INTRA-FRAME FORECASTING METHOD AND DEVICE, DECODER, AND ENCODER.

Abstract

An intra-frame forecasting method and device, a decoder, and an encoder; the method according to an embodiment of the present application comprises: using two or more different intra-frame forecasting modes to perform intra-frame forecasting for the respective blocks to be processed, so as to obtain two or more types of forecast blocks; and combining, according to a weighting matrix, the two or more types of forecast blocks obtained, so as to obtain forecast blocks of the blocks to be processed; the embodiment of the present application uses multiple intra-frame forecasting modes to determine multiple forecast blocks, thereby achieving the forecasting of complex textures, improving the quality of the intra-frame forecast, and consequently improving the compression performance;Furthermore, the intra-frame forecasting method in the modality of the present application uses a diversified weighting matrix to ensure the forecasting of complex textures, thereby improving the quality of the intra-frame forecast, and consequently improving the compression performance.

Description

The disclosure methods refer to video processing technologies, and more particularly to an intra-prognostic method and device, a decoder, and an encoder. BACKGROUND OF THE INVENTION There is a strong correlation between adjacent pixels in a video photograph, and spatial redundancy between adjacent pixels is eliminated through intra-forecast in video coding technologies to improve coding efficiency. The general intra-prognostic mode can predict simple textures, while complex textures need to be either divided into smaller blocks or encoded with more residues, which undoubtedly increases the complexity of the intra-prognostic model. In other words, in related intra-prognostic schemes, either the distortion cost is higher or the complexity is higher, leading to lower intra-prognostic quality. BRIEF DESCRIPTION OF THE INVENTION An overview of the material of interest detailed in the disclosure is provided below, which is not intended to limit the scope of protection of the claims. The disclosure provides an intra-prognostic method and device, a decoder, and an encoder, to improve the quality of intra-prognostic information. One modality of disclosure provides an intra-prognostic method, applied to a decoder and which includes the following operations. Intra-prognostic analysis is performed in a block that will be processed using two or more different intra-prognostic modes, to obtain two or more prognostic blocks corresponding to the different intra-prognostic modes. A target forecast block of the block to be processed is obtained based on weighting matrices and the two or more forecast blocks obtained. One form of disclosure provides a computer-readable storage medium that stores computer-executable instructions for implementing the intra-prognostic method described above. qq Lcnn / rznz / E / YiAi One version of the disclosure provides a decoder, which includes a memory and a processor. The memory can store instructions executable by a processor to implement the intra-prognostic method described earlier. One disclosure modality provides a decoder, which includes a decoding module, a forecasting module, and a combination module. The decoding module can be configured to decode a received bit stream, to obtain two or more different intra-forecast modes, a block to be processed, and weighting matrices. The forecasting module can be configured to perform intra-forecasting in the block to be processed by using two or more different intra-forecasting modes, to obtain two or more forecasting blocks corresponding to the different intra-forecasting modes. The combination module can be configured to obtain a target forecast block from the block to be processed based on the weighting matrices and the two or more forecast blocks obtained. One modality of disclosure provides an intra-prognostic method, applied to a coder and which includes the following operations. Intra-prognostic analysis is performed in a block that will be processed using two or more different intra-prognostic modes, to obtain two or more prognostic blocks corresponding to the different intra-prognostic modes. A target forecast block of the block to be processed is obtained based on weighting matrices and the two or more forecast blocks obtained. One form of disclosure provides a computer-readable storage medium, which stores computer-executable instructions for implementing the above intra-prognostic method applied to the encoder. One version of the disclosure provides an encoder, which includes a memory and a processor. The memory stores executable instructions for a processor to implement the aforementioned intra-prognostic method applied to the encoder. One version of the disclosure provides an encoder, which includes: a forecasting module, a combination module, and a processing module. The forecasting module can be configured to perform intra-forecasting in the block to be processed by using two or more different intra-forecasting modes, to obtain two or more forecasting blocks corresponding to the different intra-forecasting modes. The combination module can be configured to obtain a target forecast block from the block to be processed based on weighting matrices and the two or more forecast blocks obtained. qq Lcnn / eznz / E / YiAi The processing module can be configured to: test all or some of the possible combinations of forecasting modes and weight matrix derivation modes, calculate loss costs and select a combination with a low loss cost; use two or more different intra-forecasting modes and weight matrices in combination as the two or more different intra-forecasting modes and weight matrices for the intra-forecast; and write the two or more different determined intra-forecasting modes, weight matrix derivation modes and other information into a bit stream according to the syntax. One form of disclosure provides an intra-forecasting method, which includes the following operations. Intra-prognostic analysis is performed in a block that will be processed using two or more different intra-prognostic modes. During forecasting using each intra-forecasting mode, when a predefined number of pixels is predicted, a predefined number of forecast pixels is obtained from the block to be processed based on a weighting matrix and the pixels corresponding to the intra-forecasting mode that has been forecasted. A target forecast block of the block to be processed is obtained based on the predefined number of forecast pixels obtained. One form of disclosure provides an intra-prognostic device, which includes a prognostic module and a combination module. The forecasting module can be configured to perform intra-forecasting in a block that will be processed using two or more different intra-forecasting modes obtained through decoding, to obtain two or more forecasting blocks corresponding to the different intra-forecasting modes. The combination module can be configured to obtain a target forecast block from the block to be processed based on weighting matrices and the two or more forecast blocks obtained. According to the intra-prognostic method and device, the encoder and decoder provided in the disclosure modalities, intra-prognostication is performed on the block to be processed by using two or more different intra-prognostic modes to obtain two or more prognostic blocks. These two or more prognostic blocks are then combined based on weighting matrices to obtain the final prognostic block for the block to be processed. In the disclosure modalities, multiple prognostic blocks are determined using multiple intra-prognostic modes, thus achieving the prognostication of complex textures. This improves the quality of the intra-prognostication and increases compression performance. qq Lcnn / rznz / E / YiAi Furthermore, the intra-prognostic method provided in the disclosure modalities ensures the prediction of more complex textures through various weighting matrices. This improves the quality of the intra-prognostic data and increases compression performance. Additionally, the intra-prognostic method provided in the disclosure modalities is applicable to a wider range of scenarios. Other aspects and advantages of disclosure will be illustrated in the following description, and some of these will be evident from the description or understood through the implementation of disclosure. The objectives and other advantages of disclosure can be implemented or obtained by the structures specifically indicated in the description, claims, and accompanying drawings. Other aspects of the disclosure are understood after reading and understanding the attached drawings and the detailed description. BRIEF DESCRIPTION OF THE DRAWINGS The drawings described herein are used to provide further understanding of the disclosure and form part of the disclosure. The schematic representations of the disclosure and their descriptions are used to explain the disclosure and do not constitute an undue limitation of the disclosure. In the drawings: FIG. 1A is a schematic diagram of a hybrid block-based coding framework according to one disclosure modality; FIG. IB is a schematic composition block diagram of a video coding system according to a disclosure modality; FIG. 1C is a schematic composition block diagram of a video decoding system according to a disclosure modality; FIG. 2 is a schematic diagram of an intra-prognostic method according to a disclosure modality; FIG. 3 is a schematic diagram of intra-prognostic realization using four reference lines / columns according to a disclosure modality; FIG. 4 is a schematic diagram of nine intra-prognostic modes for a 4 x 4 block in H.264 according to a disclosure modality; FIG. 5 illustrates weighting diagrams of 64 geometric split mode (GPM) modes on a square block according to a disclosure modality; FIG. 6 illustrates weighting diagrams of 56 angular weighted forecast (AWP) modes in a square block according to a disclosure modality; qq Lcnn / eznz / E / YiAi FIG. 7 is a schematic flowchart of an intra-prognostic method according to a disclosure modality; FIG. 8 is a schematic diagram of intra-prognostic realization by using two different intra-prognostic modes according to a disclosure modality; FIG. 9A is a schematic diagram of positions to which the weights change, forming a straight line according to a modality of disclosure; FIG. 9B is a schematic diagram of positions to which the weights change, forming a curved line according to a modality of disclosure; FIG. 10 is a schematic flowchart of a first mode of mutual exclusion processing