Image component predicting method, device and computer storage medium

Abstract

To provide image component predicting method, device and computer storage medium that integrate a model parameter driving process without altering encoding / decoding prediction performance.SOLUTION: A method includes S501 obtaining a first reference pixel set corresponding to a predicted image component of a coding block in a video image, S502 using a preset component value as a predicted value corresponding to the predicted image component when the number of valid pixel points is less than the number of presets, S503 screening the first reference pixel set to obtain a second reference pixel set when the number of valid pixel points is larger than or equal to the number of presets, S504 using the preset component value as the predicted value corresponding to the predicted image component, and S505 implementing prediction processing for a predicted image component which corresponds to the predicted image component to be predicted by determining model parameters via the second reference pixel set, and obtaining a prediction model to be used to obtain the predicted value.SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] The present application relates to the field of video encoding and decoding technology, and in particular to a method, apparatus and computer storage medium for predicting image components. [Background technology]

[0002] With the increasing demands of people for video display quality, new forms of video applications such as high definition and ultra-high definition video have emerged. H.265 / High Efficiency Video Coding (HEVC) cannot meet the demands of the rapid development of video applications, so the Joint Video Exploration Team (JVET) has proposed the next-generation video coding standard H.266 / Versatile Video Coding (VVC), and the corresponding test model is the VVC Reference Software Test Platform (VTM: VVC Test Model).

[0003] Currently, a prediction model-based image component prediction method is integrated in VTM, and the chrominance component can be predicted by the luminance component of the current coding block (CB) through the prediction model. However, when constructing the prediction model, the number of adjacent reference pixel points used to derive the model parameters is different, which not only increases additional processing but also increases the computational complexity. Summary of the Invention [Problem to be solved by the invention]

[0004] The present embodiment provides a method, apparatus, and computer storage medium for predicting image components, which integrates the derivation process of model parameters without changing the encoding and decoding prediction performance, and at the same time, when the number of valid pixel points in the adjacent reference pixel set is smaller than the preset number, no additional processing modules are added, so that no additional processing is required and the calculation complexity is reduced. [Means for solving the problem]

[0005] The technical solution of the present embodiment can be realized as follows.

[0006] In a first aspect, the present embodiments provide a method for predicting an image component, the method comprising: Obtaining a first set of reference pixels corresponding to a predicted image component of a coding block in a video image; if the number of valid pixel points in the first reference pixel set is less than a preset number, using a preset component value as a predicted value corresponding to the predicted image component; if the number of valid pixel points in the first reference pixel set is greater than or equal to the preset number, screening the first reference pixel set to obtain a second reference pixel set, where the number of valid pixel points in the second reference pixel set is less than or equal to the preset number; When the number of valid pixel points in the second reference pixel set is smaller than the number of presets, using the preset component value as a predicted value corresponding to the predicted image component; and when the number of valid pixel points in the second reference pixel set is equal to the number of presets, determining model parameters via the second reference pixel set and obtaining a prediction model corresponding to the predicted image component according to the model parameters, wherein the prediction model is used to realize a prediction process for the predicted image component and obtain a predicted value corresponding to the predicted image component.

[0007] In a second aspect, the present embodiment provides an apparatus for predicting an image component, comprising an acquisition unit, a prediction unit and a screening unit, wherein: The acquisition unit is configured to acquire a first set of reference pixels corresponding to a predicted image component of a coding block in a video image; the prediction unit is configured to use a preset component value as a prediction value corresponding to the predicted image component when a number of valid pixel points in the first reference pixel set is smaller than a preset number; The screening unit is configured to screen the first reference pixel set to obtain a second reference pixel set when a number of valid pixel points in the first reference pixel set is greater than or equal to a preset number, where the number of valid pixel points in the second reference pixel set is less than or equal to a preset number; The prediction unit is further configured to use a preset component value as a predicted value corresponding to the predicted image component when the number of valid pixel points in the second reference pixel set is smaller than a preset number, and to determine model parameters via the second reference pixel set when the number of valid pixel points in the second reference pixel set is equal to the preset number, and to obtain a prediction model corresponding to the predicted image component according to the model parameters, where the prediction model is used to realize a prediction process for the predicted image component and obtain a predicted value corresponding to the predicted image component.

[0008] In a third aspect, embodiments of the present application provide an apparatus for predicting an image component, comprising a memory and a processor, the memory is configured to store a computer program executable by the processor; The processor, when executing the computer program, is configured to perform the method according to the first aspect.

[0009] In a fourth aspect, an embodiment of the present application provides a computer storage medium storing an image component prediction program, the image component prediction program implementing the method according to the first aspect when executed by at least one processor. Effect of the Invention

[0010] The present embodiment provides a method, an apparatus, and a computer storage medium for predicting an image component, by obtaining a first reference pixel set corresponding to a predicted image component of a coding block in a video image, and if a number of valid pixel points in the first reference pixel set is smaller than a preset number, using the preset component value as a predicted value corresponding to the predicted image component; if a number of valid pixel points in the first reference pixel set is greater than or equal to the preset number, screening the first reference pixel set to obtain a second reference pixel set, the number of valid pixel points in the second reference pixel set is smaller than or equal to the preset number, using the preset component value as a predicted value corresponding to the predicted image component; if a number of valid pixel points in the second reference pixel set is equal to the preset number, determining a model parameter via the second reference pixel set, and obtaining a prediction model corresponding to the predicted image component according to the model parameter; and, The rule is used to realize a prediction process for a predicted image component to obtain a predicted value corresponding to the predicted image component; in this way, when the number of effective pixel points of the first reference pixel set is smaller than the preset number or the number of effective pixel points of the second reference pixel set is smaller than the preset number, the preset default value is directly used as the predicted value corresponding to the predicted image component; only when the number of effective pixel points of the second reference pixel set meets the preset number, the model parameters are determined according to the first reference pixel set to establish a prediction model of the predicted image component, thereby integrating the derivation process of the model parameters; further, when the number of effective pixel points of the first reference pixel set or the second reference pixel set is smaller than the preset number, especially when the effective pixel point is 0 or 2, the additional processing module is not increased, so that the preset default value is directly used as the predicted value corresponding to the predicted image component, no additional processing is required, and the calculation complexity is reduced. [Brief description of the drawings]

[0011] [Figure 1]FIG. 1 is a schematic diagram of a distribution of effective adjacent regions according to an embodiment of the present invention. [Diagram 2] 1 is a schematic diagram of the distribution of selected areas in three modes according to an embodiment of the present invention. [Diagram 3] FIG. 1 is a block diagram of an exemplary configuration of a video encoding system according to an embodiment of the present invention. [Figure 4] FIG. 2 is a block diagram of an exemplary configuration of a video decoding system according to an embodiment of the present invention; [Diagram 5] 4 is an exemplary flowchart of a method for predicting image components according to an embodiment of the present disclosure. [Figure 6A] FIG. 13 is an exemplary structural diagram of the selection of adjacent reference pixel points in the INTRA_LT_CCLM mode according to an embodiment of the present application; [Figure 6B] FIG. 13 is an exemplary structural diagram of the selection of adjacent reference pixel points in the INTRA_L_CCLM mode according to an embodiment of the present application; [Figure 6C] FIG. 13 is an exemplary structural diagram of the selection of adjacent reference pixel points in the INTRA_T_CCLM mode according to an embodiment of the present application; [Figure 7] 13 is an exemplary flowchart of a method for predicting another image component according to an embodiment of the present disclosure. [Figure 8A] FIG. 13 is an exemplary structure diagram for generating 0 effective pixel points in INTRA_LT_CCLM mode according to an embodiment of the present application. [Figure 8B] FIG. 13 is an exemplary structure diagram for generating 0 effective pixel points in INTRA_L_CCLM mode according to an embodiment of the present application. [Figure 8C] FIG. 13 is an exemplary structure diagram for generating 0 effective pixel points in INTRA_T_CCLM mode according to an embodiment of the present application. [Figure 9A] FIG. 13 is an exemplary structure diagram of generating two effective pixel points in INTRA_LT_CCLM mode according to an embodiment of the present application. [Figure 9B] FIG. 13 is an exemplary structure diagram of generating two effective pixel points in INTRA_L_CCLM mode according to an embodiment of the present invention. [Figure 9C] FIG. 13 is an exemplary structure diagram of generating two effective pixel points in INTRA_T_CCLM mode according to an embodiment of the present application. [Figure 10] 4 is an exemplary flow chart for deriving model parameters according to an embodiment of the present disclosure. [Figure 11] 4 is an exemplary simplified flow chart for deriving model parameters according to an embodiment of the present disclosure. [Figure 12] 4 is another exemplary simplified flow chart for deriving model parameters according to an embodiment of the present disclosure. [Figure 13] FIG. 2 is an exemplary structural diagram of an image component prediction device according to an embodiment of the present disclosure; [Figure 14] 2 is an exemplary structural diagram of a specific hardware configuration of an image component prediction device according to an embodiment of the present application; [Figure 15] FIG. 2 is an exemplary structural diagram of an encoder according to an embodiment of the present invention; [Figure 16] FIG. 2 is an exemplary structural diagram of a decoder configuration according to an embodiment of the present application; DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] In order to more fully understand the features and technical contents of the present embodiment, the embodiments of the present embodiment will be described in detail below with reference to the accompanying drawings, which are for reference purposes only and are not intended to limit the present embodiment.

[0013] In a video image, a first image component, a second image component and a third image component are usually used to represent a coding block, where the three image components are one luma component, one blue chroma component and one red chroma component, respectively; specifically, the luma component is usually represented by the code Y, the blue chroma component is usually represented by the code Cb or U, and the red chroma component is usually represented by the code Cr or V; in this way, the video image can be represented in the format of YCbCr, and can also be represented in the format of YUV.

[0014] In the present embodiment, the first image component may be a luminance component, the second image component may be a blue chroma component, and the third image component may be a red chroma component, but is not particularly limited in the present embodiment.

