 2N rows is the CPV value of above-mentioned second subregion in pixel column, and reference picture 2 determines p_-1,15For the CPV of the second subregion Value.

In another embodiment,（b）A kind of distribution situation of pixel is as follows：The 1st in above-mentioned present encoding depth unit The pixel of the pixel of row Nth column, the pixel of the 1st row Nth row and Nth row Nth column belongs to above-mentioned second subregion, and on The pixel for stating the row the 1st of present encoding depth unit the 1st row belongs in the first subregion, such as the pixel distribution shown in Fig. 3, on The CPV predicted values that the first pixel is determined in reference pixel as above-mentioned first subregion are stated, are specially：Determine above-mentioned reference image The pixel of the 1st row is the CPV predicted values of above-mentioned first subregion in plain row, or, determine above-mentioned reference pixel row and reference image The intersecting pixel of element row is the CPV predicted values of above-mentioned first subregion.Reference picture 3, that is, determine p_0,-1CPV for the first subregion is pre- Measured value, or, determine p_-1,-1For the CPV predicted values of the first subregion.

Above-mentioned（b）In the case of, the CPV predicted values that the second pixel is second subregion are determined in reference pixel, Specially：（1）The pixel for determining N+1 row in above-mentioned reference pixel row is the CPV predicted values of the second subregion.Reference picture 3, i.e., Determine p_8,-1For the CPV predicted values of the second subregion.Or,（2）Compare the size of the 5th absolute value and the 6th absolute value.Wherein, 5th absolute value is the pixels arranged of N+1 in above-mentioned reference pixel row and the absolute value of the difference of the 1st row pixel, reference picture 3, i.e. the 5th absolute value be | p_8,-1-p_0,-1|；6th absolute value is N+1 rows pixel and the 1st row pixel in reference pixel row Difference absolute value, reference picture 3, i.e. the 6th absolute value be | p_-1,8-p_-1,0|.If the 5th absolute value is more than the 6th absolute value, The pixel for determining N+1 row in reference pixel row is the CPV predicted values of above-mentioned second subregion, and reference picture 3 determines p_8,-1For The CPV predicted values of second subregion.If the 5th absolute value is less than or equal to the 6th absolute value, it is determined that reference picture 3, in reference pixel row N+1 rows pixel is the CPV predicted values of above-mentioned second subregion, and reference picture 3 determines p_-1,8Predicted for the CPV of the second subregion Value.Or,（3）Compare the size of the 7th absolute value and the 8th absolute value.Wherein, the 7th absolute value is in above-mentioned reference pixel row The absolute value of the difference for the pixel that the pixel of N+1 row intersects with above-mentioned reference pixel row and above-mentioned reference pixel row, ginseng According to Fig. 3, the 5th absolute value is | p_8,-1-p_-1,-1|；8th absolute value be above-mentioned reference pixel arrange in N+1 rows pixel with it is upper The absolute value of the difference of reference pixel row and the intersecting pixel of above-mentioned reference pixel row, reference picture 3 are stated, the 8th absolute value is | p_-1,8-p_-1,-1|.If the 7th absolute value is more than the 8th absolute value, it is determined that the pixel that N+1 is arranged in reference pixel row is above-mentioned The CPV predicted values of second subregion, reference picture 3 determines p_8,-1For the CPV predicted values of the second subregion.If the 7th absolute value is less than Equal to the 8th absolute value, it is determined that the pixel of N+1 rows is the CPV predicted values of the second subregion, reference picture in reference pixel row 3, that is, determine p_-1,8For the CPV predicted values of the second subregion.

In specific implementation process, it is not limited with previous embodiment.

For example, in another embodiment of the present invention, for the 1st row Nth column in the present encoding depth unit Pixel and the pixel of the 1st row Nth row belong to first subregion or belong to the situation of second subregion, it is described Determine that the first pixel includes as the subregion constant value C PV predicted values of first subregion in reference pixel：Determine the ginseng The pixel for examining the 1st row in pixel column is the CPV predicted values of first subregion；Or, determine the reference pixel row and institute State the CPV predicted values that the intersecting pixel of reference pixel row is first subregion；Or, determine in reference pixel row the The pixel of 1 row is the CPV predicted values of first subregion.

In another embodiment of the present invention, for the pixel of the 1st row Nth column in the present encoding depth unit The picture of Nth row Nth column in first subregion, and the present encoding depth unit is belonged to the pixel of the 1st row Nth row Vegetarian refreshments belongs to the situation of the second subregion, described to determine that the CPV that the second pixel is second subregion is predicted in reference pixel Value includes：The pixel for determining 2N row in the reference pixel row is the CPV predicted values of second subregion；Or, it is determined that The pixel of 2N rows is the CPV predicted values of second subregion in the reference pixel row.

In another embodiment of the present invention, for the pixel of the 1st row Nth column in the present encoding depth unit The picture of Nth row Nth column in first subregion, and the present encoding depth unit is belonged to the pixel of the 1st row Nth row Vegetarian refreshments belongs to the situation of the second subregion, described to determine that the CPV that the second pixel is second subregion is predicted in reference pixel Value includes：Compare the size of the first absolute value and the second absolute value, wherein, first absolute value is in the reference pixel row The absolute value of the difference of the pixel of the pixel of 2N row and the 1st row, second absolute value is during the reference pixel is arranged The absolute value of the difference of the pixel of the pixel of 2N rows and the 1st row；If it is absolute that first absolute value is more than described second Value, it is determined that the pixel that 2N is arranged in the reference pixel row is the CPV value of second subregion；If first absolute value Less than or equal to second absolute value, it is determined that the pixel of 2N rows is above-mentioned second subregion in the reference pixel row CPV value.

In another embodiment of the present invention, for the pixel of the 1st row Nth column in the present encoding depth unit The picture of Nth row Nth column in first subregion, and the present encoding depth unit is belonged to the pixel of the 1st row Nth row Vegetarian refreshments belongs to the situation of the second subregion, described to determine that the CPV that the second pixel is second subregion is predicted in reference pixel Value includes：Compare the size of the 3rd absolute value and the 4th absolute value, wherein, the 3rd absolute value is in the reference pixel row The absolute value of the difference for the pixel that the pixel of 2N row intersects with the reference pixel row and reference pixel row, it is described 4th absolute value is the pixel and the reference pixel row and the reference pixel row phase of 2N rows during the reference pixel is arranged The absolute value of the difference of the pixel of friendship；If the 3rd absolute value is more than the 4th absolute value, it is determined that the reference image The pixel that 2N is arranged in plain row is the CPV value of second subregion；If it is exhausted that the 3rd absolute value is less than or equal to the described 4th To value, it is determined that the pixel of 2N rows is the CPV value of above-mentioned second subregion in the reference pixel row.

In another embodiment of the present invention, for the pixel of the 1st row Nth column in the present encoding depth unit The pixel of point, the pixel of the 1st row Nth row and Nth row Nth column belongs to second subregion, and the present encoding is deep The pixel that the 1st row the 1st is arranged in degree unit belongs to the situation of first subregion, described that the second pixel is determined in reference pixel Point is the CPV predicted values of second subregion, including：It is described the to determine pixel that N+1 in the reference pixel row arranges The CPV predicted values of two subregions；Or, the pixel for determining N+1 rows in the reference pixel row is the CPV of second subregion Predicted value.