according to the disclosure; FIG. 11 is a schematic flowchart of a second mutual exclusion processing modality according to the disclosure; FIG. 12 is a schematic diagram of intra-prognostic mode storage according to a disclosure modality; FIG. 13 is a schematic structural diagram of an intraprognostic device according to a disclosure modality; and FIG. 14 is a schematic flowchart of another intra-prognostic method according to a disclosure modality. DETAILED DESCRIPTION OF THE INVENTION The disclosure will be described in detail below with reference to the accompanying drawings and modalities. It should be noted that the modalities in the disclosure and the aspects in the modalities can be combined without conflict. The intra-prognostic method provided in the disclosure modalities is applicable to the basic process of a video codec under the hybrid block-based coding framework shown in FIG. 6(a), but is not limited to the framework and process. The basic working principle of the video codec under the hybrid block-based coding framework shown in FIG. 1A is as follows. At the encoding end, a photograph is divided into blocks, and intra-prognosis is performed on a current block to generate a prediction block of the current block. The prediction block is subtracted from an original block of the current block to obtain a residual block. The residual block is transformed and quantized to obtain a quantization coefficient matrix, and the quantization coefficient matrix is ​​encoded to entropy and discharged to a bit stream. During block splitting, each photograph is divided into encoding units Larger QQI Qnn / Q7n7 / B / YIAI units (LCUs) are enclosed in squares of the same size (e.g., 128x128, 64x64, or similar). Each LCU can be divided into rectangular coding units (CUs) according to the rules. CUs can be further divided into forecast units (PUs), transformation units (TUs), or similar. At the decoding end, intra-forecast or inter-forecast is performed on the current block to generate a forecast block. On the other end, a bitstream is parsed to obtain the quantization coefficient matrix. Inverse quantization and inverse transformation are then performed on the quantization coefficient matrix to obtain a residual block. The forecast block and the residual block are summed to obtain a reconstructed block. The reconstruction blocks form a reconstructed image, and loop filtering is performed on the reconstructed image based on the image or blocks to obtain a decoded image. Similar operations to those at the decoding end are performed at the encoding end to obtain the decoded image.The decoded image can be used as a reference image for a subsequent image during forecasting. The decoded image obtained at the encoding end is also referred to as a reconstructed image. The current block can be divided into forecast units during forecasting and into transformation units during transformation, and the division of the forecast and transformation units can be different. Block division information and mode or parameter information such as forecast, transformation, quantization, entropy coding, and loop filtering, which are determined at the encoding end, need to be downloaded to the bitstream if necessary.The decoding end determines the same block splitting information, mode information, or parameter information such as forecasting, transformation, quantization, entropy coding, and loop filtering as the encoding end through syntactic analysis and analysis based on existing information, in order to ensure that the decoded picture obtained at the encoding end is the same as the decoded picture obtained at the decoding end. The intra-prognostic method provided in the disclosure modalities is used for the intra-prognostic module in the framework shown in FIG. 1A, which can be applied to either the encoding or decoding end. At the encoding end, information such as the adopted intra-prognostic modes and weighting matrices is determined, and then the disclosure's intra-prognostic analysis is performed based on these determined modes and weighting matrices. At the decoding end, information such as the adopted intra-prognostic modes and weighting matrices is obtained by decoding the bitstream, and then the disclosure's intra-prognostic analysis is performed based on these modes. QQI Qnn / Q7n7 / B / YIAI intra-prognostic and weighting matrices obtained. Figure 1B is a schematic block diagram of a video coding system according to one disclosure modality. As shown in Figure 1B, the video coding system 11 may include: a transformation unit 111, a quantization unit 112, a mode selection and coding control logic unit 113, an intra-prognostic unit 114, an inter-prognostic unit 115 (which includes motion compensation and motion estimation), an inverse quantization unit 116, an inverse transformation unit 117, a loop filter unit 118, an encoding unit 119, and a decoded image compensation unit 110. An original input video signal is split into coding tree units (CTUs) to obtain a video reconstruction block. The mode selection and coding control logic unit 113 determines an coding mode.The residual pixel information obtained through intra- or inter-forecasting is then processed by transformation unit 111 and quantization unit 112 to transform the reconstructed video block. This includes transforming the residual information from a pixel domain to a transformation domain and quantizing the resulting transformation coefficient to further reduce the data flow per second. Intra-forecasting unit 114 is configured to perform intra-forecasting on the reconstructed video block. Intra-forecasting unit 114 is configured to determine an optimal intra-forecasting mode (i.e., target forecasting mode) for the reconstructed video block. Inter-forecasting unit 115 is configured to perform inter-forecasting encoding on the received reconstructed video block with respect to one or more blocks in one or more reference photographs to provide weather forecasting information.Motion estimation is a process of generating a motion vector (MV), and the motion of the reconstructed video block can be estimated according to the MV. Motion compensation is then performed based on the MV determined by motion estimation. After an inter-forecast mode is determined, the inter-forecast unit 115 is further configured to provide inter-forecast data selected by the encoding unit 119 and to send computationally determined MV data to the encoding unit 119. Additionally, the inverse quantization unit 116 and the inverse transformation unit 117 are configured to reconstruct the reconstructed video block to reconstruct the remaining block in the pixel domain.The loop filter unit 118 removes square-effect artifacts from the reconstructed residual block. This reconstructed residual block is then added to a forecast block in the decoded photo compensation unit 110 to generate the reconstructed video block. The encoding unit 119 is configured to encode various encoding parameters and quantized transformation coefficients. The decoded photo compensation unit 110 is configured for... QQI Qnn / Q7n7 / B / YIAI stores the reconstructed video block for forecast reference. As video frames are encoded, new reconstructed video blocks can be continuously generated, and all these reconstructed video blocks are stored in the decoded frame compensation unit 110. Figure 1C is a schematic block diagram of a video decoding system according to one disclosure modality. As shown in Figure 1C, the video decoding system 12 may include: a decoding unit 121, an inverse transform unit 127, an inverse quantization unit 122, an intra-forecast unit 123, a motion compensation unit 124, a loop filter unit 125, and a decoded image compensation unit 126. After an input video signal is encoded by the video encoding system 11, a bitstream of the video signal is downloaded. The bitstream is fed into the video decoding system 12 and is processed by the decoding unit 121 first to obtain a decoded transform coefficient.The transformation coefficient is processed by the inverse transformation unit 127 and the inverse quantization unit 122 to generate a residual block in a pixel domain. The intra-forecast unit 123 can be configured to generate forecast data for a current video decoding block based on a specified intra-forecast address and data from a previously decoded block of a current image. The motion compensation unit 124 analyzes a video frame and other associated syntactic elements to determine forecast information for the video decoding block and, using this forecast information, generates a forecast block for the video decoding block being decoded.The residual block from the inverse transform unit 127 and the inverse quantization unit 122, along with the corresponding forecast block generated by the intra-forecast unit 123 or the motion compensation unit 124, are combined to form a decoded video block. The loop filter unit 125 removes square-effect artifacts from the decoded video signal to improve video quality. The decoded video block is then stored in the decoded photo compensation unit 126, which stores the reference image for subsequent intra-forecast or motion compensation, and also for downloading the video signal to obtain a recovered original video signal. The intra-prognostic method provided in the disclosure modalities is applied to intra-prognostic unit 114 of video coding system 11 and intra-prognostic unit 123 of video decoding system 12 to forecast the current block (the block to be encoded or the block to be decoded), in order to obtain a corresponding forecast block. That is, the intra-prognostic method provided in the modalities of the QQI Qnn / Q7n7 / B / YIAI disclosure can be based on intra-prognostic video coding method or video decoding method. The intra-forecasting method forecasts a current block using reconstructed pixels around the current block that have been encoded and decoded as reference pixels. For example, as shown in Figure 2, the white 4x4 block is the current block, and the gray pixels in the leftmost row and top column are reference pixels for the current block. These reference pixels are used to forecast the current block during intra-forecasting. All of these reference pixels may be available (i.e., they have been encoded and decoded), or some may not. For example, if the current block is on the far left of the entire photograph, the reference pixels to the left of the current block are unavailable.Alternatively, if the lower left portion of the current block has not been encoded and decoded when the current block is encoded and decoded, the lower left reference pixels are unavailable. Unavailable reference pixels can