[0015] In the current video image or video encoding and decoding process, the cross-component prediction technology mainly includes a cross-component linear model prediction (CCLM) mode and a multi-directional linear model prediction (MDLM) mode. Regardless of the model parameters derived according to the CCLM mode or the model parameters derived according to the MDLM mode, the corresponding prediction models can all realize prediction between image components, such as from the first image component to the second image component, from the second image component to the first image component, from the first image component to the third image component, from the third image component to the first image component, from the second image component to the third image component, or from the third image component to the second image component.

[0016] Taking prediction from a first image component to a second image component as an example, in order to reduce the redundancy between the first image component and the second image component, CCLM mode is used in VVC, where the first image component and the second image component are of the same coding block, that is, the predicted value of the second image component is constructed according to the reconstructed value of the first image component of the same coding block, as shown in the following equation (1):

number

[0017] JPEG2025014004000003.jpg49170

[0018] For a coding block, the neighboring regions may include a left neighboring region, an upper neighboring region, a lower left neighboring region, and an upper right neighboring region. In VVC, three cross-component linear model prediction modes may be included, which are a left and upper neighboring intra-CCLM mode (which may be represented by INTRA_LT_CCLM mode), a left and lower left neighboring intra-CCLM mode (which may be represented by INTRA_L_CCLM mode), and an upper and upper right neighboring intra-CCLM mode (which may be represented by INTRA_T_CCLM mode). In these three modes, each mode may select a preset number (for example, four) of neighboring reference pixel points to be used for deriving the model parameters α and β, and the biggest difference between these three modes is that the selection regions corresponding to the neighboring reference pixel points for deriving the model parameters α and β are different.

[0019] JPEG2025014004000004.jpg98170

[0020] JPEG2025014004000005.jpg106170

[0021] Referring to FIG. 1, a schematic diagram of the distribution of effective adjacent regions according to an embodiment of the present invention is shown. In FIG. 1, the left adjacent region, the lower left adjacent region, the upper adjacent region, and the upper right adjacent region are all effective. Based on FIG. 1, the selection regions of the three modes are as shown in FIG. 2. Here, in FIG. 2, (a) shows the selection region of the INTRA_LT_CCLM mode, including the left adjacent region and the upper adjacent region, (b) shows the selection region of the INTRA_L_CCLM mode, including the left adjacent region and the lower left adjacent region, and (c) shows the selection region of the INTRA_T_CCLM mode, including the upper adjacent region and the upper right adjacent region. After the selection regions of the three modes are determined in this way, a reference point for model parameter derivation can be selected in the selection region. In this way, the selected reference point is called an adjacent reference pixel point, and the number of adjacent reference pixel points is usually four at most, and the position of the adjacent reference pixel point is usually determined for a coding block of W×H determined by one size.

[0022] However, for some special cases, such as in the case of coding block boundaries, when it is not predictable and when neighboring reference pixel points cannot be obtained due to the coding order, and further, when performing division of the coding block according to tiles or slices, the neighboring region may be invalid, so that the number of neighboring reference pixel points selected from the neighboring region may be less than four, that is, only 0 or 2 neighboring reference pixel points may be selected, thereby making the number of neighboring reference pixel points used for model parameter derivation unintegrated, thereby increasing the additional "duplication" operation and at the same time increasing the computational complexity.

[0023] In order to unify the derivation process of model parameters and reduce the computational complexity while keeping the encoding and decoding prediction performance unchanged, an embodiment of the present application provides a prediction method for an image component, by obtaining a first reference pixel set corresponding to a predicted image component of a coding block in a video image, and if the number of valid pixel points in the first reference pixel set is less than a preset number, using the preset component value as a prediction value corresponding to the predicted image component; if the number of valid pixel points in the first reference pixel set is greater than or equal to the preset number, screening the first reference pixel set to obtain a second reference pixel set, and if the number of valid pixel points in the second reference pixel set is less than or equal to the preset number, using the preset component value as a prediction value corresponding to the predicted image component; if the number of valid pixel points in the second reference pixel set is greater than or equal to the preset number, screening the first reference pixel set to obtain a second reference pixel set, and if the number of valid pixel points in the second reference pixel set is less than or equal to the preset number, using the preset component value as a prediction value corresponding to the predicted image component; If the number of effective pixel points in the first reference pixel set is equal to the number of presets, determine model parameters through the first reference pixel set, and obtain a prediction model corresponding to the predicted image component according to the model parameters, where the prediction model is used to realize a prediction process for the predicted image component to obtain a predicted value corresponding to the predicted image component; in this way, when the number of effective pixel points in the first reference pixel set is smaller than the number of presets or when the number of effective pixel points in the second reference pixel set is smaller than the number of presets, disable the CCLM mode, and directly use the default value of the preset as the predicted value corresponding to the predicted image component, without increasing additional processing modules, and at the same time, reduce the calculation complexity; and further, perform the derivation of model parameters only when the number of effective pixel points in the second reference pixel set is the number of presets, i.e., perform the CCLM mode to unify the derivation process of the model parameters.

[0024] Hereinafter, each embodiment of the present application will be described in detail with reference to the drawings.

[0025] Referring to FIG. 3, an exemplary block diagram of a configuration of a video encoding system according to an embodiment of the present application is shown. As shown in FIG. 3, the video encoding system 300 includes a transform and quantization unit 301, an intra estimation unit 302, an intra prediction unit 303, a motion compensation unit 304, a motion estimation unit 305, an inverse transform and inverse quantization unit 306, a filter control analysis unit 307, a filtering unit 308, an encoding unit 309, and a decoded image cache unit 310, etc., where the filtering unit 308 can realize deblocking filtering and Sample Adaptive Offset (SAO) filtering, and the encoding unit 309 can realize header coding and Context-based Adaptive Binary Arithmetic Coding (CABAC). For an input original video signal, a video coding block can be obtained through division of a coding tree unit (CTU), and then, for residual pixel information obtained by intra or inter prediction, the video coding block is transformed through a transform and quantization unit 301, the residual information is transformed from a pixel field to a transform field, and the obtained transform coefficients are quantized to reduce the bit rate.The intra estimation unit 302 and the intra prediction unit 303 are configured to perform intra prediction on the video coding block, specifically, the intra estimation unit 302 and the intra prediction unit 303 are configured to determine an intra prediction mode used to code the video coding block, the motion compensation unit 304 and the motion estimation unit 305 are configured to perform inter predictive coding of the received video coding block corresponding to one or more blocks in one or more reference frames to provide temporal prediction information, the motion estimation performed by the motion estimation unit 305 is a process of generating a motion vector, which can estimate the motion of the video coding block, and then the motion compensation unit 304 performs motion compensation based on the motion vector determined by the motion estimation unit 305. After determining the intra prediction mode, the intra prediction unit 303 is further configured to provide the selected intra prediction data to the encoding unit 309, and the motion estimation unit 305 also sends the calculated and determined motion vector data to the encoding unit 309. In addition, the inverse transform and inverse quantization unit 306 is configured to reconstruct the video coding block and reconstruct a residual block in a pixel field, and the reconstructed residual block is filtered through the filter control analysis unit 307 and the filtering unit 308 to remove block effect artifacts, and then adds the reconstructed residual block to one prediction block in a frame of the decoding image cache unit 310 to generate a reconstructed video coding block.The encoding unit 309 is configured to code various coding parameters and quantized transform coefficients, and in a coding algorithm based on CABAC, the content of the context can be based on neighboring coding blocks, and is configured to code the information of an intra-prediction mode determined by an instruction, and output a bitstream of the video signal, and the decoding image cache unit 310 is configured to store reconstructed video coding blocks for predicting references. With the execution of video image coding, new reconstructed video coding blocks are continuously generated, and all of these reconstructed video coding blocks are stored in the decoding image cache unit 310.

[0026] 4, an exemplary block diagram of a configuration of a video decoding system according to an embodiment of the present application is shown. As shown in FIG. 4, the video decoding system 400 includes a decoding unit 401, an inverse transform and inverse quantization unit 402, an intra prediction unit 403, a motion compensation unit 404, a filtering unit 405, and a decoded image cache unit 406, etc., where the decoding unit 401 can realize header decoding and CABAC decoding, and the filtering unit 405 can realize deblocking filtering and SAO filtering. After the input video signal is subjected to the coding process of FIG. 2, the system outputs a bitstream of the video signal. The bitstream is input to the video decoding system 400, and first passes through the decoding unit 401 to obtain decoded transform coefficients, and the transform coefficients are processed through the inverse transform and inverse quantization unit 402 to generate residual blocks in a pixel field. The intra prediction unit 403 is configured to generate prediction data of the current video decoding block according to the determined intra prediction mode and the data passing through the decoding block from the current frame or picture, and the motion compensation unit 404 is configured to determine prediction information for the video decoding block by analyzing the motion vector and other related grammar elements, and use the prediction information to generate a prediction block of the video decoding block being decoded. The residual block from the inverse transformation and inverse quantization unit 402 is added with the corresponding prediction block generated by the intra prediction unit 403 or the motion compensation unit 404 to form a video block of decoding, and the video signal of the decoding can pass through a filtering unit 405 to remove block effect artifacts and improve video quality, and then store the decoded video block in a decoding image cache unit 406, which is configured to store a reference image for subsequent intra prediction or motion compensation and output a video signal at the same time, i.e. obtain a restored original video signal.

[0027] The prediction method of the image component in the embodiment of the present application is mainly applied to the part of the intra prediction unit 303 shown in FIG. 3 and the part of the intra prediction unit 403 shown in FIG. 4, specifically, to the CCLM prediction unit in intra prediction. That is, the prediction method of the image component in the embodiment of the present application may be applied to a video encoding system, may be applied to a video decoding system, and may be simultaneously applied to a video encoding system and a video decoding system, but is not particularly limited in the embodiment of the present application. When the method is applied to the part of the intra prediction unit 303, the "coding block in the video image" specifically refers to the current coding block in intra prediction, and when the method is applied to the part of the intra prediction unit 403, the "coding block in the video image" specifically refers to the current decoding block in intra prediction.