In another embodiment of the present invention, for the pixel of the 1st row Nth column in the present encoding depth unit The pixel of point, the pixel of the 1st row Nth row and Nth row Nth column belongs to second subregion, and the present encoding is deep The pixel that the 1st row the 1st is arranged in degree unit belongs to the situation of first subregion, described that the second pixel is determined in reference pixel Point is the CPV predicted values of second subregion, including：Compare the size of the 5th absolute value and the 6th absolute value, wherein, described Five absolute values are the pixels arranged of N+1 in the reference pixel row and the absolute value of the difference of the 1st row pixel, the described 6th Absolute value is the absolute value of the difference of N+1 rows pixel and the 1st row pixel during the reference pixel is arranged；If the pentasyllabic quatrain 6th absolute value is more than to value, it is determined that the pixel that N+1 is arranged in the reference pixel row is second subregion CPV predicted values；If the 5th absolute value is less than or equal to the 6th absolute value, it is determined that N+1 in the reference pixel row Row pixel is the CPV predicted values of second subregion.

In another embodiment of the present invention, for the pixel of the 1st row Nth column in the present encoding depth unit The pixel of point, the pixel of the 1st row Nth row and Nth row Nth column belongs to second subregion, and the present encoding is deep The pixel that the 1st row the 1st is arranged in degree unit belongs to the situation of first subregion, compares the 7th absolute value and the 8th absolute value Size, the 7th absolute value is the pixel and the reference pixel row and the ginseng of N+1 row in the reference pixel row The absolute value of the difference of the intersecting pixel of pixel column is examined, the 8th absolute value is N+1 rows during the reference pixel is arranged The absolute value of the difference for the pixel that pixel intersects with the reference pixel row and reference pixel row；If the four-line poem with seven characters to a line 8th absolute value is more than to value, it is determined that the pixel that N+1 is arranged in the reference pixel row is second subregion CPV predicted values；If the 7th absolute value is less than or equal to the 8th absolute value, it is determined that N+1 rows in the reference pixel row Pixel is the CPV predicted values of second subregion.

Implement principle and scheme is similar with previous embodiment, the first pixel, the second pixel for determination can To be reasonably combined, it will not enumerate herein.

In another embodiment,（c）A kind of distribution situation of pixel is as follows：When in above-mentioned present encoding depth unit The pixel of 1 row the 1st row and the pixel of the 1st row Nth column belong to above-mentioned first subregion, and the pixel of the 1st row Nth row belongs to During above-mentioned second subregion, pixel distribution as shown in Figure 4 is can refer to, it is above-mentioned that the first pixel conduct is determined in reference pixel The CPV predicted values of above-mentioned first subregion, the CPV predicted values that the second pixel is above-mentioned second subregion are determined in reference pixel, Specially：Determine that the pixel that N/2 is arranged in above-mentioned reference pixel row is the CPV predicted values of the first subregion, determine above-mentioned reference The pixel of N+1 rows is the CPV predicted values of above-mentioned second subregion in pixel column.Reference picture 4, determines p_3,-1For the first subregion CPV predicted values, determine p_-1,8For the CPV predicted values of above-mentioned second subregion.

In another embodiment,（d）A kind of distribution situation of pixel is as follows：When the 1st in above-mentioned present encoding depth unit The pixel of row Nth column belongs to above-mentioned second subregion, and the pixel of the 1st row Nth row is when belonging to above-mentioned first subregion, such as Fig. 5 Shown pixel distribution, it is above-mentioned determined in reference pixel the first pixel as above-mentioned first subregion CPV predicted values, The CPV predicted values that the second pixel is above-mentioned second subregion are determined in reference pixel, are specially：

The pixel for determining N/2 rows in above-mentioned reference pixel row is the CPV predicted values of the first subregion, determines above-mentioned reference The pixel that N+1 is arranged in pixel column is the CPV predicted values of the second subregion.Reference picture 5, determines p_-1,3For the CPV of the first subregion Predicted value, determines p_8,-1For the CPV predicted values of the second subregion.

The pixel schematic diagram for the image intra prediction method embodiment six that Fig. 6 provides for the present invention, for above-mentioned（b）Picture Element distribution：The pixel of 1st row Nth column, the pixel and Nth row N of the 1st row Nth row in above-mentioned present encoding depth unit The pixel of row belongs to above-mentioned second subregion, and the pixel of the above-mentioned row the 1st of present encoding depth unit the 1st row belongs to first In subregion, determine that N+1 rows pixel is the second subregion in the pixel or reference pixel row that N+1 is arranged in reference pixel row CPV predicted values, CPV predicted values can also be avoided mistake occur.As shown in fig. 6, in the pixel of the first subregion and the second subregion Point is distributed as under the extreme case as shown in 6A in Fig. 6, and the actual boundary of present encoding depth unit is possible to expand in Fig. 6 " point " hatched boxes part shown in 6B, if pressing prior art in this case, p is determined according to comparative result_7,-1Or p_-1,7For The CPV predicted values of second subregion, then occur the situation of CPV predicted value mistakes, influences the accuracy of infra-frame prediction.And use this The method of invention previous embodiment, then be not in the problem.

The pixel schematic diagram for the image intra prediction method embodiment seven that Fig. 7 provides for the present invention, for above-mentioned（c）Picture Element distribution：The pixel and the pixel of the 1st row Nth column that the 1st row the 1st is arranged in above-mentioned present encoding depth unit belong to above-mentioned the One subregion, and the pixel of the 1st row Nth row belongs to above-mentioned second subregion, the pixel for determining N+1 rows in reference pixel row is The CPV predicted values of second subregion, can also avoid CPV predicted values from mistake occur.As shown in fig. 7, in the first subregion and second point The pixel in area is distributed as under the extreme case as shown in 7A in Fig. 7, and the actual boundary of present encoding depth unit is possible to expand " point " hatched boxes part shown in 7B in Fig. 7 is opened up, if pressing prior art in this case, p is determined_-1,7For the second subregion CPV predicted values, then occur the situation of CPV predicted value mistakes, influence the accuracy of infra-frame prediction.And use the present invention foregoing The method of embodiment, then be not in the problem.

The pixel schematic diagram for the image intra prediction method embodiment eight that Fig. 8 provides for the present invention, for（d）Pixel point Cloth：The pixel of the 1st row Nth column belongs to above-mentioned second subregion, and the picture of the 1st row Nth row in above-mentioned present encoding depth unit Vegetarian refreshments belongs to above-mentioned first subregion, determines that the pixel that N+1 is arranged in reference pixel row is the CPV predicted values of the second subregion, goes back CPV predicted values can be avoided mistake occur.As shown in figure 8, in the pixel of the first subregion and the second subregion is distributed as such as Fig. 8 Under extreme case shown in 8A, it is cloudy that the actual boundary of present encoding depth unit is possible to expand in Fig. 8 " point " shown in 8B Shadow square part, if pressing prior art in this case, determines p_7,-1For the CPV predicted values of the second subregion, then CPV occurs The situation of predicted value mistake, influences the accuracy of infra-frame prediction.And the method for using present invention, then be not in The problem.