be filled with available reference pixels, some values, or some methods, or left unfilled. The multi-line reference (MRL) intra-prognostic method can use more reference pixels to improve coding efficiency. For example, Figure 3 is a schematic diagram of an intra-prognostic modality that uses four reference lines / columns in the related technique. There are several intra-prognostic modes. Figure 4 shows nine intra-prognostic modes for a 4x4 block in H.264. In mode 0, the pixels above the current block are copied vertically to the current block as prognostic values; in mode 1, the reference pixels on the left side of the current block are copied horizontally as prognostic values; in mode 2 DC, an average of eight points A to D and I to L is used as the prognostic value for all points; in modes 3 to 8, the reference pixels are copied at different angles to the corresponding points in the current block. Since some positions in the current block do not exactly correspond to the reference pixels, a weighted average of the reference pixels can be used—that is, sub-pixel interpolations of the reference pixels. Additionally, there are planar, planar, and other modes in the related technique.With the development of technologies and the expansion of the blocks, the number of angular forecasting modes is increasing. For example, High Efficiency Video Coding (HEVC) uses a total of 35 intra-forecasting modes, including one planar mode, one DC mode, and 33 angular modes. As another example, Versatile Video Coding (WC) uses a total of 67 forecasting modes, including one planar mode, one DC mode, and 65 angular modes. As yet another example, China's Audio Video Coding Standard (AVS3) uses a total of 66 forecasting modes, including one DC mode, one planar mode, one bilinear mode, and 63 angular modes. qq Lcnn / eznz / E / YiAi There are also techniques for improving intra-forecast accuracy, such as enhancing a sub-pixel interpolation of a reference pixel and filtering a forecast pixel. For example, the Multiple Intra-Forecast Filter (MIPF) in AVS3 uses different filters for different block sizes to generate forecast values. For pixels in different positions within the same block, one filter is used for pixels closer to the reference pixel to generate forecast values, and another filter is used for pixels farther from the reference pixel to generate forecast values. Techniques for filtering forecast pixels can include, for example, the Intra-Forecast Filter (IPF) in AVS3, which uses reference pixels to filter forecast values. Intra-prognostic modes include DC mode, planar mode, bilinear mode, and similar modes. However, all of these modes can only predict simple textures. Angular modes are becoming increasingly common, but these can only predict by following straight lines at an angle. In the development of the WC standard (also known as H.266), Geometric Split Mode (GPM) was introduced as an inter-forecasting mode. In the development of AVS3, Angle Weighted Forecasting (AWP) was introduced as an inter-forecasting mode. The GPM or AWP uses two reference blocks of the same size as the current block. 100% of the pixel values ​​from corresponding positions in the first reference block are used in some pixel locations, 100% of the pixel values ​​from corresponding positions in the second reference block are used in some pixel locations, and pixel values ​​from corresponding positions in both reference blocks are used proportionally in the blending area. The specific allocation of weights is determined by the GPM or AWP modes. Alternatively, the GPM or AWP uses two reference blocks that are not the same size as the current block, and a desired portion of each is used as a reference block. In other words, the portion with non-zero weights is used as a reference block, and the portion with zero weights is excluded. Figure 5 illustrates weighting diagrams of 64 GPM modes in a square block according to a disclosure modality. As shown in Figure 5, black indicates that the weight of a corresponding position in a first reference block is 0%, white indicates that the weight of a corresponding position in the first reference block is 100%, and the gray area indicates that the weight of a corresponding position in the first reference block is greater than 0% and less than 100% depending on the color hue. The weight of a corresponding position in a second reference block is 100% minus the weight of the corresponding position in the first reference block. Figure 6 illustrates weighting diagrams of 56 AWP modes in a square block according to a disclosure modality. As shown in Figure 6, black indicates that the weight of a corresponding position in a first reference block is 0%, white indicates that the weight of a corresponding position in the first reference block is 100%, and the gray area indicates that the weight of a corresponding position in the first reference block is greater than 0% and less than 100% depending on the color hue. The weight of a corresponding position in a second reference block is 100% minus the weight of the corresponding position in the first reference block. The weight derivation methods of GPM and AWP are different. GPM determines an angle and offset for each mode, and then calculates a weighting matrix for each mode. AWP first creates a one-dimensional line of weights, and then expands the one-dimensional line of weights over the entire matrix using a method similar to the angular intra-forecast method. Only rectangular splitting exists for CUs, PUs, or TUs in previous encoding and decoding technologies, while GPM and AWP achieve the effect of non-rectangular splitting for forecasting without the actual splitting. GPM and AWP use a weight mask of the two reference blocks, i.e., the aforementioned weight diagram. This mask determines the weights of the two reference blocks when a forecast block is generated. This can be understood simply as some positions in the forecast block being derived from the first reference block and some from the second, while the merging area is obtained by weighting the corresponding positions in the two reference blocks, thus achieving a smoother merging.The GPM and AWP do not divide the current block into two CUs or PUs by the divide line, so the transformation, quantization, inverse transformation, and inverse quantization of the residuals after forecasting are also performed on the current block as a whole. The intra-forecasting method provided in the disclosure modalities may include the following operations. Intra-forecasting is performed on a block to be processed using two or more different intra-forecasting modes to obtain two or more forecast blocks corresponding to the different intra-forecasting modes. A forecast block from the block to be processed is obtained by combining the two or more forecast blocks obtained based on weighting matrices. In the disclosure modalities, multiple forecast blocks are determined using multiple intra-forecasting modes, thus achieving the forecasting of complex textures. In this way, the quality of the intra-forecasting is improved, and compression performance is increased. Furthermore, the intra-prognostic method provided in the disclosure modalities ensures the prediction of more complex textures through various weighting matrices. This improves the quality of the intra-prognostic data and increases compression performance, and also makes the method applicable to a wider range of scenarios. qq Lcnn / eznz / E / YiAi Figure 7 is a schematic flowchart of an intra-prognostic method according to one disclosure modality. As shown in Figure 7, the method includes the following operations. In 700, the intra-prognostic process is carried out in a block that will be processed by using two or more different intra-prognostic modes respectively, to obtain two or more prognostic blocks corresponding to the different intra-prognostic modes. In this type of disclosure, the block to be processed can be a block that will be encoded, that is, processed by an encoder, or a block that will be decoded, that is, processed by a decoder. In an exemplary modality, intra-prognostic modes may include, but are not limited to, a DC mode, a planar mode, a planar mode, a bilinear mode, an angular (AP) prognostic mode, and other intra-prognostic modes, as well as techniques for improving intra-prognostic, for example, improving a sub-pixel interpolation of a reference pixel, or filtering a prognostic pixel, such as MIPF and IPF. An intraprognostic mode that generates the prognostic block independently of other intraprognostic modes is referred to as a first intraprognostic mode (or a basic intraprognostic mode in the disclosure), and may include a DC mode, a planar mode, a planar mode, a bilinear mode, and an AP mode. In other words, for the basic intraprognostic mode, if the reference pixel and the basic intraprognostic mode are determined, the prognostic block can be determined. An intraprognostic mode that relies on a basic intraprognostic mode to determine a prognostic block is referred to as a second intraprognostic mode (or an enhanced intraprognostic mode in the disclosure), and may include techniques such as MIPF and IPF that enhance the intraprognostic process. In other words, enhanced intraprognostic modes cannot generate the prognostic block independently.For example, a basic intra-prognostic mode such as AP mode can generate a prognostic block based on reference pixels, while an enhanced intra-prognostic mode such as MIPF can generate or determine a prognostic block by using different filters for pixels in different positions based on the aforementioned AP mode. In an exemplary modality, the two or more different modes of intra-prognostic include at least one basic mode of intra-prognostic. In an exemplary modality, taking two different intra-prognostic modes that are used to perform intra-prognostic analysis in the block to be processed as an example, the two different intra-prognostic modes are both basic intra-prognostic modes. In an exemplary modality, the basic intra-prognostic mode is used in conjunction with the enhanced intra-prognostic mode, i.e., the basic intra-prognostic mode adopted qq Lcnn / eznz / E / YiAi can be further combined with the enhanced intra-prognostic mode to forecast the block to be processed. In an exemplary scenario, let's consider two different intra-prognostic modes used to forecast the block to be processed. These two modes include a basic intra-prognostic mode and an enhanced intra-prognostic mode. For instance, the first and second intra-prognostic modes both use the same AP mode, but the first mode does not use an enhanced intra-prognostic mode (e.g., IPF does not