[0028] Based on the application scenario example of Fig. 3 or Fig. 4 above, referring to Fig. 5, an exemplary flowchart of an image component prediction method according to an embodiment of the present application is shown. As shown in Fig. 5, the method may include the following steps:

[0029] In S501, a first reference pixel set corresponding to a predicted image component of a coding block in a video image is obtained.

[0030] It should be noted that a video image can be divided into multiple coding blocks, each coding block can include a first image component, a second image component and a third image component, and the coding block in the present embodiment is a current block in a video image to be coded. When the first image component needs to be predicted through a prediction model, the predicted image component is the first image component, when the second image component needs to be predicted through a prediction model, the predicted image component is the second image component, and when the third image component needs to be predicted through a prediction model, the predicted image component is the third image component.

[0031] Furthermore, it should be noted that, when the left neighboring region, the lower-left neighboring region, the upper neighboring region and the upper-right neighboring region are all valid regions, for the INTRA_LT_CCLM mode, the first reference pixel set is constituted by adjacent reference pixel points in the left neighboring region and the upper neighboring region of the coding block as shown in (a) of FIG. 2, for the INTRA_L_CCLM mode, the first reference pixel set is constituted by adjacent reference pixel points in the left neighboring region and the lower-left neighboring region of the coding block as shown in (b) of FIG. 2, and for the INTRA_T_CCLM mode, the first reference pixel set is formed by adjacent reference pixel points in the upper neighboring region and the upper-right neighboring region of the coding block as shown in (c) of FIG. 2.

[0032] In some embodiments, illustratively, for S501, obtaining a first reference pixel set corresponding to a predicted image component of a coding block in the video image may include the following steps.

[0033] In S501a-1, a reference pixel point adjacent to at least one side of the coding block is obtained, where the at least one side includes a left side of the coding block and / or a top side of the coding block.

[0034] At S501a-2, a first reference pixel set corresponding to the predicted image component is formed based on the reference pixel points.

[0035] At least one edge of the coding block may include the left edge of the coding block and / or the top edge of the coding block, i.e., at least one edge of the coding block may point to the top edge of the coding block, may point to the left edge of the coding block, or may further point to both the top edge and the left edge of the coding block, and it should be noted that this is not particularly limited in the present embodiment.

[0036] In this way, for the INTRA_LT_CCLM mode, if the left adjacent region and the upper adjacent region are all valid regions, the first reference pixel set can be formed by the reference pixel points adjacent to the left edge of the coding block and the reference pixel points adjacent to the upper edge of the coding block; if the left adjacent region is a valid region and the upper adjacent region is an invalid region, the first reference pixel set can be formed by the reference pixel points adjacent to the left edge of the coding block; if the left adjacent region is an invalid region and the upper adjacent region is a valid region, the first reference pixel set can be formed by the reference pixel points adjacent to the upper edge of the coding block.

[0037] In some embodiments, illustratively, for S501, obtaining a first reference pixel set corresponding to a predicted image component of a coding block in the video image may include the following steps.

[0038] In S501b-1, a reference pixel point in a reference row or reference column adjacent to the coding block is obtained, where the reference row is formed by rows adjacent to the upper side and upper right side of the coding block, and the reference column is formed by columns adjacent to the left side and lower left side of the coding block.

[0039] At S501b-2, a first reference pixel set corresponding to the predicted image component is formed based on the reference pixel points.

[0040] The reference row adjacent to the coding block may be formed by the row adjacent to the upper side and the upper right side of the coding block, and the reference column adjacent to the coding block may be formed by the column adjacent to the left side and the lower left side of the coding block, and the reference row or the reference column adjacent to the coding block may refer to the reference row adjacent to the upper side of the coding block, may refer to the reference column adjacent to the left side of the coding block, or may further refer to the reference row or the reference column adjacent to the other edge of the coding block, and it should be noted that there is no particular limitation in the present embodiment. For convenience of explanation, in the present embodiment, the reference row adjacent to the coding block is described by taking the reference row adjacent to the upper side as an example, and the reference column adjacent to the coding block is described by taking the reference column adjacent to the left side as an example.

[0041] Here, the reference pixel points in the reference row adjacent to the coding block may include reference pixel points adjacent to the upper edge and the upper right edge (which may also be referred to as adjacent reference pixel points corresponding to the upper edge and the upper right edge), where the upper edge represents the upper edge of the coding block, and the upper right edge represents the length of the edge that extends horizontally to the right from the upper edge of the coding block and has the same height as the current coding block, and the reference pixel points in the reference column adjacent to the coding block may further include reference pixel points adjacent to the left edge and the lower left edge (which may also be referred to as adjacent reference pixel points corresponding to the left edge and the lower left edge), where the left edge represents the left edge of the coding block, and the lower left edge represents the length of the edge that extends vertically downward from the left edge of the coding block and has the same width as the current decoded block, although this is not particularly limited in the present embodiment.

[0042] In this way, for INTRA_L_CCLM mode, if the left adjacent region and the lower left adjacent region are valid regions, the first reference pixel set can be formed by reference pixel points in a reference column adjacent to the coding block, and for INTRA_T_CCLM mode, if the upper adjacent region and the upper right adjacent region are valid regions, the first reference pixel set can be formed by reference pixel points in a reference row adjacent to the coding block.

[0043] In S502, if the number of valid pixel points in the first reference pixel set is smaller than the number of presets, the component value of the preset is used as a predicted value corresponding to the image component to be predicted.

[0044] It should be noted that the number of valid pixel points can be determined according to the validity of the adjacent regions, and can be determined according to the number of valid pixel points in the selected region. For some special cases, such as the boundary of the coding block, when it is not predictable and when the coding order is not able to obtain the adjacent reference pixel points, and further, when performing the division of the coding block according to tile and slice, the left adjacent region, the lower left adjacent region, the upper adjacent region and the upper right adjacent region are not all valid regions, and there may be invalid regions, so that the number of valid pixel points in the selected region is smaller than the preset number, and the number of valid pixel points in the first reference pixel set is smaller than the preset number.

[0045] In addition, the number of presets is a predetermined number of effective pixel points, which is used to measure whether the predicted image components are to perform the steps of deriving model parameters and constructing a prediction model, where it is noted that the number of presets can be four, but is not particularly limited in the present embodiment. In this way, assuming that the number of presets is four, that is, when the number of effective pixel points in the first reference pixel set is 0 or 2, the component value of the preset is directly used as the predicted value corresponding to the predicted image components, thereby reducing the calculation complexity.

[0046] Furthermore, the preset component value is used to represent a fixed value (which may also be referred to as a default value) corresponding to a previously set predicted image component. Here, the preset component value mainly relates to bit information of a current video image. Thus, in some embodiments, for S502, when the number of valid pixel points in the first reference pixel set is smaller than the number of presets, using the preset component value as a predicted value corresponding to the predicted image component may include the following steps:

[0047] At S502a, a preset component range corresponding to the predicted image component is determined based on bit information of a video image.

[0048] In S502b, a median value of the preset component range is determined according to the preset component range, and the median value is used as a predicted value corresponding to the predicted image component, where the median value is indicated in a preset component value.

[0049] It should be noted that in the present embodiment, the intermediate value of the preset component range corresponding to the predicted image component can be used as the preset component value, and then used as the predicted value corresponding to the predicted image component. Here, assuming that the bit depth of the predicted image component is represented by BitDepthC, it can be obtained that the calculation method of the intermediate value of the predicted image component is 1<<(BitDepthC-1), and the calculation method can be set according to actual cases, and is not particularly limited in the present embodiment.

[0050] For example, take the predicted image component as chrominance component, assuming that the current video image is an 8-bit video, the component range corresponding to the chrominance component is 0-255, where the intermediate value is 128, in this case the preset component value may be 128, i.e. the default value may be 128; assume that the current video image is a 10-bit video, the component range corresponding to the chrominance component is 0-1023, where the intermediate value is 512, in this case the preset component value may be 512, i.e. the default value is 512. In the embodiment of the present application, the bit information of the video image is taken as 10 bits, for example, that is the preset component value is 512.

[0051] Furthermore, in some embodiments, after using the preset component values ​​as predicted values ​​corresponding to the predicted image components in S502, the method may further include the following steps:

[0052] At S502c, for each pixel point in the coding block, a predicted value fill is performed for the predicted image component of each pixel point using the preset component values.

[0053] Please note that for cases where the number of valid pixel points in the first reference pixel set is smaller than the preset number, a fixed default value is directly used to perform filling of predicted values ​​for predicted image components in the coding block without the need to add additional processing modules.

[0054] For example, assuming that the preset component value is 512 and the predicted image component is a chrominance component, for the chrominance prediction value corresponding to each pixel point in the coding block, 512 can be directly used to fill in the chrominance prediction value.

[0055] In S503, if the number of valid pixel points in the first reference pixel set is greater than or equal to a preset number, the first reference pixel set is screened to obtain a second reference pixel set.

[0056] It should be noted that in the first parameter pixel set, there may be some unimportant reference pixel points (for example, these reference pixel points have poor relevance) or some abnormal reference pixel points, and in order to ensure the accuracy of the prediction model, these reference pixel points need to be deleted to obtain the second reference pixel set, where the number of effective pixel points in the second reference pixel set is less than or equal to the number of presets. Here, the number of effective pixel points included in the second reference pixel set is, in practical application, usually selected as the number of presets to be four, but is not particularly limited in the present embodiment.

[0057] In addition, if the number of effective pixel points in the first reference pixel set is greater than or equal to the number of presets, the first reference pixel set can also be screened to obtain a second reference pixel set, and after obtaining the second reference pixel set, it is still necessary to make a judgment according to the number of effective pixel points in the second reference pixel set and the number of presets. Here, if the number of effective pixel points in the second reference pixel set is less than the number of presets, the component value of the preset can be adopted as the prediction value corresponding to the image component to be predicted, and if the number of effective pixel points in the second reference pixel set is equal to the number of presets, the model parameters can be derived according to the second reference pixel set.