The structural representation for the image infra-frame prediction device embodiment one that Fig. 9 provides for the present invention, as shown in figure 9, the dress Put including：Zoning unit 901, determining unit 902 and coding unit 903.Wherein：

Zoning unit 901, for being the first subregion and the second subregion by present encoding depth dividing elements.

Determining unit 902, for the distribution according to pixel in first subregion and the second subregion, in reference pixel Determine the first pixel as first subregion subregion constant value C PV predicted values, the second pixel is determined in reference pixel Point is the CPV predicted values of second subregion.

Coding unit 903, for the CPV predicted values according to first subregion and the CPV predicted values of second subregion Infra-frame prediction is carried out to the present encoding depth unit.

It should be noted that the present encoding depth unit includes N × N number of pixel；The reference pixel includes ginseng Examine pixel column and reference pixel row；The reference pixel row is not belonging to the present encoding depth unit, and the reference pixel Row is the one-row pixels point adjacent with the 1st row pixel of the present encoding depth unit, the 1st row of the reference pixel row with 1st column alignment of the present encoding depth unit；The reference pixel row are not belonging to the present encoding depth unit, and institute It is a row pixel adjacent with the 1st row pixel of the present encoding depth unit to state reference pixel row, and the reference image 1st row of element row aligns with the 1st row of the present encoding depth unit；The reference pixel row and the reference pixel row phase The pixel of friendship is the starting point that the reference pixel row and the reference pixel are arranged.

The device is used to perform preceding method embodiment, and its implementing principle and technical effect is similar, will not be repeated here.

In another embodiment, determining unit 902, specifically can be used for the 1st in the present encoding depth unit When the pixel of row Nth column and the pixel of the 1st row Nth row belong to first subregion or belong to second subregion, Determine the first pixel as the CPV predicted values of first subregion in the reference pixel row.

In another embodiment, determining unit 902, specifically can be used for the 1st in the present encoding depth unit When the pixel of row Nth column and the pixel of the 1st row Nth row belong to first subregion or belong to second subregion, Determine the second pixel as the subregion constant value C PV predicted values of second subregion in the reference pixel row.

In another embodiment, determining unit 902, specifically can be used for the 1st in the present encoding depth unit The pixel of the pixel of row Nth column, the pixel of the 1st row Nth row and the 1st row the 1st row belongs to first subregion, and institute When the pixel for stating Nth row Nth column in present encoding depth unit belongs to second subregion, determine in the reference pixel row The pixel of 1st row is the CPV predicted values of first subregion；Or, determine the reference pixel row and the reference pixel The intersecting pixel of row is the CPV predicted values of first subregion.

In another embodiment, determining unit 902, specifically can be used for the 1st in the present encoding depth unit The pixel of the pixel of row Nth column, the pixel of the 1st row Nth row and the 1st row the 1st row belongs to first subregion, and institute When the pixel for stating Nth row Nth column in present encoding depth unit belongs to the second subregion, 2N in the reference pixel row is determined The pixel of row is the CPV predicted values of second subregion.

In another embodiment, determining unit 902, specifically can be used for the 1st in the present encoding depth unit The pixel of the pixel of row Nth column, the pixel of the 1st row Nth row and the 1st row the 1st row belongs to first subregion, and institute When the pixel for stating Nth row Nth column in present encoding depth unit belongs to the second subregion, compare the first absolute value and second absolute The size of value, wherein, first absolute value is the pixel of 2N row in the reference pixel row and the pixel of the 1st row The absolute value of difference, second absolute value is the pixel of the pixel of 2N rows and the 1st row during the reference pixel is arranged The absolute value of difference；If first absolute value is more than second absolute value, it is determined that 2N is arranged in the reference pixel row Pixel be second subregion CPV value；If first absolute value is less than or equal to second absolute value, it is determined that institute The pixel for stating 2N rows in reference pixel row is the CPV value of above-mentioned second subregion.

In another embodiment, determining unit 902, specifically can be used for the 1st in the present encoding depth unit The pixel of the pixel of row Nth column, the pixel of the 1st row Nth row and the 1st row the 1st row belongs to first subregion, and institute When the pixel for stating Nth row Nth column in present encoding depth unit belongs to the second subregion, compare the 3rd absolute value and the 4th absolute The size of value, wherein, the 3rd absolute value is the pixel and the reference pixel row of 2N row in the reference pixel row Arrange the absolute value of the difference of intersecting pixel with the reference pixel, the 4th absolute value is the in reference pixel row The absolute value of the difference for the pixel that the pixel of 2N rows intersects with the reference pixel row and reference pixel row；If described 3rd absolute value is more than the 4th absolute value, it is determined that the pixel that 2N is arranged in the reference pixel row is described second point The CPV value in area；If the 3rd absolute value is less than or equal to the 4th absolute value, it is determined that 2N rows in the reference pixel row Pixel be above-mentioned second subregion CPV value.

In another embodiment, determining unit 902, specifically can be used for the 1st in the present encoding depth unit The pixel of the pixel of row Nth column, the pixel of the 1st row Nth row and Nth row Nth column belongs to second subregion, and institute When the pixel for stating the 1st row the 1st row in present encoding depth unit belongs to first subregion, determine in the reference pixel row The pixel of 1st row is the CPV predicted values of first subregion；Or, determine the reference pixel row and the reference pixel The intersecting pixel of row is the CPV predicted values of first subregion.

In another embodiment, determining unit 902, specifically can be used for the 1st in the present encoding depth unit The pixel of row Nth column, the pixel of the 1st row Nth row, the pixel of Nth row Nth column belong to second subregion, and institute When the pixel for stating the 1st row the 1st row in present encoding depth unit belongs to first subregion, determine in the reference pixel row The pixel of N+1 row is the CPV predicted values of second subregion.

In another embodiment, determining unit 902, specifically can be used for the 1st row in the present encoding depth unit The pixel of Nth column, the pixel of the 1st row Nth row, the pixel of Nth row Nth column belong to second subregion, and described The pixel that the 1st row the 1st is arranged in present encoding depth unit is when belonging to first subregion, compares the 5th absolute value and the 6th exhausted To the size of value, wherein, the 5th absolute value is the pixel and the 1st row pixel of N+1 row in the reference pixel row Difference absolute value, the 6th absolute value be the reference pixel arrange in N+1 rows pixel and the 1st row pixel difference The absolute value of value；If the 5th absolute value is more than the 6th absolute value, it is determined that N+1 is arranged in the reference pixel row Pixel is the CPV predicted values of second subregion；If the 5th absolute value is less than or equal to the 6th absolute value, it is determined that N+1 rows pixel is the CPV predicted values of second subregion in the reference pixel row.

In another embodiment, determining unit 902, specifically can be used for the 1st in the present encoding depth unit The pixel of row Nth column, the pixel of the 1st row Nth row, the pixel of Nth row Nth column belong to second subregion, and institute When the pixel for stating the 1st row the 1st row in present encoding depth unit belongs to first subregion, compare the 7th absolute value and the 8th The size of absolute value, the 7th absolute value is the pixel and the reference pixel row of N+1 row in the reference pixel row Arrange the absolute value of the difference of intersecting pixel with the reference pixel, the 8th absolute value is the in reference pixel row The absolute value of the difference for the pixel that the pixel of N+1 rows intersects with the reference pixel row and reference pixel row；If institute The 7th absolute value is stated more than the 8th absolute value, it is determined that the pixel that N+1 is arranged in the reference pixel row is described the The CPV predicted values of two subregions；If the 7th absolute value is less than or equal to the 8th absolute value, it is determined that the in reference pixel row The pixel of N+1 rows is the CPV predicted values of second subregion.