use the enhanced intra-prognostic mode), while the second mode does. Alternatively, both the first and second intra-prognostic modes use the same AP mode, but the first mode uses one option of an enhanced intra-prognostic mode, and the second mode uses a different option. In the dissemination methods, at least two different intra-prognostic modes are used for the forecast of the block to be processed, so that the forecast of the block to be processed can be made from multiple perspectives, which is suitable for the forecast of complex textures and facilitates improving the quality of the intra-prognostic. The process of the previous intra-prognostic method is applicable to the encoder as well as the decoder. In an exemplary mode, at the decoder end, the block to be processed is a block to be decoded, and before the operation at 700, the method may also include the following operation. A stream of bits is syntactically analyzed to obtain the two or more different intra-forecast modes, the block to be processed, and the weighting matrices. In an exemplary form, at the encoder end, before the operation at 700, the method may also include the following operation. All or some possible combinations of forecasting modes and weighting matrix derivation modes are tried, the loss costs of the combinations are calculated, and a combination with a low loss cost is selected; and two or more different intra-forecasting modes and weighting matrices are used in the combination as the two or more different intra-forecasting modes and weighting matrices for the intra-forecasting. The method may also include the following operation. Information such as the two or more different intra-forecast modes and weight matrix derivation modes determined is written to a bit stream according to the syntax. All possible combinations herein can include: all possible Lcnn / rznz / E / YiAi combinations of all possible first intraprognostic modes, all possible second intraprognostic modes, and all possible weight matrix derivation modes. For example, assuming there are 66 intraprognostic modes available in total, there are 66 possible first intraprognostic modes, the second intraprognostic mode is definitely not the same as the first intraprognostic mode, and thus there are 65 possibilities. There are 56 weight matrix derivation modes (in the case of AWP). Thus, the total number of combinations of any two different intraprognostic modes and any one weight matrix derivation mode is 66 x 65 x 56. In an exemplary modality, the method for calculating loss costs may include one or any combination of the following algorithms: sum of absolute differences (SAD), sum of absolute transformed differences (SATD), or velocity distortion optimization (RDO). In an exemplary modality, a first selection such as coarse selection is made using SATD and / or SAD to determine the candidate combinations of all or some possible combinations of the forecasting modes and weight matrix derivation modes; and then a second selection such as fine selection is made using the RDO to determine a combination with a minimum loss cost from the candidate combinations. In one modality, coarse selection may also include: using some fast algorithms to reduce attempt times. For example, when an angular intra-forecast mode results in a high cost, a predefined number of intra-forecast modes adjacent to the angular intra-forecast mode are no longer attempted. In an exemplary modality, before attempting combinations of forecasting modes and weight matrix derivation modes, the method may also include the following operation. The texture of the current block to be processed is analyzed, for example using gradients. The operation that attempts combinations of forecasting modes and weight matrix derivation modes may also include the following operation. The intra-prognostic modes to be attempted are determined based on the result of the texture analysis. For example, in a direction with a stronger texture (e.g., greater than a predefined high threshold) of the current block to be processed, during the rough selection to attempt combinations of forecast modes and weight matrix derivation modes, intra-forecast modes in similar directions (i.e., the direction plus or minus a predefined angle) with a strong texture are selected to attempt as many as possible. As another example, in a direction with a weaker texture (e.g., less than a predefined low threshold) of the current block to be processed, intra-forecast modes in similar directions with a weaker texture are not selected to attempt as many as possible. It should be noted that the loss cost includes not only the cost of codewords occupied in the bitstream by the first intra-forecasting mode, the second intra-forecasting mode, and the weight matrix derivation modes, but also the cost of flags and quantization coefficients to be transmitted in the bitstream for transformation, quantization, entropy coding, or similar operations on the forecast residuals, and the distortion cost of the reconstructed blocks. Generally speaking, the cost is not the space occupied, but rather the distortion cost—that is, the difference between the forecast block and the original block, or the distortion difference between the original snapshot and the snapshot obtained after encoding or decoding. Minimum cost means minimum distortion, that is, minimal loss in the compression process and the highest encoding quality. In an exemplary modality, after determining the combination with the minimum loss cost, before the operation at 700, the method may also include the following operation. If the selected minimum loss cost is less than or equal to the costs of other forecasting modes, and the other forecasting modes may include other intra-forecasting modes or inter-forecasting modes, the coder selects the intra-forecasting modes in combination with the selected minimum loss cost according to the disclosure as the forecasting modes for the block to be processed; if the selected minimum loss cost is greater than the costs of other forecasting modes, the coder selects some other forecasting modes as the forecasting modes for the block to be processed. In an exemplary form, the method at the encoder end may also include the following operation. Information such as the two or more different intra-prognostic modes and determined weight matrix derivation modes is written into a bit stream according to the syntax. The method may also include the following operation. Intra-prognostic and subsequent coding processing are performed on the block to be processed based on the two or more different intra-prognostic modes and weighting matrices determined according to the disclosure's intra-prognostic method. In 701, a target forecast block is obtained from the block to be processed based on weighting matrices and the two or more forecast blocks obtained. In one exemplary mode, at the encoder end, the weighting matrices can be determined by calculating the loss cost as described in 700. At the decoding end, the bit stream is parsed syntactically based on the syntax and the weighting matrices are obtained based on the weighting matrix derivation modes obtained. The weighting matrix can be determined by reference to the GPM or AWP weight derivation method in the inter-forecast. If the GPM or AWP forecasting mode is used in the same codec standard or codec, the GPM or AWP weight derivation method can be used to determine the weighting matrix in this disclosure mode, allowing for the reuse of some of the same logic. For example, if the inter-forecast in AVS3 uses AWP, an AWP weight derivation method can be used to determine the weighting matrix in this disclosure mode. The method for determining the weighting matrix in this disclosure mode can also differ from the GPM or AWP method used in the same codec standard or codec. For example, a different number of modes, a different combination area algorithm, or a different parameter can be used. In an exemplary mode, as shown in FIG. 8, taking that the intra-prognostic is performed in the block to be processed by using two different intra-prognostic modes to obtain a first prognostic block and a second prognostic block as an example, the operation in 701 may include the following operations. A first product is calculated from a matrix corresponding to the first forecast block and the first weighting matrix, and a second product is calculated from a matrix corresponding to the second forecast block and the second weighting matrix. A sum is calculated from the first product, the second product, and a predefined value. The calculated sum is normalized to obtain the target forecast block. In an exemplary mode, the second weighting matrix is ​​a difference between the maximum weight (e.g., 8) and the first weighting matrix. The normalization process may include: right-shifting the calculated sum by a predefined number of bits (e.g., 3 bits), to obtain the target forecast block of the block to be processed. For example, a value of an element predMatrixSawp[x][y] in predMatrixSawp = ((predMatrixO[x][y] * AwpWeightArrayY[x][y] + predMatrixl[x][y] * ( 8 - AwpWeightArrayY[x][y]) + 4 ) >> 3). predMatrixSawp indicates the target forecast block, predMatrixSawp[x][y] indicates an array of the target forecast block, predMatrixO[x][y] indicates an array corresponding to the first forecast block, predMatrixl[x][y] indicates an array corresponding to the second forecast block, and AwpWeightArrayY[x][y] indicates the first weighting array. In one modality, after the operation in 701, the method may also include the following operation. Intra-prognostication is performed in the target prognostic block of the block to be processed by using an enhanced intra-prognostic mode, and a prognostic result is used as the target prognostic block of the block to be processed. In one exemplary instance, in all the possible weighting matrices determined, not all points in each of the weighting matrices have the same weight. In other words, at least one of all the possible weighting matrices includes at least two different weights. In an exemplary modality, all possible weighting matrices include at least two different weights. In an exemplary configuration, at least one of all possible weighting matrices includes at least two different weights, and at least one of all possible weighting matrices includes only the same weights. For example, if the minimum weight is 0 and the maximum weight is 8, in one weighting matrix, some points have a weight of 0 and some points have a weight of 8; in another weighting matrix, all points have a weight of 4. A weighting matrix that includes only the same weights can have any