[0058] Further, in some embodiments, for S503, screening the first set of reference pixels to obtain a second set of reference pixels further comprises: determining a selected pixel point location based on pixel locations and / or image component intensities corresponding to each neighboring reference pixel point in the first reference pixel set; According to the determined selected pixel point position, selecting an effective pixel point corresponding to the selected pixel point position from the first reference pixel set, and forming a second reference pixel set with the effective pixel points obtained by selection, wherein the number of effective pixel points in the second reference pixel set is less than or equal to a preset number.

[0059] Specifically, the first reference pixel set may be screened according to the position of the selected reference pixel point, or may be screened according to the image component intensity (luminance value, chromaticity value, etc.), so that the second reference pixel set is formed by the screened selected reference pixel point. The following will be described taking the selected reference pixel point position as an example.

[0060] Assuming that the number of available pixel point samples in the upper and upper right regions adjacent to the current coding block is numSampT, and the number of available pixel point samples in the left and lower left regions adjacent to the current coding block is numSampL, the screening process is as follows (where availT represents the availability of the row adjacent to the top of the current coding block, availL represents the availability of the column adjacent to the left of the current coding block, nTbW represents the width of the current coding block, and nTbH represents the height of the current coding block):

[0061] If the intra prediction mode of the current block is INTRA_LT_CCLM mode, numSampT=availT?nTbW:0 numSampL=availL?nTbH:0, If not, numSampT=(availT&&predModeIntra==INTRA_T_CCLM)?(nTbW+Min(numTopRight,nTbH)):0 numSampL=(availL&&predModeIntra==INTRA_L_CCLM)?(nTbH+Min(numLeftBelow,nTbW)):0.

[0062] Here, numTopRight represents the number of valid pixel points in the upper right nTbW range, and numLeftBelow represents the number of valid pixel points in the lower left nTbH range. The number of pixels screened on each edge is indicated by cntN, the start point position is indicated by startPosN, the selection interval is indicated by pickStepN, and the selected pixel point position is indicated by pickPosN[pos], and the derivation process is as follows:

[0063] The variable numIs4N indicates whether to screen the pixel points on one side or not: numIs4N=((availT&&availL&&predModeIntra==INTRA_LT_CCLM)?0:1) The variable startPosN represents the starting position: startPosN=numSampN>>(2+numIs4N) The variable pickStepN represents the selection interval: pickStepN=Max(1,numSampN>>(1+numIs4N)) Here, N is replaced by T and L, respectively, which represent the case of screening pixel points on the top side and the case of screening on the left side, that is, N side here represents T side or L side. Here, when the availability of N side availN is TRUE and the selected intra mode predModeIntra is INTRA_LT_CCLM mode or INTRA_N_CCLM mode, the number of pixel points screened on N side cntN and the selected pixel point position pickPosN[pos] are as follows (it should be noted that the total number of screened pixel points should be cntT+cntL):

[0064] cntN=Min(numSampN,(1+numIs4N)<<1) pickPosN[pos]=(startPosN+pos*pickStepN),with pos=0...cntN-1 Otherwise, cntN is set to 0, that is, the number of screened pixel points is 0.

[0065] Assuming the predicted samples of the current coding block are predSamples[x][y] with x=0...nTbW-1, y=0...nTbH-1, the derivation is as follows:

[0066] If numSampL and numSampT are both invalid, they default to preset values, as follows:

[0067] predSamples[x][y]=1<<(BitDepth C -1) Otherwise, In the first step, obtain the luminance reconstruction samples pY[x][y] with x=0...nTbW*2-1, y=0...nTbH*2-1 of the same luminance block; In the second step, we obtain the adjacent luminance reconstruction samples pY[x][y], In the third step, obtain the downsampled luminance reconstruction samples pDsY[x][y] with x=0...nTbW-1, y=0...nTbH-1; In the fourth step, if numSampL is greater than 0, we set the chromaticity value pSelC[idx] of the point selected by the left side to p[-1][pickPosL[idx]] with idx=0...cntL-1 and get the downsampled reconstructed luminance value pSelDsY[idx] with idx=0...cntL-1 of the point selected by the left side.

[0068] In the fifth step, if numSampT is greater than 0, set the chromaticity value pSelC[idx] of the point selected by the top to p[pickPosT[idx-cntL]][-1] with idx=cntL...cntL+cntT-1, and get the downsampled reconstructed luminance value pSelDsY[idx] with idx=0...cntL+cntT-1 of the point selected by the top.

[0069] In step 6, if cntT+cntL is different from 0, the derivation of variables minY, maxY, minC and maxC is as follows:

[0070] If cntT+cntL is equal to 2, then let pSelComp[3] be pSelComp[0], pSelComp[2] be pSelComp[1], pSelComp[0] be pSelComp[1], and pSelComp[1] be pSelComp[3], where Comp is replaced by DsY and C, respectively, to represent the reconstructed luma and chroma of the selected adjacent sample.

[0071] The derivation of the arrays minGrpIdx and maxGrpIdx is as follows:

[0072] minGrpIdx[0]=0 minGrpIdx[1]=2 maxGrpIdx[0]=1 maxGrpIdx[1]=3 If pSelDsY[minGrpIdx[0]] is greater than pSelDsY[minGrpIdx[1]], minGrpIdx[0] is swapped with minGrpIdx[1], (minGrpIdx[0],minGrpIdx[1])=Swap(minGrpIdx[0],minGrpIdx[1]) If pSelDsY[maxGrpIdx[0]] is greater than pSelDsY[maxGrpIdx[1]], then maxGrpIdx[0] is swapped with maxGrpIdx[1], (maxGrpIdx[0],maxGrpIdx[1])=Swap(maxGrpIdx[0],maxGrpIdx[1]) If pSelDsY[minGrpIdx[0]] is greater than pSelDsY[maxGrpIdx[1]], the array minGrpIdx is replaced with maxGrpIdx, (minGrpIdx,maxGrpIdx)=Swap(minGrpIdx,maxGrpIdx) If pSelDsY[minGrpIdx[1]] is greater than pSelDsY[maxGrpIdx[0]], minGrpIdx[1] is swapped with maxGrpIdx[0], (minGrpIdx[1],maxGrpIdx[0])=Swap(minGrpIdx[1],maxGrpIdx[0]) The calculations of the variables maxY, maxC, minY and minC are as follows (each representing the average of the two groups):

[0073] maxY=(pSelDsY[maxGrpIdx[0]]+pSelDsY[maxGrpIdx[1]]+1)>>1 maxC=(pSelC[maxGrpIdx[0]]+pSelC[maxGrpIdx[1]]+1)>>1 minY=(pSelDsY[minGrpIdx[0]]+pSelDsY[minGrpIdx[1]]+1)>>1 minC=(pSelC[minGrpIdx[0]]+pSelC[minGrpIdx[1]]+1)>>1 In the seventh step, the derivation process of the linear model parameters a, b and k is as follows (where a is the slope (the ratio of the chromaticity difference to the luminance difference), b is the intercept, and k is the shift of a, storing a as an integer):

[0074] If numSampL is equal to 0 and numSampT is equal to 0, k=0 a=0 b=1<<(BitDepth C -1) If not, diff=maxY-minY If diff is different from 0, diffC=maxC-minC x=Floor(Log2(diff)) normDiff=((diff<<4)>>x)&15 x+=(normDiff !=0)?1:0 y=Floor(Log2(Abs(diffC)))+1 a=(diffC*(divSigTable[normDiff]|8)+2y-1)>>y k=((3+xy)<1)?1:3+xy a=((3+xy)<1)?Sign(a)*15:a b=minC-((a*minY)>>k) Here, divSigTable[ ] is divSigTable[ ]={0,7,6,5,5,4,4,3,3,2,2,1,1,1,1,0}, Otherwise (if diff is equal to 0), k=0 a=0 b=minC In the eighth step, the chrominance prediction samples predSamples[x][y] with x=0...nTbW-1, y=0...nTbH-1 are obtained via the following calculation (where Clip1C limits the prediction values ​​between 0-1023): predSamples[x][y]=Clip1C(((pDsY[x][y]*a)>>k)+b)

[0075] JPEG2025014004000006.jpg49170

[0076] JPEG2025014004000007.jpg99170

[0077] JPEG2025014004000008.jpg85170

[0078] JPEG2025014004000009.jpg86170

[0079] In this way, for cases where the number of valid pixel points in the first reference pixel set is greater than or equal to a preset number, a second reference pixel set can be obtained by screening the first reference pixel set, and the second reference pixel set includes four valid pixel points.

[0080] In S504, if the number of valid pixel points in the second reference pixel set is smaller than the number of presets, the component value of the preset is used as a predicted value corresponding to the image component to be predicted.

[0081] In S505, if the number of effective pixel points in the second reference pixel set is equal to the number of presets, a model parameter is determined through the second reference pixel set, and a prediction model corresponding to the predicted image component is obtained according to the model parameter.

[0082] It should be noted that after screening the first reference pixel set, a second reference pixel set is obtained. The number of effective pixel points in the second reference pixel set may be smaller than the number of presets, or may be larger than or equal to the number of presets. If the number of effective pixel points in the second reference pixel set is smaller than the number of presets, the component value of the preset is directly used as the predicted value corresponding to the predicted image component; if the number of effective pixel points in the second reference pixel set is larger than or equal to the number of presets, the model parameters are determined through the second reference pixel set, and a prediction model corresponding to the predicted image component is obtained according to the model parameters. It should be noted that the reference pixel points used to derive the model parameters are usually four, so the second reference pixel set obtained by screening is smaller than the number of presets (the second reference pixel set is smaller than four effective pixel points) or equal to the number of presets (the second reference pixel set includes four effective pixel points).