In another embodiment, determining unit 902, specifically can be used for the 1st in the present encoding depth unit When the pixel of row Nth column and the pixel of the 1st row Nth row belong to first subregion or belong to second subregion, The first pixel is determined as the CPV predicted values of first subregion in reference pixel row, and in the reference pixel Determine the second pixel as the CPV predicted values of second subregion in row.

In another embodiment, determining unit 902, specifically can be used for the 1st in the present encoding depth unit When the pixel of row Nth column and the pixel of the 1st row Nth row belong to first subregion or belong to second subregion, The first pixel is determined in the reference pixel row as the CPV predicted values of first subregion, and in the reference pixel Determine the second pixel as the CPV predicted values of second subregion in row.

In another embodiment, determining unit 902, specifically can be used for the 1st in the present encoding depth unit The pixel and the pixel of the 1st row Nth column that row the 1st is arranged belong to first subregion, and the pixel of the 1st row Nth row belongs to During second subregion, the pixel for determining N/2 row in the reference pixel row is the CPV predicted values of first subregion, The pixel for determining N+1 rows in the reference pixel row is the CPV predicted values of second subregion.

In another embodiment, determining unit 902, specifically can be used for the 1st in the present encoding depth unit The pixel of row Nth column belongs to second subregion, and the 1st arranges the pixel of the 1st row and the pixel of the 1st row Nth row belongs to During first subregion, the pixel for determining N/2 rows in the reference pixel row is the CPV predicted values of first subregion, The pixel for determining N+1 row in the reference pixel row is the CPV predicted values of second subregion.

In another embodiment, determining unit 902, specifically can be used for the 1st in the present encoding depth unit When the pixel of row Nth column and the pixel of the 1st row Nth row belong to first subregion or belong to second subregion, The pixel for determining the 1st row in the reference pixel row is the CPV predicted values of first subregion；Or, determine the reference Pixel column and the CPV predicted values that the intersecting pixel of reference pixel row is first subregion；Or, determine the reference The pixel of the 1st row is the CPV predicted values of first subregion in pixel column.

In another embodiment, determining unit 902, specifically can be used for the 1st row in the present encoding depth unit The pixel of Nth column and the pixel of the 1st row Nth row are belonged in first subregion, and the present encoding depth unit When the pixel of Nth row Nth column belongs to the second subregion, determine that the pixel that 2N is arranged in the reference pixel row is described second The CPV predicted values of subregion；Or, determine that the pixel of 2N rows in the reference pixel row is pre- for the CPV of second subregion Measured value.

In another embodiment, determining unit 902, specifically can be used for the 1st row in the present encoding depth unit The pixel of Nth column and the pixel of the 1st row Nth row are belonged in first subregion, and the present encoding depth unit When the pixel of Nth row Nth column belongs to the second subregion, compare the size of the first absolute value and the second absolute value, wherein, described One absolute value is the pixel and the absolute value of the difference of the 1st pixel arranged that 2N is arranged in the reference pixel row, described the Two absolute values are the absolute value of the difference of the pixel of the pixel of 2N rows and the 1st row during the reference pixel is arranged；If described First absolute value is more than second absolute value, it is determined that the pixel that 2N is arranged in the reference pixel row is described second point The CPV value in area；If first absolute value is less than or equal to second absolute value, it is determined that 2N rows in the reference pixel row Pixel be above-mentioned second subregion CPV value.

In another embodiment, determining unit 902, specifically can be used for the 1st in the present encoding depth unit The pixel of row Nth column and the pixel of the 1st row Nth row belong to first subregion, and the present encoding depth unit When the pixel of middle Nth row Nth column belongs to the second subregion, compare the size of the 3rd absolute value and the 4th absolute value, wherein, it is described 3rd absolute value is the pixel and the reference pixel row and the reference pixel row phase of 2N row in the reference pixel row The absolute value of the difference of the pixel of friendship, the 4th absolute value be the reference pixel arrange in 2N rows pixel with it is described The absolute value of the difference of reference pixel row and the intersecting pixel of reference pixel row；If the 3rd absolute value is more than described 4th absolute value, it is determined that the pixel that 2N is arranged in the reference pixel row is the CPV value of second subregion；If described Three absolute values are less than or equal to the 4th absolute value, it is determined that the pixel of 2N rows is above-mentioned second in the reference pixel row The CPV value of subregion.

In another embodiment, determining unit 902, specifically can be used for the 1st in the present encoding depth unit The pixel of the pixel of row Nth column, the pixel of the 1st row Nth row and Nth row Nth column belongs to second subregion, and institute When the pixel for stating the 1st row the 1st row in present encoding depth unit belongs to first subregion, determine in the reference pixel row The pixel of N+1 row is the CPV predicted values of second subregion；Or, determine the picture of N+1 rows in the reference pixel row Vegetarian refreshments is the CPV predicted values of second subregion.

In another embodiment, determining unit 902, specifically can be used for the 1st in the present encoding depth unit The pixel of the pixel of row Nth column, the pixel of the 1st row Nth row and Nth row Nth column belongs to second subregion, and institute When the pixel for stating the 1st row the 1st row in present encoding depth unit belongs to first subregion, compare the 5th absolute value and the 6th The size of absolute value, wherein, the 5th absolute value is the pixel and the 1st row pixel of N+1 row in the reference pixel row The absolute value of the difference of point, the 6th absolute value is N+1 rows pixel and the 1st row pixel during the reference pixel is arranged The absolute value of difference；If the 5th absolute value is more than the 6th absolute value, it is determined that N+1 is arranged in the reference pixel row Pixel be second subregion CPV predicted values；If the 5th absolute value is less than or equal to the 6th absolute value, really N+1 rows pixel is the CPV predicted values of second subregion in the fixed reference pixel row.

In another embodiment, determining unit 902, specifically can be used for the 1st in the present encoding depth unit The pixel of the pixel of row Nth column, the pixel of the 1st row Nth row and Nth row Nth column belongs to second subregion, and institute When the pixel for stating the 1st row the 1st row in present encoding depth unit belongs to first subregion, compare the 7th absolute value and the 8th The size of absolute value, the 7th absolute value is the pixel and the reference pixel row of N+1 row in the reference pixel row Arrange the absolute value of the difference of intersecting pixel with the reference pixel, the 8th absolute value is the in reference pixel row The absolute value of the difference for the pixel that the pixel of N+1 rows intersects with the reference pixel row and reference pixel row；If institute The 7th absolute value is stated more than the 8th absolute value, it is determined that the pixel that N+1 is arranged in the reference pixel row is described the The CPV predicted values of two subregions；If the 7th absolute value is less than or equal to the 8th absolute value, it is determined that the in reference pixel row The pixel of N+1 rows is the CPV predicted values of second subregion.