value greater than the minimum weight and less than the maximum weight. In an exemplary configuration, for one or more weighting matrices that include at least two different weights, according to a minimum and a maximum weight, each point in the block to be processed is obtained by weighting forecast values ​​derived from two intra-forecasting modes. For example, there are eight values ​​for the weights, i.e., from 0 to 8. For example, if intra-forecasting is performed in the block to be processed using two different intra-forecasting modes, 0 indicates that the point is derived exclusively from a forecast value derived from one intra-forecasting mode, and 8 indicates that the point is derived exclusively from a forecast value derived from the other intra-forecasting mode. Assuming that the minimum weight is set to 1 and the maximum weight is set to 7, all points in this weighting matrix are obtained by weighting the forecast values ​​derived from the two intra-forecasting modes.However, not all points carry the same weight. In one exemplary mode, intra-forecasting is performed within the block to be processed using two different intra-forecasting modes. Only one of the possible weighting matrices includes just two types of weights: one weight type indicates that a forecast value for a corresponding point is derived exclusively from a corresponding point value in the first forecast block, and the other weight type indicates that a forecast value for a corresponding point is derived exclusively from a corresponding point value in the second forecast block. For example, the two weight types are 0 and 1. In one exemplary mode, intra-forecasting is performed on the block to be decoded using two different intra-forecasting modes. One of all possible weighting matrices includes a plurality of weight types. A maximum weight and a minimum weight (e.g., 0) indicate, respectively, that the forecast values ​​of the corresponding points are derived exclusively from the values ​​of the corresponding points in the first forecast block and the values ​​of the corresponding points in the second forecast block. A weight other than the maximum or minimum weight indicates that a forecast value for a corresponding point is derived from a weighted average of the values ​​of the corresponding point in the first forecast block and the corresponding point in the second forecast block. An area consisting of weights other than the maximum or minimum weight can be referred to as a combination area. In one exemplary mode, as shown in FIG. 9A, in all possible weighting matrices, if a weighting matrix includes only two types of weights, the positions where the weights change form a straight line; if a weighting matrix includes multiple types of weights, the positions with the same weights in the combination area form a straight line. In one mode, the above straight lines are all horizontal-vertical, or the above straight lines are not all horizontal-vertical. In an exemplary mode, as shown in FIG. 9B, in all possible weighting matrices, if a weighting matrix includes only two types of weights, the positions where the change in weights forms a curved line; if a weighting matrix includes multiple types of weights, the positions with the same weights in the combination area form a curved line. The various weighting matrices provided in the disclosure modalities ensure the forecasting of more diverse forecast blocks and enable the intra-forecasting method provided in the disclosure modalities to be applicable to more scenarios. In one mode, for example, IAWP is used in AVS3 for inter-prognosis and includes 56 weighting matrices. In another mode of disclosure, 64 weighting matrices are used for intra-prognosis, where 56 of these weighting matrices are identical to the AWP weighting matrices. For example, the first 56 weighting matrices are identical to the AWP weighting matrices, and each of the remaining eight weighting matrices includes only one type of weight, i.e., 1, 2, ..., 7, and 8 respectively. For the eight weighting matrices, the total weight is 16; that is, a weight of 1 indicates a 1:15 weighting and a weight of 2 indicates a 2:14 weighting. In this way, when the number of modes of the 64 weighting matrices is binarized, a six-bit codeword can be used for each weighting matrix. Alternatively, the total weight is 8.In that case, 8 is the maximum weight, that is, a weight of 1 indicates a 1:7 weighting and a weight of 2 indicates a 2:6 weighting. Since inter-forecasting uses correlation in the time domain, the reconstructed image of the reference image is used as the reference block. Intra-forecasting uses correlation in the space domain, and the reconstructed pixels around the block to be processed are used as the reference pixels. Closer distances in the space domain indicate stronger correlation, and farther distances indicate weaker correlation. Therefore, if a weighting matrix makes all pixel positions used for a forecast block very far from the reference pixels, such a weighting matrix may not be used in the disclosure modalities to ensure the intra-forecasting effect. In an exemplary modality, the block size (e.g., the block to be processed) in the intra-prognostic method provided in the disclosure modalities may include, but is not limited to, the following. A block width is greater than or equal to a first threshold TH1, a block height is greater than or equal to a second threshold TH2, and the first threshold TH1 and the second threshold TH2 can be 8, 16, 32, 64, 128, or similar. The first threshold TH1 can be equal to the second threshold TH2, for example, first threshold TH1 = second threshold TH2 = 8. Alternatively, a block pixel count is greater than or equal to a third threshold TH3, and the third threshold TH3 can be 8, 16, 32, 64, 128, or similar. Alternatively, the block width is less than or equal to a fourth TH4 threshold, the block height is less than or equal to a fifth TH5 threshold, and the fourth TH4 threshold and the fifth TH5 threshold can be 8, 16, 32, 64, 128, or similar. The fourth TH4 threshold can be equal to the fifth TH5 threshold. Alternatively, a number of pixels in the block is less than or equal to the sixth TH6 threshold, and the sixth TH6 threshold can be 8, 16, 32, 64, 128, or similar. By limiting the block size, the impact of complexity due to the use of multiple forecasting modes for the forecast in the total system is reduced, and the trade-off between compression performance and complexity is well achieved by discarding some blocks of inapplicable sizes, thus better ensuring the applicability of the disclosure. In one exemplary mode, block splitting becomes increasingly flexible as technology evolves. Besides square blocks, the splitting method can also support blocks with aspect ratios such as 1:2, 1:4, 1:8, 2:1, 4:1, 8:1, and similar ratios. In the intra-prognostic method provided in the disclosure modes, the inventors of the disclosure found that blocks with certain aspect ratios, or blocks with certain sizes, such as 1:4 or 4:1, 1:8 or 8:1, 8x32, 8x64, 32x8, or 64x8, may not provide good or significant compression performance. In one disclosure mode, the block size can be adjusted by configuring the aspect ratio. For example, the width-to-height ratio is less than or equal to a predefined THR threshold, and the width-to-height ratio is less than or equal to the THR threshold. In one example, the block size and block aspect ratio can be set simultaneously. For instance, the intra-prognostic method provided in the disclosure modalities can be used if the block size satisfies the following criteria: block height is greater than or equal to 8, block width is greater than or equal to 8, block width-to-height ratio is less than or equal to 4, and block height-to-width ratio is less than or equal to 4; otherwise, the intra-prognostic method provided in the disclosure modalities is not used by default. In an exemplary modality, at the encoder end, before the operation at 700 in the disclosure, the method may also include the following operation. A flag is set at the photograph level to indicate whether to use the intra-prognostic method in the disclosure modalities for the current photograph to be processed, i.e., whether to proceed with the operation on 700, and the flag is written into the bit stream according to the syntax, so that the intra-prognostic method is performed at the decoder end based on the flag. In one modality, if the intra-prognostic method in the disclosure modalities is used for intra-photographs (such as Photograph I) but not for inter-photographs (such as Photograph B and Photograph P), when the flag indicates that the current photograph to be processed is an intra-photograph, it indicates that the decoder end proceeds with the operation at 700; when the flag indicates that the current photograph to be processed is an inter-photograph, it indicates that the decoder end exits the disclosure process, and relevant intra-prognostic techniques can be used. As another example, if the intra-forecast method in disclosure modalities is not used for intra-photographs (such as photograph I) but is used for inter-photographs (such as photograph B and photograph P), when the flag indicates that the current photograph to be processed is an intra-photograph, it indicates that the decoder end exits the disclosure process, and relevant intra-forecast techniques can be used; when the flag indicates that the current photograph to be processed is an inter-photograph, it indicates that the decoder end proceeds with the operation at 700. As another example, if the disclosure method is used for some inter-photographs but not for other inter-photographs, when the flag indicates that the current photograph to be processed is from some inter-photographs, it indicates that the decoder end proceeds with the operation at 600; when the flag indicates that the current photograph to be processed is from the other inter-photographs, it indicates that the decoder end exits the disclosure process, and relevant intra-forecast techniques can be used. Correspondingly, at the decoder end, before the operation at 700 of the intra-forecast method shown in FIG. 7, the method may further include: syntactically analyzing the bit stream according to the syntax to obtain the flag. QQI Qnn / Q7n7 / B / YIAI For example, if the intra-prognostic method in the disclosure modalities is used for intra-photographs (such as photograph I) but not for inter-photographs (such as photograph B and photograph P), when the flag obtained through decoding indicates that the current photograph to be processed is an intra-photograph, the operation in 700 is carried out; when