[0083] Furthermore, it should be noted that the prediction model may be a linear model or a nonlinear model. Here, the nonlinear model may be a nonlinear form such as a quadratic curve, or a nonlinear form formed by multiple linear models, for example, the cross-component prediction technique of multi-model CCLM (MMLM: Multiple Model CCLM) is a nonlinear form formed by multiple linear models, and is not particularly limited in the present embodiment. Here, the prediction model can be used to perform a prediction process on the image component to be predicted to obtain a prediction value corresponding to the image component to be predicted.

[0084] After obtaining the second reference pixel set, when the number of effective pixel points in the second reference pixel set is equal to the number of presets, the model parameters can be determined according to the second reference pixel set. After deriving the model parameters, a prediction model corresponding to the chrominance components can be further obtained according to the model parameters as shown in Equation (1), and then the prediction model is used to perform a prediction process on the chrominance components to obtain a predicted value corresponding to the chrominance components.

[0085] Further, in some embodiments, for S505, after obtaining a prediction model corresponding to the predicted image component according to the model parameters, the method further comprises: The method may further include performing a prediction process on predicted image components of each pixel point in the encoding block based on the prediction model to obtain predicted values ​​corresponding to the predicted image components of each pixel point.

[0086] Please note that for the case where the number of effective pixel points in the first reference pixel set is greater than or equal to the preset number, in order to obtain a predicted value corresponding to the predicted image component of each pixel point in the coding block, it is necessary to determine model parameters (such as α and β) through the first reference pixel set, and then obtain a prediction model corresponding to the predicted image component according to the model parameters. For example, if the predicted image component is a chrominance component, it is possible to obtain a prediction model corresponding to the chrominance component shown in formula (1) according to the model parameters α and β, and then use the prediction model shown in formula (1) to perform a prediction process on the chrominance component of each pixel point in the coding block, thus obtaining a predicted value corresponding to the chrominance component of each pixel point.

[0087] In the present embodiment, for the left adjacent region, the lower left adjacent region, the upper adjacent region and the upper right adjacent region, there may be a valid region and there may be an invalid region, so that the number of valid pixel points selected from the adjacent regions may be smaller than the preset number. Therefore, in some embodiments, referring to FIG. 7, an exemplary flowchart of a prediction method of another image component according to the present embodiment is shown. As shown in FIG. 7, after S501, the method may further include the following steps:

[0088] In S701, the number of valid pixel points in the first reference pixel set is determined, and it is determined whether the number of valid pixel points is smaller than a preset number.

[0089] Furthermore, after S503, the method may further include the following steps:

[0090] In S702, it is determined whether the number of valid pixel points in the second reference pixel set is less than a preset number.

[0091] It should be noted that the determination of the number of effective pixel points can be judged according to the validity of the neighboring region. In this way, after determining the number of effective pixel points, by comparing the number of effective pixel points with the preset number, if the number of effective pixel points in the first reference pixel set is smaller than the preset number, execute step S502, if the number of effective pixel points in the second reference pixel set is smaller than the preset number, execute step S504, if the number of effective pixel points in the second reference pixel set is greater than or equal to the preset number, execute step S505.

[0092] Furthermore, it should be noted that the number of presets can be 4. The following will take the number of presets as 4 as an example for detailed description.

[0093] In one possible embodiment, when the number of effective pixel points in the first reference pixel set is greater than or equal to the preset number, since the reference pixel points used to derive the model parameters are usually four, the first reference pixel set is first screened to make the number of effective pixel points in the first reference pixel set four, and then model parameters are derived according to the four effective pixel points, and a prediction model corresponding to the predicted image component is obtained according to the model parameters, so as to obtain a predicted value corresponding to the predicted image component.

[0094] Specifically, assume that the image components to be predicted are chrominance components, and the chrominance components are predicted through the luminance components. Assume that the numbers of the four effective pixel points selected by screening are 0, 1, 2, and 3, respectively. By comparing the four selected effective pixel points, two pixel points with relatively large luminance values ​​(which may include the pixel point with the largest luminance value and the pixel point with the second largest luminance value) and two pixel points with relatively small luminance values ​​(which may include the pixel point with the smallest luminance value and the pixel point with the second smallest luminance value) can be further selected based on the four comparisons. Furthermore, two arrays, minIdx[2] and maxIdx[2], are set to respectively store two groups of pixel points, and initially, the effective pixel points with numbers 0 and 2 are first put into minIdx[2], and the effective pixel points with numbers 1 and 3 are put into maxIdx[2], as follows:

[0095] Init:minIdx[2]={0,2},maxIdx[2]={1,3} Then, through four comparisons, the two pixel points with relatively small luminance values ​​are stored in minIdx[2], and the two pixel points with relatively large luminance values ​​are stored in maxIdx[2], as shown below.

[0096] Step 1: if(L[minIdx[0]]>L[minIdx[1]],swap(minIdx[0],minIdx[1]) Step 2: if(L[maxIdx[0]]>L[maxIdx[1]],swap(maxIdx[0],maxIdx[1]) Step 3: if(L[minIdx[0]]>L[maxIdx[1]],swap(minIdx,maxIdx) Step 4: if(L[minIdx[1]]>L[maxIdx[0]],swap(minIdx[1],maxIdx[0]) In this way, two pixel points with relatively small brightness values ​​can be obtained, and the corresponding brightness values ​​can be expressed as luma 0 min and luma 1 minThe corresponding chromaticity values ​​are shown in 0 min and chroma 1 min At the same time, two pixels with relatively large luminance values ​​can be obtained, and the corresponding luminance values ​​are respectively represented as luma 0 max and luma 1 max The corresponding chromaticity values ​​are shown in 0 max and chroma 1 max In addition, the luminance values ​​corresponding to two relatively small pixel points can be averaged to obtain the luminance value corresponding to the first average point, and the luminance value corresponding to the first average point can be calculated as follows: min Then, the luminance values ​​corresponding to the two larger pixel points are averaged to obtain the luminance value corresponding to the second average point. max Similarly, we can obtain the chromaticity values ​​corresponding to the two average points, which are min and chroma max Specifically, the details are as follows:

[0097] luma min =(luma 0 min +luma 1 min +1)>>1 luma max =(luma 0 max +luma 1 max +1)>>1 chroma min =(chroma 0 min +chroma 1 min +1)>>1 chroma max =(chroma 0 max +chroma 1 max +1)>>1 In other words, the average point of the two (luma min ,chroma min) and (luma max ,chroma max After obtaining α, the model parameters can be obtained by the two points in a calculation manner of "two points determine a line". Specifically, the model parameters α and β can be calculated and obtained by Equation (2).

number

[0098] Here, the model parameter α is the slope in the prediction model, and the model parameter β is the intercept in the prediction model. After deriving the model parameters in this way, a prediction model corresponding to the chrominance components can be obtained according to the model parameters as shown in formula (1), and then the prediction model is used to perform a prediction process on the chrominance components to obtain a predicted value corresponding to the chrominance components.

[0099] In another possible embodiment, for some special cases, such as the case of the boundary of the coding block, when it is not predictable and when adjacent reference pixel points cannot be obtained by the coding order, and further, when the division of the coding block is performed according to tiles and slices, the left adjacent region, the lower left adjacent region, the upper adjacent region and the upper right adjacent region are not all valid regions, and there may be invalid regions, thereby making the number of valid pixel points in the first reference pixel set smaller than the preset number.

[0100] In this way, since the number of presets is four, in some embodiments, for S502, when the number of valid pixel points in the first reference pixel set is smaller than the number of presets, using the component value of the preset as the predicted value corresponding to the predicted image component is: If the number of valid pixel points in the first reference pixel set is 0 or 2, a preset component value may be used as a predicted value corresponding to the predicted image component.

[0101] In some embodiments, for S504, if the number of valid pixel points in the current second reference pixel set is smaller than the number of presets, using a preset component value as a predicted value corresponding to the predicted image component, If the number of valid pixel points in the second reference pixel set is 0 or 2, the prediction may include using a preset component value as a predicted value corresponding to the image component to be predicted.

[0102] In other words, when the number of presets is four, even if the number of valid pixel points in the first reference pixel set is smaller than the number of presets, or even if the number of valid pixel points in the second reference pixel set is smaller than the number of presets, the number of valid pixel points is 0 or 2.

[0103] Specifically, if the sum of the adjacent reference pixel points in the selection region used for the coding block is 0, select an effective pixel point of 0. In the following three special cases, obtain an effective pixel point of 0.

[0104] JPEG2025014004000011.jpg33170

[0105] JPEG2025014004000012.jpg34170

[0106] JPEG2025014004000013.jpg34170

[0107] It should be further noted that the valid pixel point is determined to be 0 according to the validity of the adjacent region, that is, the number of valid pixel points in the first reference pixel set can be determined according to the validity of the adjacent region. If the number of valid pixel points is 0, the model parameter α can be set to 0, and the model parameter β can be set to a preset component value corresponding to the image component to be predicted.

[0108] JPEG2025014004000014.jpg80170

[0109] JPEG2025014004000015.jpg59170

[0110] JPEG2025014004000016.jpg46170

[0111] JPEG2025014004000017.jpg40170

[0112] JPEG2025014004000018.jpg41170

[0113] It should be further noted that the number of effective pixel points is two, and can be determined according to the validity of the adjacent region, can be determined according to the number of effective pixel points in the selection region, and can also be determined according to other judgment conditions, and is not particularly limited in the present embodiment. In this way, the number of effective pixel points in the first reference pixel set can be determined according to the validity of the adjacent region.

[0114] In the prior art technical solution, when there are two effective pixel points, it is necessary to obtain four pixel points by duplicating the two effective pixel points. For example, assume that the numbers of the four pixel points are 0, 1, 2, and 3, number 0 is the second effective pixel point selected, number 1 is the first effective pixel point selected, number 2 is the second effective pixel point selected, and number 3 is the first effective pixel point selected. According to the four pixel points whose numbers are 0, 1, 2, and 3, the model parameters α and β can be determined to establish the prediction model shown in formula (1), and through the prediction model, a prediction value corresponding to the image component to be predicted can be obtained.