The structural representation for the image infra-frame prediction device embodiment two that Figure 10 provides for the present invention, as shown in Figure 10, should Device includes：Memory 110 and processor 111, wherein：

Memory 110 is used for store instruction collection.The processor 111 is configured as calling the instruction set in memory 110, with Perform following flow：It is the first subregion and the second subregion by present encoding depth dividing elements；According to first subregion and The distribution of pixel in two subregions, determines the first pixel as the subregion constant value of first subregion in reference pixel CPV predicted values, determine in reference pixel the CPV predicted values that the second pixel is second subregion；According to described first point The CPV predicted values in area and the CPV predicted values of second subregion carry out infra-frame prediction to the present encoding depth unit.

Processor 111, is arranged specifically for the pixel of the 1st row Nth column in the present encoding depth unit and the 1st When the pixel of N rows belongs to first subregion or belongs to second subregion, is determined in the reference pixel row One pixel as first subregion CPV predicted values.

Alternatively, processor 111, specifically in the present encoding depth unit pixel of the 1st row Nth column and When the pixel of 1st row Nth row belongs to first subregion or belongs to second subregion, in the reference pixel row Determine the second pixel as the subregion constant value C PV predicted values of second subregion.

Processor 111, specifically in the present encoding depth unit pixel of the 1st row Nth column, the 1st arrange N The pixel of capable pixel and the 1st row the 1st row belongs to N in first subregion, and the present encoding depth unit When the pixel of row Nth column belongs to second subregion, the pixel for determining the 1st row in the reference pixel row is described first The CPV predicted values of subregion；Or, determine that the reference pixel row and the intersecting pixel of reference pixel row are described first The CPV predicted values of subregion.

Processor 111, specifically in the present encoding depth unit pixel of the 1st row Nth column, the 1st arrange N The pixel of capable pixel and the 1st row the 1st row belongs to N in first subregion, and the present encoding depth unit When the pixel of row Nth column belongs to the second subregion, determine that the pixel that 2N is arranged in the reference pixel row is described second point The CPV predicted values in area.

Processor 111, specifically in the present encoding depth unit pixel of the 1st row Nth column, the 1st arrange N The pixel of capable pixel and the 1st row the 1st row belongs to N in first subregion, and the present encoding depth unit When the pixel of row Nth column belongs to the second subregion, compare the size of the first absolute value and the second absolute value, wherein, described first Absolute value is the pixels arranged of 2N in the reference pixel row and the absolute value of the difference of the pixel of the 1st row, described second Absolute value is the absolute value of the difference of the pixel of the pixel of 2N rows and the 1st row during the reference pixel is arranged；If described One absolute value is more than second absolute value, it is determined that the pixel that 2N is arranged in the reference pixel row is second subregion CPV value；If first absolute value is less than or equal to second absolute value, it is determined that 2N rows in the reference pixel row Pixel is the CPV value of above-mentioned second subregion.

Processor 111, specifically in the present encoding depth unit pixel of the 1st row Nth column, the 1st arrange N The pixel of capable pixel and the 1st row the 1st row belongs to N in first subregion, and the present encoding depth unit When the pixel of row Nth column belongs to the second subregion, compare the size of the 3rd absolute value and the 4th absolute value, wherein, the described 3rd Absolute value is that the pixel of 2N row in the reference pixel row intersects with the reference pixel row and reference pixel row The absolute value of the difference of pixel, the 4th absolute value is the pixel of 2N rows and the reference during the reference pixel is arranged The absolute value of the difference of pixel column and the intersecting pixel of reference pixel row；If the 3rd absolute value is more than the described 4th Absolute value, it is determined that the pixel that 2N is arranged in the reference pixel row is the CPV value of second subregion；If the described 3rd is exhausted 4th absolute value is less than or equal to value, it is determined that the pixel of 2N rows is above-mentioned second subregion in the reference pixel row CPV value.

Further, processor 111, specifically for the pixel of the 1st row Nth column in the present encoding depth unit The pixel of point, the pixel of the 1st row Nth row and Nth row Nth column belongs to second subregion, and the present encoding is deep When the pixel that the 1st row the 1st is arranged in degree unit belongs to first subregion, the pixel of the 1st row in the reference pixel row is determined Point is the CPV predicted values of first subregion；Or, determine that the reference pixel row arranges the pixel intersected with the reference pixel Point is the CPV predicted values of first subregion.

Processor 111, specifically in the present encoding depth unit pixel of the 1st row Nth column, the 1st arrange N Capable pixel, the pixel of Nth row Nth column belong to the 1st row in second subregion, and the present encoding depth unit When the pixel of 1st row belongs to first subregion, determine that the pixel that N+1 is arranged in the reference pixel row is described second The CPV predicted values of subregion.

Processor 111, specifically for the pixel of the 1st row Nth column, the 1st row Nth row in the present encoding depth unit Pixel, the pixel of Nth row Nth column belong to the 1st row in second subregion, and the present encoding depth unit When the pixel of 1 row belongs to first subregion, compare the size of the 5th absolute value and the 6th absolute value, wherein, the described 5th Absolute value is the pixels arranged of N+1 in the reference pixel row and the absolute value of the difference of the 1st row pixel, and the described 6th is exhausted To being worth for the absolute value of the difference of N+1 rows pixel and the 1st row pixel in reference pixel row；If the described 5th is absolute Value is more than the 6th absolute value, it is determined that the pixel that N+1 is arranged in the reference pixel row is the CPV of second subregion Predicted value；If the 5th absolute value is less than or equal to the 6th absolute value, it is determined that N+1 row pictures in the reference pixel row Vegetarian refreshments is the CPV predicted values of second subregion.

Processor 111, specifically in the present encoding depth unit pixel of the 1st row Nth column, the 1st arrange N Capable pixel, the pixel of Nth row Nth column belong to the 1st row in second subregion, and the present encoding depth unit When the pixel of 1st row belongs to first subregion, compare the size of the 7th absolute value and the 8th absolute value, the four-line poem with seven characters to a line pair It is worth the pixel that the pixel arranged for N+1 in the reference pixel row intersects with the reference pixel row and reference pixel row The absolute value of the difference of point, the 8th absolute value is the pixel and the reference image of N+1 rows during the reference pixel is arranged The absolute value of the difference of plain row and the intersecting pixel of reference pixel row；If it is exhausted that the 7th absolute value is more than the described 8th To value, it is determined that the pixel that N+1 is arranged in the reference pixel row is the CPV predicted values of second subregion；If described Seven absolute values are less than or equal to the 8th absolute value, it is determined that the pixel of N+1 rows is described second point in the reference pixel row The CPV predicted values in area.

Alternatively, processor 111, specifically in the present encoding depth unit pixel of the 1st row Nth column and When the pixel of 1st row Nth row belongs to first subregion or belongs to second subregion, in reference pixel row The first pixel is determined as the CPV predicted values of first subregion, and the second pixel is determined in reference pixel row It is used as the CPV predicted values of second subregion.

Alternatively, processor 111, specifically in the present encoding depth unit pixel of the 1st row Nth column and When the pixel of 1st row Nth row belongs to first subregion or belongs to second subregion, in the reference pixel row The first pixel is determined as the CPV predicted values of first subregion, and the second pixel is determined in reference pixel row It is used as the CPV predicted values of second subregion.