the flag obtained through decoding indicates that the current photograph to be processed is an inter-photograph, the disclosure process ends, and relevant intra-prognostic techniques can be used. As another example, if the intra-prognostic method in the disclosure modalities is not used for intra-photographs (such as photograph I) but is used for inter-photographs (such as photograph B and photograph P), when the flag obtained through decoding indicates that the current photograph to be processed is an intra-photograph, the disclosure process ends, and relevant intra-prognostic techniques can be used; when the flag obtained through decoding indicates that the current photograph to be processed is an inter-photograph, the operation in 700 proceeds. As another example, if the disclosure method is used for some Interphotographs but not for other Interphotographs, when the flag obtained through decoding indicates that the current photograph to be processed is from some Interphotographs, the operation proceeds in 700; when the flag obtained through decoding indicates that the current photograph to be processed is from the other Interphotographs, the disclosure process ends, and relevant techniques for intra-forecast can be used. In an exemplary modality, at the encoder end, before the operation at 700 of the disclosure, the method may also include the following operation. A flag is set below a photograph level and above a CU level (e.g., mosaic, cut, patch, or LCU) to indicate whether the intra-prognostic method of disclosure modalities is used for a specified area, and the flag is written into the bit stream according to the syntax, to instruct the decoder to perform the corresponding intra-prognostic method based on the flag. By configuring different levels of flags in the disclosure modes, you can flexibly optimize the settings according to different application scenarios and video content, and also reduce the data throughput per second. For example, if there are multiple CUs in a LCU and none of them use the intra-prognostic method provided in the disclosure modes, you only need to set one flag at the LCU level to indicate that all CUs in the LCU do not use the intra-prognostic method provided in the disclosure modes. In other words, there is no need to set a separate flag for each CU; you only need to set one flag instead of multiple flags. QQI Qnn / Q7n7 / B / YIAI Correspondingly, at the decoder end, before the operation on 700, the method may further include: syntactically parsing the bit stream according to the syntax to obtain the flag. In an exemplary modality, at the encoder end, before the operation at 700 in the disclosure, the method may also include the following operation. A mutually exclusive enhanced prognostic mode is established with the intra-prognostic method provided in the disclosure modalities, to better determine the intra-prognostic modes during the intra-prognostic process. At the decoder end, the mutually exclusive enhanced forecast mode, established with the intra-forecast method provided in the disclosure modalities, is obtained by syntactically analyzing the bitstream. If it is determined that the intra-forecast method provided in the disclosure modalities is implemented in the block to be processed, the mutually exclusive enhanced forecast mode is not used; or, if it is determined that the mutually exclusive enhanced forecast mode is used for the block to be processed, the intra-forecast method of the disclosure modalities is not implemented.By configuring the mutually exclusive enhanced forecasting mode, it is not necessary to transmit a flag in the bitstream to indicate whether to use the mutually exclusive enhanced forecasting mode, thus avoiding unnecessary transmission of the flag in the bitstream and generally achieving better compression performance. In an exemplary modality, the mutually exclusive enhanced forecasting mode may include, for example, IPF and derived tree (DT). For example, the mutually exclusive enhanced prognostic mode is DT, a technique in AVS3. DT can divide the current UC into rectangular Pus, and correspondingly, into smaller Tus. When disclosure modalities are performed in combination with DT, the intra-prognostic method provided in the disclosure modalities can be used on one or multiple PUs divided based on DT, but complexity increases. For example, the mutually exclusive enhanced prognostic mode is an enhanced intra-prognostic (IIP), a technique in AVS3, and IIP can use more complex filters to obtain prognostic values. The inventors of the disclosure found during testing of the intra-prognostic process provided in the disclosure modalities that using IIP, DT, or IPF increases the calculation or complexity of the intra-prognostic process. Therefore, in the modalities, by configuring the mutually exclusive relationship between these enhanced prognostic modes and the disclosure's intra-prognostic process, a good balance between performance and complexity is achieved, thus better ensuring the disclosure's applicability. The intra-prognostic method in disclosure and the IPF are taken as an example qq Lcnn / eznz / E / YiAi to illustrate the case of mutual exclusion. In an exemplary modality, it is assumed that the disclosure intra-prognostic method is mutually exclusive with the IPF. The flag used in the disclosure modalities to indicate whether to perform the intra-prognostic method in the disclosure modalities on the current photograph to be processed is decoded first, and then the IPF flag is decoded. As shown in FIG. 10, the process mainly includes the following operations. If the disclosure intra-forecasting method is used for the current block, there is no need to decode the IPF flag; that is, there is no need to transmit the IPF flag in the bitstream. If the disclosure intra-forecasting method is not used for the current block, the IPF flag is additionally decoded to determine whether the IPF needs to be used. If the current block uses the IPF, the current block is forecasted using the IPF in combination with another intra-forecasting method. If the current block does not use the IPF, the current block is forecasted using another intra-forecasting method. In an exemplary modality, it is assumed that the disclosure intra-prognostic method is not mutually exclusive with the IPF. The flag used in the disclosure modalities to indicate whether to perform the disclosure intra-prognostic method on the current photograph to be processed is decoded first, and then the IPF flag is decoded. As shown in FIG. 11, the process mainly includes the following operations. The IPF flag needs to be decoded, regardless of whether the disclosure's intra-prognostic method is used. Furthermore, if both the disclosure's intra-prognostic method and the IPF are used, the current block is forecasted using both the disclosure's intra-prognostic method and the IPF. It should be noted that the modalities shown in Figures 10 and 11 are only examples where the intraprognostic method described in the disclosure is mutually exclusive with one technology. If the intraprognostic method described in the disclosure is also mutually exclusive with other technologies, the process is more complex, but the principle remains the same, which can be easily understood by someone skilled in the technique based on the modalities shown in Figures 10 and 11 of the disclosure. Details are not described herein. In one exemplary mode, the disclosure mode method may further include: storing intra-prognostic mode information used in the intra-prognostic process for use in the encoding and decoding of a neighboring block. For example, in an MPM mode, a reference to an intra-prognostic mode of a neighboring block is required. In other words, subsequent encoding / decoding blocks of the current image may use intra-prognostic modes of previously encoded / decoded blocks, such as neighboring blocks, based on the relationship of adjacent position. A chroma block (encoding unit) may use an intra-prognostic mode of a previously encoded / decoded luma block (encoding unit) based on a position relationship. The stored information is for reference by subsequent encoding / decoding blocks.Since coding mode information within the same block (coding unit) can be obtained directly, but coding mode information in different blocks (coding units) cannot, the intra-prognostic mode information used in intra-prognostic analysis needs to be stored. This way, the information can be read by subsequent coding / decoding blocks based on their position. In an exemplary modality, two different intra-prognostic modes are used to perform intra-prognostic analysis in the block to be decoded respectively, and the operation in which the intra-prognostic modes are stored in the intra-prognostic analysis may include the following operation. At least one minimal unit stores one of the two different intraprognostic modes, and at least one minimal unit stores the other of the two different intraprognostic modes. That is, at least two minimal units store the different intraprognostic modes. The minimal unit can be a predefined array of a fixed size (for example, a 4 x 4 array). Each minimal unit stores one intraprognostic mode. In this way, each time a block is encoded / decoded, the minimal units corresponding to the block's position can be used to store the intraprognostic modes for that block. For example, as shown in FIG. 12, if an intra-prognostic mode 5 is used to perform luma prognostication in a current 16x16 block, all corresponding 4x4 minimum units in the block store intra-prognostic mode 5. Taking the YUV format as an example, generally, the intra-prognostic luma mode is stored, which may include an intra-prognostic luma mode of a block that contains both a luma component and a chroma component, and an intra-prognostic luma mode of a block that contains only a luma component. As another example, in AVS3, disclosure modes can store two different intra-forecast modes using logic similar to that used by AWP to store two different movement information. That is, if a position corresponding to a minimum unit only uses a forecast block determined by one of the two intra-forecast modes, the minimum unit stores that intra-forecast mode. If a position corresponding to a minimum unit only uses a forecast block determined by the other of the two intra-forecast modes, the minimum unit stores that other intra-forecast mode.If a position corresponding to a minimum unit uses both a forecast block determined by the first intra-forecasting mode and a forecast block determined by the second intra-forecasting mode, the minimum unit can store one of the two intra-forecasting modes according