[0115] JPEG2025014004000019.jpg75170

[0116] Therefore, in the prior art technical solution, when the number of effective pixel points is two, an additional "duplicate" operation is required to use the same processing module, and four pixel points are obtained, so that the model parameters can be derived according to the same operation process when the number of effective pixel points is four, and the additional "duplicate" operation can be increased, and the four obtained pixel points still need to perform four comparisons and four average calculations, thereby making the calculation complexity relatively high. In the embodiment of the present application, the process when the number of effective pixel points is two is matched with the process when the number of effective pixel points is 0, and in this case, the same processing module can be directly used without increasing additional operations, thereby reducing the calculation complexity.

[0117] Referring to Fig. 10, an exemplary flow chart of deriving model parameters according to an embodiment of the present application is shown. In Fig. 10, assuming that the image component to be predicted is a chrominance component, firstly obtain adjacent reference pixel points from the selected region to form a first adjacent reference pixel set, then determine the number of valid pixel points in the first adjacent reference pixel set, if the number of valid pixel points is greater than or equal to 4, perform a screening process on the first reference pixel set to obtain a second reference pixel set, then determine the number of valid pixel points in the second adjacent reference pixel set, if the number of valid pixel points in the first reference pixel set or the second reference pixel set is 0, set the model parameter α to 0, and set the model parameter β to a default value, in this case, the predicted value corresponding to the chrominance component is filled with the default value. If the number of valid pixel points in the first reference pixel set or the second reference pixel set is 2, the processing steps are the same as the processing steps when the number of valid pixel points is 0. When the number of effective pixel points in the second reference pixel set is four, first, two pixel points of the relatively large chrominance component and two pixel points of the relatively small chrominance component are obtained by four comparisons, and then the two average points are obtained, and model parameters α and β are derived based on the two average points, and the prediction process of the chrominance component is performed according to the constructed prediction model. Therefore, only the coding blocks in which the number of effective pixel points in the second reference pixel set is four can derive the model parameters in the CCLM mode, and for the coding blocks in which the number of effective pixel points is less than four, the default value filling method is directly used to reduce the calculation complexity when the number of reference pixel points in the selected region is four or less, and the encoding and decoding performance can be basically maintained unchanged.

[0118] In the present embodiment, the process of deriving model parameters is integrated, that is, for the number of effective pixel points in the first reference pixel set used for deriving model parameters, if the number of effective pixel points in the first reference pixel set is greater than or equal to the preset number, perform screening of effective pixel points on the first reference pixel set to obtain a second reference pixel set, and then determine the number of effective pixel points in the second adjacent reference pixel set; if the number of effective pixel points in the second reference pixel set meets the preset number, the current coding block needs to perform the steps of deriving model parameters in CCLM and building a prediction model; if the number of effective pixel points in the first reference pixel set or the second reference pixel set is less than the preset number, the current coding block can use default values ​​to fill in the predicted values ​​corresponding to the predicted image components of the coding block. Therefore, as shown in FIG. 11, the present embodiment can further provide a simplified process of deriving model parameters.

[0119] Compared with FIG. 10, the derivation process of the model parameters shown in FIG. 11 is more streamlined. In FIG. 11, assuming that the image component to be predicted is a chrominance component and the preset component value is 512, firstly, obtain adjacent reference pixel points from the selected region to form a first adjacent reference pixel set, then determine the number of valid pixel points in the first adjacent reference pixel set, if the number of valid pixel points in the first reference pixel set is greater than or equal to the preset number, perform a screening process on the first reference pixel set to obtain a second reference pixel set. Then, determine the number of valid pixel points in the second adjacent reference pixel set, if the number of valid pixel points in the first reference pixel set or the second reference pixel set is less than the preset number, set the model parameter α to 0, and set the model parameter β to 512, in this case the predicted value corresponding to the chrominance component is filled in 512. If the number of effective pixel points in the second reference pixel set meets the preset number, first obtain two pixel points of the relatively large chrominance component and two pixel points of the relatively small chrominance component by four comparisons, then obtain the two average points, derive model parameters α and β based on the two average points, and perform the prediction process of the chrominance component according to the constructed prediction model. It should be noted that only the coding blocks in which the number of effective pixel points in the second reference pixel set meets the preset number can perform the derivation of model parameters in CCLM mode, and for the coding blocks in which the number of effective pixel points is smaller than the preset number, the method of filling with default values ​​can be directly used to reduce the calculation complexity when the number of reference pixel points in the selected region is equal to or smaller than the preset number, and further, the encoding and decoding performance can be basically kept unchanged. Usually, the number of presets in the present embodiment can be four.

[0120] Furthermore, in the present embodiment, the number of valid pixel points in the first reference pixel set can be determined according to the number of valid pixel points in the selection region. Therefore, as shown in FIG. 12, the present embodiment can further provide a simplified process of deriving another model parameter.

[0121] In FIG. 12, assuming that the image component to be predicted is a chrominance component and the preset component value is 512, firstly determine a selection region to obtain a first reference pixel set, then determine the number of effective pixel points in the first reference pixel set, if the number of effective pixel points in the first reference pixel set is greater than or equal to the preset number, perform a screening process on the first reference pixel set to obtain a second reference pixel set. Then determine the number of effective pixel points in the second adjacent reference pixel set, if the number of effective pixel points in the first reference pixel set or the second reference pixel set is less than the preset number, the prediction value corresponding to the chrominance component is filled in 512, if the number of effective pixel points in the second reference pixel set meets the preset number, firstly obtain two pixel points of the relatively large chrominance component and two pixel points of the relatively small chrominance component by four comparisons, then obtain two average value points, derive model parameters α and β based on the two average value points, and perform a prediction process of the chrominance component according to the constructed prediction model. Therefore, only for the coding blocks in which the number of effective pixel points in the second reference pixel set meets the preset number, the derivation of model parameters in the CCLM mode can be performed, and for the coding blocks in which the number of effective pixel points is smaller than the preset number, the method of filling with default values ​​is directly used, so that the calculation complexity can be reduced when the number of reference pixel points in the selected region is equal to or smaller than the preset number, and furthermore, the encoding and decoding performance can be kept basically unchanged. Usually, the number of presets in the present embodiment can be four.

[0122] JPEG2025014004000020.jpg123170

[0123] In addition, VVC defines the case where the variables numSampL and numSampT are both 0 (in this case, the number of effective pixel points to be screened and used to derive the model parameters can be 0), and the predicted value corresponding to the chromaticity component is directly set to a default value; otherwise, the model parameters need to be derived, specifically as follows:

[0124] if(numSampL==0&&numSampT==0) Setting the predicted values ​​corresponding to the chrominance components to default values; else The model parameters are derived and the constructed prediction model is used to predict the chrominance components to obtain predicted values ​​corresponding to the chrominance components.

[0125] JPEG2025014004000021.jpg55170

[0126] if(numSampL+numSampT<number of presets) Setting the predicted values ​​corresponding to the chrominance components to default values; else The model parameters are derived and the constructed prediction model is used to predict the chrominance components to obtain predicted values ​​corresponding to the chrominance components.

[0127] The present embodiment includes obtaining a first reference pixel set corresponding to a predicted image component of a coding block in a video image, and if the number of valid pixel points in the first reference pixel set is smaller than a preset number, using the preset component value as a predicted value corresponding to the predicted image component; if the number of valid pixel points in the first reference pixel set is greater than or equal to the preset number, screening the first reference pixel set to obtain a second reference pixel set, the number of valid pixel points in the second reference pixel set is smaller than or equal to the preset number; if the number of valid pixel points in the second reference pixel set is smaller than the preset number, using the preset component value as a predicted value corresponding to the predicted image component; When the number of effective pixel points in the adjacent reference pixel set is equal to the number of presets, a model parameter is determined through a second reference pixel set, and a prediction model corresponding to a predicted image component is obtained according to the model parameter, and the prediction model is used to realize a prediction process for the predicted image component to obtain a predicted value corresponding to the predicted image component; in this way, the derivation process of the model parameter is integrated under the premise that the encoding and decoding prediction performance is not changed; at the same time, when the number of effective pixel points in the adjacent reference pixel set is smaller than the number of presets, when the effective pixel points are 0 or 2, no additional processing module is added, so that no additional processing is required and the calculation complexity is reduced.

[0128] In another embodiment of the present application, when the number of valid pixel points in the first reference pixel set is less than the preset number, the CCLM mode can be directly disabled, and the component value of the preset can be used as the predicted value corresponding to the predicted image component. Thus, in some embodiments, after screening the first reference pixel set to obtain a second reference pixel set, the method includes: if the number of valid pixel points in the first reference pixel set is smaller than a preset number or the number of valid pixel points in the second reference pixel set is smaller than a preset number, using a preset component value as a predicted value corresponding to the predicted image component; The method further includes adopting a CCLM mode to realize a prediction process for the image component to be predicted when the number of valid pixel points in the second reference pixel set is greater than or equal to the preset number.

[0129] It should be noted that when the number of valid pixel points in the first reference pixel set or the second reference pixel set is smaller than the number of presets, in this case, the CCLM mode can be disabled, for example, the identifier for whether to use the CCLM mode is set to ``disable CCLM mode'', in which case the prediction value corresponding to the predicted image component is directly filled with a default value, and only when the number of valid pixel points in the second reference pixel set is greater than or equal to the number of presets, the CCLM mode is used, for example, the identifier for whether to use the CCLM mode is set to ``enable CCLM mode'', in which case prediction processing for the predicted image component can be realized via the CCLM mode.

[0130] Furthermore, assuming that the predicted image components are chromaticity components and the number of presets is four, for all cases in which two valid pixel points may be generated (here, the judgment method for generating two valid pixel points is not particularly limited in the present embodiment), it is further noted that the model parameter α can be set to 0 and the model parameter β can be set to an intermediate value (which may also be referred to as a default value) corresponding to the chromaticity component, thereby filling the predicted values ​​corresponding to the chromaticity components of all pixel points in the coding block with the default value; furthermore, for all cases in which two valid pixel points may be generated, numSampL and numSampT can both be set to 0, thereby filling the predicted values ​​corresponding to the chromaticity components of all pixel points in the coding block with the default value.