Alternatively, processor 111, specifically in the present encoding depth unit pixel of the 1st row Nth column and When the pixel of 1st row Nth row belongs to first subregion or belongs to second subregion, in reference pixel row The first pixel is determined as the CPV predicted values of first subregion, and the second pixel is determined in the reference pixel row It is used as the CPV predicted values of second subregion.

In another embodiment, processor 111, specifically for the picture that the 1st row the 1st is arranged in the present encoding depth unit Vegetarian refreshments and the pixel of the 1st row Nth column belong to first subregion, and the pixel of the 1st row Nth row belongs to second subregion When, determine that the pixel that N/2 is arranged in the reference pixel row is the CPV predicted values of first subregion, determine the reference The pixel of N+1 rows is the CPV predicted values of second subregion in pixel column.

In another embodiment, processor 111, specifically for the picture of the 1st row Nth column in the present encoding depth unit Vegetarian refreshments belongs to second subregion, and the 1st arranges the pixel of the 1st row and the pixel of the 1st row Nth row belongs to first subregion When, the pixel for determining N/2 rows in the reference pixel row is the CPV predicted values of first subregion, determines the reference The pixel that N+1 is arranged in pixel column is the CPV predicted values of second subregion.

Processor 111, is arranged specifically for the pixel of the 1st row Nth column in the present encoding depth unit and the 1st When the pixel of N rows belongs to first subregion or belongs to second subregion, the 1st row in the reference pixel row are determined Pixel be first subregion CPV predicted values；Or, determine that the reference pixel row and reference pixel row are intersecting Pixel be first subregion CPV predicted values；Or, the pixel for determining the 1st row in the reference pixel row is institute State the CPV predicted values of the first subregion.

Processor 111, N is arranged specifically for the pixel and the 1st of the 1st row Nth column in the present encoding depth unit The pixel that capable pixel belongs to Nth row Nth column in first subregion, and the present encoding depth unit belongs to During two subregions, the pixel for determining 2N row in the reference pixel row is the CPV predicted values of second subregion；Or, really The pixel of 2N rows is the CPV predicted values of second subregion in the fixed reference pixel row.

Processor 111, N is arranged specifically for the pixel and the 1st of the 1st row Nth column in the present encoding depth unit The pixel that capable pixel belongs to Nth row Nth column in first subregion, and the present encoding depth unit belongs to During two subregions, compare the size of the first absolute value and the second absolute value, wherein, first absolute value is the reference pixel row In 2N arrange pixel with the 1st row pixel difference absolute value, second absolute value be the reference pixel row In 2N rows pixel and the 1st row pixel difference absolute value；If it is exhausted that first absolute value is more than described second To value, it is determined that the pixel that 2N is arranged in the reference pixel row is the CPV value of second subregion；If described first is absolute Value is less than or equal to second absolute value, it is determined that the pixel of 2N rows is above-mentioned second subregion in the reference pixel row CPV value.

Processor 111, is arranged specifically for the pixel of the 1st row Nth column in the present encoding depth unit and the 1st The pixel that the pixel of N rows belongs to Nth row Nth column in first subregion, and the present encoding depth unit belongs to During the second subregion, compare the size of the 3rd absolute value and the 4th absolute value, wherein, the 3rd absolute value is the reference pixel The absolute value of the difference for the pixel that the pixel that 2N is arranged in row intersects with the reference pixel row and reference pixel row, 4th absolute value is the pixel and the reference pixel row and the reference pixel of 2N rows during the reference pixel is arranged The absolute value of the difference of the intersecting pixel of row；If the 3rd absolute value is more than the 4th absolute value, it is determined that the ginseng Examine the CPV value that the pixel that 2N is arranged in pixel column is second subregion；If the 3rd absolute value is less than or equal to described the Four absolute values, it is determined that the pixel of 2N rows is the CPV value of above-mentioned second subregion in the reference pixel row.

Processor 111, specifically in the present encoding depth unit pixel of the 1st row Nth column, the 1st arrange N Capable pixel and the pixel of Nth row Nth column are belonged to the 1st in second subregion, and the present encoding depth unit The pixel that row the 1st is arranged is when belonging to first subregion, and it is described the to determine N+1 is arranged in the reference pixel row pixel The CPV predicted values of two subregions；Or, the pixel for determining N+1 rows in the reference pixel row is the CPV of second subregion Predicted value.

Processor 111, specifically in the present encoding depth unit pixel of the 1st row Nth column, the 1st arrange N Capable pixel and the pixel of Nth row Nth column are belonged to the 1st in second subregion, and the present encoding depth unit When the pixel that row the 1st is arranged belongs to first subregion, compare the size of the 5th absolute value and the 6th absolute value, wherein, it is described 5th absolute value is the absolute value of the difference of the pixel that N+1 is arranged in the reference pixel row and the 1st row pixel, described the Six absolute values are the absolute value of the difference of N+1 rows pixel and the 1st row pixel during the reference pixel is arranged；If the described 5th Absolute value is more than the 6th absolute value, it is determined that the pixel that N+1 is arranged in the reference pixel row is second subregion CPV predicted values；If the 5th absolute value is less than or equal to the 6th absolute value, it is determined that N+ in the reference pixel row 1 row pixel is the CPV predicted values of second subregion.

Processor 111, specifically in the present encoding depth unit pixel of the 1st row Nth column, the 1st arrange N Capable pixel and the pixel of Nth row Nth column are belonged to the 1st in second subregion, and the present encoding depth unit When the pixel that row the 1st is arranged belongs to first subregion, compare the size of the 7th absolute value and the 8th absolute value, the four-line poem with seven characters to a line To being worth the picture that the pixel arranged for N+1 in the reference pixel row intersects with the reference pixel row and reference pixel row The absolute value of the difference of vegetarian refreshments, the 8th absolute value is the pixel of N+1 rows and the reference during the reference pixel is arranged The absolute value of the difference of pixel column and the intersecting pixel of reference pixel row；If the 7th absolute value is more than the described 8th Absolute value, it is determined that the pixel that N+1 is arranged in the reference pixel row is the CPV predicted values of second subregion；If described 7th absolute value is less than or equal to the 8th absolute value, it is determined that the pixel of N+1 rows is described second in the reference pixel row The CPV predicted values of subregion.

, can be by it in several embodiments provided by the present invention, it should be understood that disclosed apparatus and method Its mode is realized.For example, device embodiment described above is only schematical, for example, the division of the unit, only Only a kind of division of logic function, can there is other dividing mode when actually realizing, such as multiple units or component can be tied Another system is closed or is desirably integrated into, or some features can be ignored, or do not perform.It is another, it is shown or discussed Coupling each other or direct-coupling or communication connection can be the INDIRECT COUPLINGs or logical of device or unit by some interfaces Letter connection, can be electrical, machinery or other forms.

The unit illustrated as separating component can be or may not be it is physically separate, it is aobvious as unit The part shown can be or may not be physical location, you can with positioned at a place, or can also be distributed to multiple On NE.Some or all of unit therein can be selected to realize the mesh of this embodiment scheme according to the actual needs 's.

In addition, each functional unit in each embodiment of the invention can be integrated in a processing unit, can also That unit is individually physically present, can also two or more units it is integrated in a unit.Above-mentioned integrated list Member can both be realized in the form of hardware, it would however also be possible to employ hardware adds the form of SFU software functional unit to realize.