to a predefined determination method. For example, for a 4x4 minimum unit, a point is selected, for example, point (2, 2), and if a weight of the first intra-forecasting mode at this point is greater than or equal to that of the second intra-forecasting mode, the first intra-forecasting mode is stored; otherwise, the second intra-forecasting mode is stored.As another example, a sum of the weights of the first intraprognostic mode and a sum of the weights of the second intraprognostic mode are calculated for all points in a block of the minimum unit. If the sum of the weights of the first intraprognostic mode is greater than or equal to the sum of the weights of the second intraprognostic mode, the first intraprognostic mode is stored; otherwise, the second intraprognostic mode is stored. The method of storing related information in GPM or AWP is used in the disclosure modalities. In this way, part of the same logic is reused. In an exemplary modality, two different intra-prognostic modes are used to perform intra-prognostic analysis in the block to be processed respectively, and the operation in which the intra-prognostic modes are stored in the intra-prognostic analysis may include the following operation. The same intra-prognostic mode is selected and stored for all the minimum units corresponding to the entire block to be processed. This reduces complexity. In an exemplary modality, the operation in which the same intra-prognostic mode is selected and stored for all the minimum units corresponding to the entire block to be processed may include the following operation. It is determined whether all the minimal units of the block to be processed store one of the two intra-prognostic modes or the other, based on the weight matrix derivation modes obtained by syntactically analyzing the bit stream. For example, all weight matrix derivation modes select the first intra-prognostic mode; as another example, all weight matrix derivation modes select the second intra-prognostic mode; as yet another example, all the minimal units corresponding to some weight matrix derivation modes select the first intra-prognostic mode, and all the minimal units corresponding to some other weight matrix derivation modes select the second intra-prognostic mode. The weight matrix derivation mode is a method for deriving weight matrices. For a block with a given width and height, each weight matrix derivation mode can be used to derive a weight matrix, and different weight matrix derivation modes can be used to derive different weight matrices for the same-sized block. For example, in AVS3, AWP has 56 weight matrix derivation modes; in WC, GPM has 64 weight matrix derivation modes. In an exemplary mode, the operation in which the same intra qq Lcnn / eznz / E / YiAi forecast mode is selected and stored for all the minimum units corresponding to the entire block to be processed may include the following operation. It is determined whether all the minimal units of the block to be processed store one of the two intra-prognostic modes or the other of the two intra-prognostic modes based on a mode number from the weight matrix derivation mode obtained by syntactically parsing the bitstream. In one modality, it can be determined whether all the minimal units of the block to be processed store the first intra-prognostic mode or the second intra-prognostic mode based on a mode number lookup box from the weight matrix derivation mode. For example, the disclosure modalities use the same weight matrix derivation modes as those of the AWP, as shown in Table 1.All minimum units corresponding to weight matrix derivation modes with mode number 0 in Table 1 can select the first intra-prognostic mode to store, and all minimum units corresponding to weight matrix derivation modes with mode number 1 in Table 1 can select the second intra-prognostic mode to store. TABLE 1 qq Lcnn / eznz / E / YiAi 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 0 1 1 1 0 1 1 1 0 1 1 1 0 1 In one modality, a coding method is provided, which includes the following operations. A target photograph is obtained and the block is divided in the target photograph to obtain a block that will be processed. Intra-prognostication is performed on the block to be processed using two or more different intra-prognostic modes, to obtain two or more prognostic blocks corresponding to the different intra-prognostic modes; and an objective prognostic block is obtained from the block to be processed based on weighting matrices and the two or more prognostic blocks obtained. The encoding is done based on the block to be processed and the target forecast block to generate a bit stream. In another modality, a decoding method is provided, which includes the following operations. A stream of bits is parsed to obtain a block to be processed and weighting matrices. The intra-prognostic process is carried out in the block to be processed by using two or more different intra-prognostic modes, to obtain two or more prognostic blocks corresponding to the different intra-prognostic modes. A target forecast block is obtained from the block to be processed based on the weighting matrices and the two or more forecast blocks obtained. Decoding is performed based on the target forecast block and the block to be processed, to obtain a reconstruction block corresponding to the block to be processed. One form of disclosure provides a computer-readable storage medium, which stores computer-executable instructions to implement the above intra-prognostic method or the decoding method applicable to the decoder end. One disclosure modality provides a computer-readable storage medium, which stores computer-executable instructions to implement the applicable intra-prognostic or coding method at the encoder end. One version of the disclosure provides a decoder, which includes a memory and a processor. The memory stores executable instructions for a processor to implement the operations of the intra-forecasting method or the aforementioned decoding method applicable at the decoder end. One disclosure modality provides an encoder, which includes a memory and a processor. The memory stores instructions executable by a processor to implement the operations of the intra-prognostic method or the aforementioned coding method applicable at the encoder end. Figure 13 is a schematic structural diagram of an intraprognostic device according to the disclosure. As shown in Figure 13, the device includes at least one prognostic module and one combination module. The forecasting module is configured to perform intra-forecasting in a block that will be processed by using two or more different intra-forecasting modes, to obtain two or more forecasting blocks corresponding to the different intra-forecasting modes. The combination module is configured to obtain a target forecast block from the block to be processed based on weighting matrices and the two or more forecast blocks obtained. The intra-prognostic device provided in this disclosure modality may be arranged in an encoder or a decoder. qq Lcnn / eznz / E / YiAi The intra-prognostic device provided in this disclosure modality may be arranged in the decoder, and may also include a decoding module. The decoding module is configured to decode a received bit stream to obtain the two or more different intra-forecast modes, the block to be processed, and the weighting matrices. The intra-prognostic device provided in this disclosure modality may be arranged in the encoder, and may also include a processing module. The processing module is configured to: try all or some possible combinations of forecasting modes and weight matrix derivation modes, calculate loss costs and select a combination with a low loss cost; use two or more different intra-forecasting modes and weight matrices in the combination as the two or more different intra-forecasting modes and weight matrices for the intra-forecast; and write to a bit stream according to the syntax, information about the two or more different intra-forecasting modes and weight matrix derivation modes determined. In an exemplary modality, the two or more different modes of intra-prognostic include at least one basic mode of intra-prognostic. In one exemplary modality, two different intra-prognostic modes are used to perform intra-prognostic analysis on the block to be processed, and the two different intra-prognostic modes are both basic intra-prognostic modes. In an exemplary modality, the basic intra-prognostic mode used can be further combined with an enhanced intra-prognostic mode to predict the block to be processed. In one exemplary modality, two different intra-prognostic modes are used to forecast the block to be processed, and the two different intra-prognostic modes include a basic intra-prognostic mode and an enhanced intra-prognostic mode. In the dissemination methods, at least two different intra-prognostic modes are used for the forecast of the block to be processed, so that the forecast of the block to be processed can be made from multiple perspectives, which is suitable for the forecast of complex textures and facilitates improving the quality of the intra-prognostic. In one exemplary instance, not every weighting matrix in all possible weighting matrices has the same weight at all points. In other words, at least one of all possible weighting matrices includes at least two different weights. In an exemplary modality, all possible weighting matrices include at least two different weights. In an exemplary mode, at least one of all possible weighting matrices qq Lcnn / eznz / E / YiAi includes at least two different weights, and at least one of all possible weighting matrices includes only the same weights. In one exemplary mode, intra-forecasting is performed in the block to be processed by using two different intra-forecasting modes, and only one of all possible weighting matrices includes only two types of weights, one type of weights indicates that a forecast value of a corresponding point is derived exclusively from a value of the corresponding point in the first forecast block, and the other type of weights indicates that a forecast value of a corresponding point is derived exclusively from a value of the corresponding point in the second forecast block. In one exemplary mode, intra-forecasting is performed on the block to be processed using two different intra-forecasting modes, and one of all possible weighting matrices includes a plurality of weight types. A maximum weight and a minimum weight (e.g., 0) respectively indicate that forecast values ​​for corresponding points are derived exclusively from values ​​of corresponding points in the first