[0131] Further, for all cases that may generate two effective pixel points, the predicted values ​​corresponding to the chrominance components may be directly filled with default values, or for all cases that may generate two effective pixel points, the CCLM mode may be disabled, or for all cases that may generate two or zero effective pixel points, the CCLM mode may be disabled, or for all cases that may generate two or zero effective pixel points, the predicted values ​​corresponding to the chrominance components may be directly filled with default values.

[0132] Therefore, when the number of effective pixel points used for deriving the model parameters is different, the unification of the derivation process of the model parameters is realized. Specifically, when the number of effective pixel points is two, no additional processing is required, and the existing processing module is directly read (i.e., the processing when the number of effective pixel points is two is matched with the processing when the number of effective pixel points is zero), thereby reducing the computational complexity.

[0133] The image component prediction method in this application example is based on the VVC latest reference software VTM5.0, and is under All intra conditions. The test sequence required for JVET is in accordance with general test conditions. The average BD-rate changes for the Y, Cb and Cr components are 0.00%, 0.02% and 0.02%, respectively. This indicates that there is essentially no impact on the encoding and decoding performance.

[0134] On the premise that it does not affect the encoding and decoding performance, this application can have the following beneficial effects:

[0135] First, the present application can unify the derivation process of model parameters in CCLM mode. In the prior art technical solution, for the case where there are two effective pixel points, an additional "duplicate" operation needs to be performed to generate four usable pixel points, so that the same operation can be performed for the case where there are four effective pixel points, and the model parameters can be derived. However, the present application can reduce the additional "duplicate" operation, and at the same time, the processing for the case where the number of effective pixel points is two can be matched with the processing for the case where the number of effective pixel points is zero, and in this case, the same processing module can be directly used without increasing additional operations, so that the derivation process of linear model parameters can be unified.

[0136] JPEG2025014004000022.jpg93170

[0137] This embodiment provides a method for predicting an image component, and through the technical solution of this embodiment, the number of effective pixel points in a first reference pixel set is compared with a preset number, and when the number of effective pixel points in the first reference pixel set or the second reference pixel set is smaller than the preset number, the default value of the preset is directly used as the predicted value corresponding to the predicted image component; when the number of effective pixel points in the second reference pixel set is greater than or equal to the preset number, the model parameters are determined according to the first reference pixel set to build a prediction model of the predicted image component, thereby integrating the derivation process of the model parameters; and further, for the case when the number of effective pixel points in the first reference pixel set is smaller than the preset number, no additional processing modules are increased, thereby reducing the calculation complexity.

[0138] Based on the same inventive concept as the above embodiment, referring to FIG. 13, an exemplary structural diagram of the configuration of an image component prediction device 130 according to the embodiment of the present application is shown. The image component prediction device 130 can include an acquisition unit 1301, a prediction unit 1302 and a screening unit 1303, in which: The obtaining unit 1301 is configured to obtain a first reference pixel set corresponding to a predicted image component of a coding block in a video image; The prediction unit 1302 is configured to use a preset component value as a prediction value corresponding to the predicted image component when a number of valid pixel points in the first reference pixel set is smaller than a preset number; The screening unit 1303 is configured to screen the first reference pixel set to obtain a second reference pixel set when a number of valid pixel points in the first reference pixel set is greater than or equal to a preset number, where the number of valid pixel points in the second reference pixel set is less than or equal to a preset number; The prediction unit 1302 is further configured to: use a preset component value as a predicted value corresponding to the predicted image component when the number of valid pixel points in the second reference pixel set is smaller than a preset number; and when the number of valid pixel points in the second reference pixel set is equal to the preset number, determine model parameters via the second reference pixel set and obtain a prediction model corresponding to the predicted image component according to the model parameters, where the prediction model is used to realize a prediction process for the predicted image component and obtain a predicted value corresponding to the predicted image component.

[0139] In the above technical solution, the acquisition unit 1301 is specifically configured to acquire reference pixel points adjacent to at least one side of the coding block, where the at least one side includes the left side of the coding block and / or the top side of the coding block, and form a first reference pixel set corresponding to the predicted image component based on the reference pixel points.

[0140] In the above technical solution, the obtaining unit 1301 is specifically configured to obtain reference pixel points in a reference row or a reference column adjacent to the coding block, where the reference row is formed by a row adjacent to an upper edge and an upper right edge of the coding block, and the reference column is formed by a column adjacent to a left edge and a lower left edge of the coding block, and form a first reference pixel set corresponding to the predicted image component based on the reference pixel points.

[0141] In the above technical solution, the screening unit 1303 is specifically configured to determine a selected pixel point position based on a pixel position and / or image component intensity corresponding to each adjacent reference pixel point in the first reference pixel set, select an effective pixel point corresponding to the selected pixel point position from the first reference pixel set according to the determined selected pixel point position, and form a second reference pixel set with the effective pixel points obtained by selection, where the number of effective pixel points in the second reference pixel set is less than or equal to a preset number.

[0142] In the above technical proposal, referring to FIG. 13, the image component prediction device 130 further includes a determination unit 1304 configured to determine a preset component range corresponding to the predicted image component based on bit information of the video image, determine an intermediate value of the preset component range according to the preset component range, and use the intermediate value as a predicted value corresponding to the predicted image component, where the intermediate value is indicated in a preset component value.

[0143] In the above technical proposal, referring to FIG. 13, the image component prediction device 130 further includes a filling unit 1305, configured to perform, for each pixel point in the encoding block, filling of a predicted value for the predicted image component of each pixel point using the preset component value.

[0144] In the above technical solution, the prediction unit 1302 is further configured to perform a prediction process on a predicted image component of each pixel point in the coding block based on the prediction model, to obtain a predicted value corresponding to the predicted image component of each pixel point.

[0145] In the above technical solution, the value of the number of presets is 4; and the prediction unit 1302 is further configured to use a preset component value as a prediction value corresponding to the image component to be predicted when the number of effective pixel points in the first reference pixel set is 0 or 2; Correspondingly, the prediction unit 1302 is further configured to use a preset component value as a prediction value corresponding to the predicted image component when the number of valid pixel points in the second reference pixel set is 0 or 2.

[0146] In the above technical solution, referring to FIG. 13, the image component prediction device 130 further includes a judgment unit 1306 configured to use a preset component value as a prediction value corresponding to the predicted image component when the number of effective pixel points in the first reference pixel set is smaller than a preset number or the number of effective pixel points in the second reference pixel set is smaller than a preset number, and to adopt a CCLM mode to realize a prediction process for the predicted image component when the number of effective pixel points in the second reference pixel set is greater than or equal to the preset number.

[0147] In this embodiment, it should be understood that a "unit" may be a part of a circuit, a part of a processor, a part of a program or software, and may be modular or non-modular. Furthermore, each component in this embodiment may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The integrated unit described above may be realized in the form of hardware or in the form of a software functional module.

[0148] If the integrated unit is implemented in the form of a software function module and is not sold or used as an independent product, it can be stored in one computer-readable storage medium. Based on this understanding, the technical solution of this embodiment, in essence or a part that contributes to the prior art, or all or a part of the solution of the technology can be embodied in the form of a software product, and the computer software product includes several instructions stored in one storage medium to cause a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor to execute all or a part of the steps of the method described in this embodiment. The aforementioned storage medium includes various media that can store program code, such as U disk, mobile hard disk, read only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0149] Therefore, this embodiment provides a computer storage medium, which stores an image component prediction program, which, when executed by at least one processor, realizes the steps of the method described in any one of the above embodiments.

[0150] Based on the above combination of the image component prediction device 130 and computer storage medium, referring to FIG. 14, a specific hardware structure of the image component prediction device 130 according to the embodiment of the present application is shown, which includes a network interface 1401, a memory 1402 and a processor 1403, and each component is coupled together via a bus system 1404. It can be understood that the bus system 1404 is used to realize the connection communication between these components. In addition to a data bus, the bus system 1404 includes a power bus, a control bus and a status signal bus. However, for the sake of clarity, various buses are represented as the bus system 1404 in FIG. 14. Here, the network interface 1401 is configured to send and receive signals in the process of sending and receiving information to and from another external network element; The memory 1402 is configured to store a computer program executable by the processor 1403. The processor 1403 is configured to execute the computer program to execute the following steps, which include: Obtaining a first set of reference pixels corresponding to a predicted image component of a coding block in a video image; if the number of valid pixel points in the first reference pixel set is less than a preset number, using a preset component value as a predicted value corresponding to the predicted image component; if the number of valid pixel points in the first reference pixel set is greater than or equal to the preset number, screening the first reference pixel set to obtain a second reference pixel set, where the number of valid pixel points in the second reference pixel set is less than or equal to the preset number; When the number of valid pixel points in the second reference pixel set is smaller than the number of presets, using the preset component value as a predicted value corresponding to the predicted image component; and when the number of valid pixel points in the second reference pixel set is equal to the number of presets, determining model parameters via the second reference pixel set and obtaining a prediction model corresponding to the predicted image component according to the model parameters, wherein the prediction model is used to realize a prediction process for the predicted image component and obtain a predicted value corresponding to the predicted image component.

[0151] It should be understood that the memory 1402 in the present embodiment may be volatile or non-volatile memory, or may include both volatile and non-volatile memory. Here, the non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory may be random access memory (RAM), used as an external cache. Various forms of RAM may be used, such as, by way of example and not limitation, static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus random access memory (DRRAM). The memory 1402 of the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0152] The processor 1403 may be an integrated circuit chip having the ability to process signals. In the realization process, each step of the above method can be completed by an instruction in the form of a hardware integrated logic circuit or software in the processor 1403. The above processor 1403 may be a general-purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Each method, step and logic block diagram disclosed in the embodiments of the present application can be realized or executed. The general-purpose processor may be a microprocessor, or the processor may be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present application may be directly executed by a hardware decoding processor, or may be executed by a combination of hardware and software modules in the decoding processor. The software modules may be located in conventional storage media such as random access memory, flash memory, read-only memory, programmable read-only memory, or electrically erasable programmable memory, registers, etc. The storage medium is placed in the memory 1402, and the processor 1403 reads the information in the memory 1402 and completes the steps of the method in combination with its hardware.