The above-mentioned integrated unit realized in the form of SFU software functional unit, can be stored in an embodied on computer readable and deposit In storage media.Above-mentioned SFU software functional unit is stored in a storage medium, including some instructions are to cause a computer Equipment（Can be personal computer, server, or network equipment etc.）Or processor（processor）Perform the present invention each The part steps of embodiment methods described.And foregoing storage medium includes：USB flash disk, mobile hard disk, read-only storage（Read- Only Memory, ROM）, random access memory（Random Access Memory, RAM）, magnetic disc or CD etc. it is various Can be with the medium of store program codes.

Finally it should be noted that：Various embodiments above is merely illustrative of the technical solution of the present invention, rather than its limitations；To the greatest extent The present invention is described in detail with reference to foregoing embodiments for pipe, it will be understood by those within the art that：Its according to The technical scheme described in foregoing embodiments can so be modified, or which part or all technical characteristic are entered Row equivalent substitution；And these modifications or replacement, the essence of appropriate technical solution is departed from various embodiments of the present invention technology The scope of scheme.

Claims

1. a kind of image intra prediction method, it is characterised in that including：

According to the distribution of pixel in first subregion and the second subregion, determine the first pixel as institute in reference pixel State the subregion constant value C PV predicted values of the first subregion, CPV of second pixel for second subregion is determined in reference pixel Predicted value；

According to the CPV predicted values of first subregion and the CPV predicted values of second subregion to the present encoding depth list Member carries out infra-frame prediction；

Wherein, the present encoding depth unit includes N × N number of pixel；The reference pixel includes reference pixel row and ginseng Examine pixel column；The reference pixel row is not belonging to the present encoding depth unit, and the reference pixel row is worked as with described The adjacent one-row pixels point of the preceding row pixel of coding depth unit the 1st, the 1st row of the reference pixel row and the present encoding 1st column alignment of depth unit；The reference pixel row are not belonging to the present encoding depth unit, and the reference pixel is arranged A row pixel adjacent with the 1st row pixel of the present encoding depth unit, and the 1st row of reference pixel row with The 1st row alignment of the present encoding depth unit；The reference pixel row and the reference pixel arrange intersecting pixel The starting point of reference pixel row and the reference pixel row；

Wherein, the pixel and the pixel of the 1st row Nth column that the 1st row the 1st is arranged in the present encoding depth unit belong to described First subregion, and the pixel of the 1st row Nth row belongs to second subregion；

It is described to determine the first pixel as the subregion constant value C PV predicted values of first subregion, in ginseng in reference pixel The CPV predicted values for determining that the second pixel is second subregion in pixel are examined, including：

Determine that the pixel that N/2 is arranged in the reference pixel row is the CPV predicted values of first subregion, determine the reference The pixel of N+1 rows is the CPV predicted values of second subregion in pixel column；Or,

Wherein, the pixel of the 1st row Nth column belongs to second subregion in the present encoding depth unit, and the 1st arranges the 1st Capable pixel and the pixel of the 1st row Nth row belong to first subregion；

The pixel for determining N/2 rows in the reference pixel row is the CPV predicted values of first subregion, determines the reference The pixel that N+1 is arranged in pixel column is the CPV predicted values of second subregion；Or,

Wherein, the pixel of the 1st row Nth column and the pixel of the 1st row Nth row belong to institute in the present encoding depth unit State the first subregion or belong to second subregion；

It is described to determine that the first pixel includes as the subregion constant value C PV predicted values of first subregion in reference pixel：

Determine the CPV predicted values of the reference pixel row and the intersecting pixel of reference pixel row for first subregion；

Or,

2. according to the method described in claim 1, it is characterised in that the picture of the 1st row Nth column in the present encoding depth unit Vegetarian refreshments and the pixel of the 1st row Nth row belong to Nth row Nth column in first subregion, and the present encoding depth unit Pixel belong to the second subregion；

3. according to the method described in claim 1, it is characterised in that the picture of the 1st row Nth column in the present encoding depth unit Vegetarian refreshments and the pixel of the 1st row Nth row belong to Nth row Nth column in first subregion, and the present encoding depth unit Pixel belong to the second subregion；

Compare the size of the first absolute value and the second absolute value, wherein, first absolute value is the in the reference pixel row The pixel of 2N row and the absolute value of the difference of the pixel of the 1st row, second absolute value are the during the reference pixel is arranged The absolute value of the difference of the pixel of the pixel of 2N rows and the 1st row；

If first absolute value is more than second absolute value, it is determined that the pixel of 2N row is in the reference pixel row The CPV value of second subregion；

If first absolute value is less than or equal to second absolute value, it is determined that the pixel of 2N rows in the reference pixel row Point is the CPV value of above-mentioned second subregion.

4. according to the method described in claim 1, it is characterised in that the 1st row Nth column in the present encoding depth unit Pixel and the pixel of the 1st row Nth row belong to Nth row N in first subregion, and the present encoding depth unit The pixel of row belongs to the second subregion；

Compare the size of the 3rd absolute value and the 4th absolute value, wherein, the 3rd absolute value is the in the reference pixel row Pixel and the reference pixel row and the reference pixel of 2N row arrange the absolute value of the difference of the pixel intersected, and described the Four absolute values are that the pixel of 2N rows intersects with the reference pixel row and reference pixel row during the reference pixel is arranged Pixel difference absolute value；

If the 3rd absolute value is more than the 4th absolute value, it is determined that the pixel of 2N row is in the reference pixel row The CPV value of second subregion；

If the 3rd absolute value is less than or equal to the 4th absolute value, it is determined that the pixel of 2N rows in the reference pixel row Point is the CPV value of above-mentioned second subregion.

5. according to the method described in claim 1, it is characterised in that the picture of the 1st row Nth column in the present encoding depth unit The pixel of vegetarian refreshments, the pixel of the 1st row Nth row and Nth row Nth column belongs to second subregion, and the present encoding The pixel that the 1st row the 1st is arranged in depth unit belongs to first subregion；

Or,

6. according to the method described in claim 1, it is characterised in that the picture of the 1st row Nth column in the present encoding depth unit The pixel of vegetarian refreshments, the pixel of the 1st row Nth row and Nth row Nth column belongs to second subregion, and the present encoding The pixel that the 1st row the 1st is arranged in depth unit belongs to first subregion；

Compare the size of the 5th absolute value and the 6th absolute value, wherein, the 5th absolute value is N in the reference pixel row The absolute value of the pixel of+1 row and the difference of the 1st row pixel, the 6th absolute value is N+1 in reference pixel row The absolute value of the difference of row pixel and the 1st row pixel；

If the 5th absolute value is more than the 6th absolute value, it is determined that the pixel that N+1 is arranged in the reference pixel row For the CPV predicted values of second subregion；

If the 5th absolute value is less than or equal to the 6th absolute value, it is determined that N+1 row pixels in the reference pixel row Point is the CPV predicted values of second subregion.

7. according to the method described in claim 1, it is characterised in that the picture of the 1st row Nth column in the present encoding depth unit The pixel of vegetarian refreshments, the pixel of the 1st row Nth row and Nth row Nth column belongs to second subregion, and the present encoding The pixel that the 1st row the 1st is arranged in depth unit belongs to first subregion；

Compare the size of the 7th absolute value and the 8th absolute value, the 7th absolute value is N+1 row in the reference pixel row Pixel and the reference pixel row and the absolute value of the difference of pixel that intersects of reference pixel row, the described 8th is exhausted To being worth the picture intersected is arranged for the pixel of N+1 rows in reference pixel row with the reference pixel row and the reference pixel The absolute value of the difference of vegetarian refreshments；

If the 7th absolute value is more than the 8th absolute value, it is determined that the pixel that N+1 is arranged in the reference pixel row For the CPV predicted values of second subregion；

If the 7th absolute value is less than or equal to the 8th absolute value, it is determined that the pixel of N+1 rows in the reference pixel row For the CPV predicted values of second subregion.

8. a kind of image infra-frame prediction device, it is characterised in that including：

Determining unit, for the distribution according to pixel in first subregion and the second subregion, determines in reference pixel One pixel determines that the second pixel is institute as the subregion constant value C PV predicted values of first subregion, in reference pixel State the CPV predicted values of the second subregion；

Coding unit, works as the CPV predicted values according to first subregion and the CPV predicted values of second subregion to described Preceding coding depth unit carries out infra-frame prediction；

Wherein, the determining unit, specifically for the pixel and the 1st that the 1st row the 1st is arranged in the present encoding depth unit The pixel of row Nth column belongs to first subregion, and the pixel of the 1st row Nth row is when belonging to second subregion, determines institute The CPV predicted values that the pixel that N/2 in reference pixel row arranges is first subregion are stated, the is determined in the reference pixel row The pixel of N+1 rows is the CPV predicted values of second subregion；Or,

Wherein, the determining unit, belongs to specifically for the pixel of the 1st row Nth column in the present encoding depth unit Second subregion, and the 1st arrange the pixel of the 1st row and when the pixel of the 1st row Nth row belongs to first subregion, it is determined that The pixel of N/2 rows is the CPV predicted values of first subregion in the reference pixel row, is determined in the reference pixel row The pixel of N+1 row is the CPV predicted values of second subregion；Or,

Wherein, the determining unit, specifically for the pixel and the 1st of the 1st row Nth column in the present encoding depth unit When the pixel of row Nth row belongs to first subregion or belongs to second subregion, determine in the reference pixel row The pixel of 1st row is the CPV predicted values of first subregion；Or, determine the reference pixel row and the reference pixel The intersecting pixel of row is the CPV predicted values of first subregion；Or, determine the pixel of the 1st row in the reference pixel row Point is the CPV predicted values of first subregion.

9. device according to claim 8, it is characterised in that the determining unit, deep specifically for the present encoding The pixel of the 1st row Nth column and the pixel of the 1st row Nth row belong to first subregion, and the current volume in degree unit When the pixel of Nth row Nth column belongs to the second subregion in code depth unit, the pixel that 2N is arranged in the reference pixel row is determined Point is the CPV predicted values of second subregion；Or, the pixel for determining 2N rows in the reference pixel row is described second The CPV predicted values of subregion.

10. device according to claim 8, it is characterised in that the determining unit, deep specifically for the present encoding The pixel of the 1st row Nth column and the pixel of the 1st row Nth row belong to first subregion, and the current volume in degree unit When the pixel of Nth row Nth column belongs to the second subregion in code depth unit, compare the big of the first absolute value and the second absolute value It is small, wherein, first absolute value is the pixel of 2N row in the reference pixel row and the difference of the 1st pixel arranged Absolute value, second absolute value is the difference of the pixel of the pixel of 2N rows and the 1st row during the reference pixel is arranged Absolute value；If first absolute value is more than second absolute value, it is determined that the pixel that 2N is arranged in the reference pixel row Point is the CPV value of second subregion；If first absolute value is less than or equal to second absolute value, it is determined that the reference The pixel of 2N rows is the CPV value of above-mentioned second subregion in pixel column.

11. device according to claim 8, it is characterised in that the determining unit, specifically in the present encoding The pixel of the 1st row Nth column and the pixel of the 1st row Nth row belong to first subregion in depth unit, and described current When the pixel of Nth row Nth column belongs to the second subregion in coding depth unit, compare the big of the 3rd absolute value and the 4th absolute value It is small, wherein, the 3rd absolute value is pixel and the reference pixel row that 2N is arranged in the reference pixel row and described The absolute value of the difference of the intersecting pixel of reference pixel row, the 4th absolute value is 2N rows during the reference pixel is arranged The absolute value of the difference for the pixel that pixel intersects with the reference pixel row and reference pixel row；If the described 3rd is exhausted 4th absolute value is more than to value, it is determined that the pixel that 2N is arranged in the reference pixel row is second subregion CPV value；If the 3rd absolute value is less than or equal to the 4th absolute value, it is determined that the picture of 2N rows in the reference pixel row Vegetarian refreshments is the CPV value of above-mentioned second subregion.

12. device according to claim 8, it is characterised in that the determining unit, specifically in the present encoding The pixel of the 1st row Nth column, the pixel of the 1st row Nth row and the pixel of Nth row Nth column belong to described in depth unit When the pixel that the 1st row the 1st is arranged in second subregion, and the present encoding depth unit belongs to first subregion, institute is determined State the CPV predicted values that the pixel that N+1 is arranged in reference pixel row is second subregion；Or, determine the reference pixel The pixel of N+1 rows is the CPV predicted values of second subregion in row.

13. device according to claim 8, it is characterised in that the determining unit, specifically in the present encoding The pixel of the 1st row Nth column, the pixel of the 1st row Nth row and the pixel of Nth row Nth column belong to described in depth unit The pixel that the 1st row the 1st is arranged in second subregion, and the present encoding depth unit is when belonging to first subregion, compares the The size of five absolute values and the 6th absolute value, wherein, the 5th absolute value is the pixel of N+1 row in the reference pixel row Point and the absolute value of the difference of the 1st row pixel, the 6th absolute value be in the reference pixel row N+1 rows pixel and The absolute value of the difference of 1st row pixel；If the 5th absolute value is more than the 6th absolute value, it is determined that the reference image The pixel that N+1 is arranged in plain row is the CPV predicted values of second subregion；If the 5th absolute value is less than or equal to described the Six absolute values, it is determined that N+1 rows pixel is the CPV predicted values of second subregion in the reference pixel row.

14. device according to claim 8, it is characterised in that the determining unit, specifically in the present encoding The pixel of the 1st row Nth column, the pixel of the 1st row Nth row and the pixel of Nth row Nth column belong to described in depth unit The pixel that the 1st row the 1st is arranged in second subregion, and the present encoding depth unit is when belonging to first subregion, compares the The size of seven absolute values and the 8th absolute value, the 7th absolute value be the reference pixel row in N+1 arrange pixel with The absolute value of the difference of the intersecting pixel of reference pixel row and the reference pixel row, the 8th absolute value is described The difference for the pixel that the pixel of N+1 rows intersects with the reference pixel row and reference pixel row in reference pixel row Absolute value；If the 7th absolute value is more than the 8th absolute value, it is determined that the picture that N+1 is arranged in the reference pixel row Vegetarian refreshments is the CPV predicted values of second subregion；If the 7th absolute value is less than or equal to the 8th absolute value, it is determined that the ginseng The pixel for examining N+1 rows in pixel column is the CPV predicted values of second subregion.