[0153] It can be understood that the embodiments described herein can be implemented in hardware, software, firmware, middleware, microcode, or a combination thereof. For a hardware implementation, the processing unit is implemented in one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processing (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field-Programmable Gate Arrays (FPGAs), general-purpose processors, controllers, microcontrollers, microprocessors, other electronic units configured to perform the functions described herein, or a combination thereof.

[0154] In a software implementation, the techniques described herein may be realized through modules (processes, functions, etc.) that perform the functions described herein. The software code may be stored in a memory and executed by a processor. The memory may be implemented within the processor or external to the processor.

[0155] Illustratively, as another embodiment, the processor 1403 is further configured to execute the computer program to perform the steps of the method according to any one of the above embodiments.

[0156] Referring to Fig. 15, an exemplary structural diagram of an encoder according to an embodiment of the present invention is shown. As shown in Fig. 15, the encoder 150 may include at least the image component prediction device 130 according to any one of the above embodiments.

[0157] Referring to Fig. 16, an exemplary structural diagram of a decoder according to an embodiment of the present application is shown. As shown in Fig. 16, the decoder 160 may include at least the image component prediction device 130 according to any one of the above embodiments.

[0158] It should be noted that, as used herein, the terms "comprising," "including," or any other variations thereof, are intended to have a non-exclusive inclusion, whereby a process, method, item, or apparatus that includes a set of elements not only includes those elements, but also includes other elements not expressly listed, or includes the inherent elements of those processes, methods, items, or apparatus. Without more limitation, an element qualified with the phrase "comprising a..." does not exclude the presence of other related elements in the process, method, item, or apparatus that includes that element.

[0159] The numbers of the above embodiments of the present invention do not indicate superiority or inferiority of the embodiments, but are used for convenience of explanation.

[0160] The methods disclosed in the several method embodiments of the present application may be combined in any manner without conflict to obtain new method embodiments.

[0161] Features disclosed in the various product embodiments according to the present application may be combined in any non-conflicting manner to obtain new product embodiments.

[0162] Features disclosed in any method or apparatus embodiment according to the present application may be combined in any non-conflicting manner to obtain new method or apparatus embodiments.

[0163] The above are only specific embodiments of the present application, and the scope of protection of the present application is not limited thereto. Any modifications or replacements that a person skilled in the art can easily think of within the technical scope disclosed in the present application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application shall be subject to the scope of protection of the claims. [Industrial Applicability]

[0164] In an embodiment of the present application, a first reference pixel set corresponding to a predicted image component of a coding block in a video image is obtained, and if a number of valid pixel points in the first reference pixel set is smaller than a preset number, a component value of the preset is used as a predicted value corresponding to the predicted image component; if a number of valid pixel points in the first reference pixel set is greater than or equal to the preset number, the first reference pixel set is screened to obtain a second reference pixel set, and if a number of valid pixel points in the second reference pixel set is smaller than or equal to the preset number, a component value of the preset is used as a predicted value corresponding to the predicted image component; if a number of valid pixel points in the second reference pixel set is equal to the preset number, a model parameter is determined via the second reference pixel set, and a prediction model corresponding to the predicted image component is obtained according to the model parameter, and the prediction model is applied to the predicted image component. A prediction process is realized for the first reference pixel set to obtain a predicted value corresponding to the predicted image component; in this way, when the number of effective pixel points in the first reference pixel set is smaller than the preset number or the number of effective pixel points in the second reference pixel set is smaller than the preset number, the preset default value is directly used as the predicted value corresponding to the predicted image component; only when the number of effective pixel points in the second reference pixel set meets the preset number, the model parameters are determined according to the first reference pixel set to establish a prediction model of the predicted image component, thereby integrating the derivation process of the model parameters; and further, when the number of effective pixel points in the first reference pixel set or the second reference pixel set is smaller than the preset number, especially when the effective pixel points are 0 or 2, no additional processing module is increased, so that the preset default value is directly used as the predicted value corresponding to the predicted image component, no additional processing is required, and the calculation complexity is reduced.

Claims

1. A method for predicting an image component, applied to a decoder, comprising the steps of: obtaining a first reference sample set corresponding to a predicted image component of a decoded block in a video image; if a number of valid sample points in the first reference sample set is equal to 0, using a preset component value as a predicted value corresponding to the predicted image component; If the number of valid sample points in the first reference sample set is greater than or equal to four, processing the first reference sample set to obtain a second reference sample set, where the number of valid sample points in the second reference sample set is less than or equal to a preset number, the preset number being four; determining model parameters based on the second reference sample set when a number of valid sample points in the second reference sample set is equal to a preset number; The method for predicting image components further comprises: obtaining a prediction model corresponding to the predicted image component according to the model parameters; performing a prediction process on the predicted image component based on the prediction model to obtain a predicted value corresponding to the predicted image component; Obtaining a first reference sample set corresponding to a predicted image component of a decoded block in the video image includes: Obtaining reference sample points adjacent to at least one side of the decoded block, the at least one side including a left side of the decoded block and / or a top side of the decoded block; determining a first reference sample set corresponding to the image component to be predicted based on the reference sample points.

2. Obtaining a first reference sample set corresponding to a predicted image component of a decoded block in the video image includes: obtaining reference sample points in reference rows and / or columns adjacent to the decoding block, the reference rows being formed by rows adjacent to an upper edge and a right-upper edge of the decoding block, and the reference columns being formed by columns adjacent to a left edge and a left-lower edge of the decoding block; determining a first reference sample set corresponding to the predicted image component based on the reference sample points; The method for predicting an image component according to claim 1 .

3. Processing the first reference sample set includes screening the first reference sample set; Screening the first reference sample set to obtain a second reference sample set includes: determining selected sample point locations based on sample locations corresponding to adjacent reference sample points in the first reference sample set; According to the determined selected sample point positions, selecting valid sample points corresponding to the selected sample point positions from the first reference sample set, and determining a second reference sample set based on the valid sample points obtained by the selection, wherein the number of valid sample points in the second reference sample set is less than or equal to a preset number; The method for predicting an image component according to claim 1 or 2.

4. Using the preset component values ​​as predicted values ​​corresponding to the predicted image components includes: determining the preset component values ​​based on a bit depth of the video image; The preset component values ​​are 1<<(BitDepth-1), where BitDepth is the bit depth of the predicted image component. The method for predicting an image component according to claim 1 .

5. The method for predicting image components comprises: and performing, for each sample point in the decoded block, a fill-in of a predicted value for a predicted image component of each sample point using the preset component value. The method for predicting an image component according to claim 1 .

6. After obtaining a prediction model corresponding to the predicted image component according to the model parameters, the method for predicting the image component includes: and performing a prediction process on a predicted image component of each sample point in the decoded block based on the prediction model to obtain a predicted value corresponding to the predicted image component of each sample point. The method for predicting an image component according to claim 1 .

7. A method for predicting image components, applied to an encoder, comprising the steps of: obtaining a first reference sample set corresponding to a predicted image component of a coding block in a video image; if a number of valid sample points in the first reference sample set is equal to 0, using a preset component value as a predicted value corresponding to the predicted image component; If the number of valid sample points in the first reference sample set is greater than or equal to four, processing the first reference sample set to obtain a second reference sample set, where the number of valid sample points in the second reference sample set is less than or equal to a preset number, the preset number being four; determining model parameters based on the second reference sample set when a number of valid sample points in the second reference sample set is equal to a preset number; The method for predicting image components further comprises: obtaining a prediction model corresponding to the predicted image component according to the model parameters; performing a prediction process on the predicted image component based on the prediction model to obtain a predicted value corresponding to the predicted image component; Obtaining a first reference sample set corresponding to predicted image components of a coding block in the video image includes: obtaining reference sample points in reference rows and / or columns adjacent to the coding block, the reference rows being formed by rows adjacent to an upper edge and a right-upper edge of the coding block, and the reference columns being formed by columns adjacent to a left edge and a left-lower edge of the coding block; determining a first reference sample set corresponding to the image component to be predicted based on the reference sample points.

8. Processing the first reference sample set includes screening the first reference sample set; Screening the first reference sample set to obtain a second reference sample set includes: determining selected sample point locations based on sample locations corresponding to adjacent reference sample points in the first reference sample set; According to the determined selected sample point positions, selecting valid sample points corresponding to the selected sample point positions from the first reference sample set, and determining a second reference sample set based on the valid sample points obtained by the selection, wherein the number of valid sample points in the second reference sample set is less than or equal to a preset number; The method for predicting an image component according to claim 7.

9. Using the preset component values ​​as predicted values ​​corresponding to the predicted image components includes: determining the preset component values ​​based on a bit depth of the video image; The preset component values ​​are 1<<(BitDepth-1), where BitDepth is the bit depth of the predicted image component. The method for predicting an image component according to claim 7.

10. The method for predicting image components comprises: and performing, for each sample point in the coding block, a fill-in of a predicted value for a predicted image component of each sample point using the preset component values. The method for predicting an image component according to claim 7.

11. After obtaining a prediction model corresponding to the predicted image component according to the model parameters, the method for predicting the image component includes: and performing a prediction process on predicted image components of each sample point in the coding block based on the prediction model to obtain a predicted value corresponding to the predicted image component of each sample point. The method for predicting an image component according to claim 7.

12. 1. A decoder comprising a memory and a processor, the memory is configured to store a computer program executable by the processor; The decoder, wherein the processor, when executing the computer program, is configured to perform the method for predicting an image component according to any one of claims 1 to 6.

13. 1. An encoder comprising a memory and a processor, the memory is configured to store a computer program executable by the processor; The encoder, wherein the processor, when executing the computer program, is configured to perform the method for predicting an image component according to any one of claims 7 to 11.

Images