forecast block and values ​​of corresponding points in the second forecast block; and a weight other than the maximum or minimum weight indicates that a forecast value for a corresponding point is derived from a weighted average of values ​​of the corresponding point in the first forecast block and the corresponding point in the second forecast block. An area consisting of weights other than the maximum or minimum value can be referred to as a combination area. In one exemplary mode, in all possible weighting matrices, if a weighting matrix includes only two types of weights, the positions where the weights change form a straight line; if a weighting matrix includes multiple types of weights, the positions with the same weights in the combination area form a straight line. In one mode, the aforementioned straight lines are all horizontal-vertical, or the aforementioned straight lines are not all horizontal-vertical. In an exemplary mode, in all possible weighting matrices, if a weighting matrix includes only two types of weights, the positions where the weights change form a curved line; if a weighting matrix includes multiple types of weights, the positions with the same weights in the combination area form a curved line. The various weighting matrices provided in the disclosure modalities ensure the forecasting of more diverse forecast blocks and enable the intra-forecasting method provided in the disclosure modalities to be applicable to more scenarios. In an exemplary modality, in the intra-prognostic device provided in this modality of disclosure, the block size may satisfy, but is not limited to, the following conditions. A block width is greater than or equal to a first threshold TH1, a block height is greater than or equal to a second threshold TH2, and the first threshold TH1 and the second threshold TH2 can be 8, 16, 32, 64, 128, or similar. The first threshold TH1 can be equal to the second threshold TH2, for example, first threshold TH1 = second threshold TH2 = 8. Alternatively, a block pixel count is greater than or equal to a third threshold TH3, and the third threshold TH3 can be 8, 16, 32, 64, 128, or similar. Alternatively, the block width is less than or equal to a fourth TH4 threshold, the block height is less than or equal to a fifth TH5 threshold, and the fourth TH4 threshold and the fifth TH5 threshold can be 8, 16, 32, 64, 128, or similar. The fourth TH4 threshold can be equal to the fifth TH5 threshold. Alternatively, a number of pixels in the block is less than or equal to the sixth TH6 threshold, and the sixth TH6 threshold can be 8, 16, 32, 64, 128, or similar. By limiting the block size, the impact of complexity due to the use of multiple forecasting modes for the forecast in the total system is reduced, and the trade-off between compression performance and complexity is well achieved by discarding some blocks of inapplicable sizes, thus better ensuring the applicability of the disclosure. In one exemplary mode, block splitting becomes increasingly flexible as technology evolves. Besides square blocks, the splitting method can also support blocks with aspect ratios such as 1:2, 1:4, 1:8, 2:1, 4:1, 8:1, and similar ratios. In the intra-prognostic method provided in the disclosure modes, the inventors of the disclosure found that blocks with certain aspect ratios, or blocks with certain sizes, such as 1:4 or 4:1, 1:8 or 8:1, 8x32, 8x64, 32x8, or 64x8, may not provide good or significant compression performance. In one disclosure mode, the block size can be adjusted by configuring the aspect ratio. For example, the width-to-height ratio is less than or equal to a predefined THR threshold, and the width-to-height ratio is less than or equal to the THR threshold. In one example, the block size and block aspect ratio can be set simultaneously. For instance, the intra-prognostic method provided in the disclosure modalities can be used if the block size satisfies the following criteria: block height is greater than or equal to 8, block width is greater than or equal to 8, block width-to-height ratio is less than or equal to 4, and block height-to-width ratio is less than or equal to 4; otherwise, the intra-prognostic method provided in the disclosure modalities is not used by default. In an exemplary mode, two different intra-forecasting modes are used to perform intra-forecasting on the block to be decoded, and the combination module can be qq Lcnn / eznz / E / YiAi configured specifically to: calculate a first product of a matrix corresponding to the first forecast block and the first weighting matrix and a second product of a matrix corresponding to the second forecast block and the second weighting matrix; calculate a sum of the first product, the second product and a predefined value; and normalize the calculated sum to obtain the target forecast block. In an exemplary mode, the intra-prognostic device provided in this disclosure mode can be arranged in the encoder, and the processing module can be further configured to: set a flag at the photograph level to indicate whether to use the intra-prognostic method in the disclosure modes for the current photograph to be processed. The decoding module in the decoder can also be configured to: determine, based on a flag, whether to proceed with the intra-prognostic analysis in the block to be processed by using the two or more different intra-prognostic modes obtained through decoding. In an exemplary modality, the intra-prognostic device provided in this disclosure modality may be disposed of in the encoder, and the processing module may be further configured to: set a flag below a photograph level and above a CU level (e.g., mosaic, slice, patch, or LCU), to indicate whether to use the intra-prognostic method in the disclosure modalities for a specified area. Accordingly, the decoding module in the decoder can be further configured to: determine, based on a flag, whether to proceed with the intra-prognostic analysis in the block to be processed by using the two or more different intra-prognostic modes obtained through decoding. By configuring different levels of flags in the disclosure modes, on the one hand, the optimal configuration is flexibly carried out according to different application scenarios and video content, and on the other hand, the data flow per second is further reduced. In an exemplary modality, the intra-prognostic device provided in this disclosure modality may be arranged in the encoder, and the processing module may be further configured to establish an enhanced prognostic mode i.e. mutually exclusive with the intra-prognostic method provided in the disclosure modalities. Correspondingly, the decoding module in the decoder may be further configured to: parse the bit stream to obtain the mutually exclusive enhanced forecast mode with the intra-forecast method provided in the disclosure modalities; and if it is determined that the intra-forecast method provided in qq Lcnn / rznz / E / YiAi disclosure modalities is used for the block to be processed, the mutually exclusive enhanced forecast mode is not used; or, if it is determined that the mutually exclusive enhanced forecast mode is used for the block to be processed, the intra-forecast method of the disclosure modalities is not used.By configuring the mutually exclusive enhanced forecasting mode, it is not necessary to transmit a flag in the bitstream to indicate whether the mutually exclusive enhanced forecasting mode is being used, thus avoiding unnecessary transmission of the flag in the bitstream and generally achieving better compression performance. In the modalities, by configuring the mutually exclusive relationship between these improved forecasting modes and the intra-forecast of the disclosure, the exchange between performance and complexity is well achieved, thus better ensuring the applicability of the disclosure. In an exemplary mode, the combination module can also be configured to store intra-prognostic mode information used in intra-prognostic processing, for use in a neighboring block's encoding process. In an exemplary modality, two different intra-prognostic modes are used to perform intra-prognostic analysis in the block to be processed, and the operation in which the intra-prognostic modes are stored in the intra-prognostic analysis may include the following operation. At least one minimal unit stores one of the two different intraprognostic modes, and at least one minimal unit stores the other of the two different intraprognostic modes. That is, at least two minimal units store different intraprognostic modes. In an exemplary mode, two different intra-prognostic modes are used to perform intra-prognostic analysis on the block to be processed, and the intra-prognostic modes used in intra-prognostic analysis that are stored may include the following operation. The same intra-prognostic mode is selected and stored for all the minimum units corresponding to the entire block to be processed. The decoder provided in the disclosure modes performs intra-prognostication on the block to be processed by using two or more different intra-prognostic modes to obtain two or more prognostic blocks. It then combines these two or more prognostic blocks based on weighting matrices to obtain the final prognostic block for the block to be processed. The disclosure modes use multiple intra-prognostic modes to determine multiple prognostic blocks, thereby achieving prognostication of complex textures, improving intra-prognostic quality, and thus enhancing compression performance. qq Lcnn / rznz / E / YiAi Furthermore, the decoder provided in the disclosure methods ensures the prediction of more complex textures using various weighting matrices. This improves the quality of the intra-prognosis and increases compression performance. Additionally, the intra-prognosis method provided in the disclosure methods is applicable to a wider range of scenarios. One version of the disclosure also provides a decoder, which includes: a decoding module, a forecasting module, and a combination module. The decoding module is configured to decode a received bit stream, to obtain two or more different intra-forecast modes, a block to be processed, and weighting matrices. The forecasting module is configured to perform intra-forecasting in the block to be processed by using two or more different intra-forecasting modes, to obtain two or more forecasting blocks corresponding to the different intra-forecasting modes. The combination module is configured to obtain a target forecast block from the block to be processed based on the weighting matrices and the two or more forecast blocks obtained. In an exemplary mode, the decoding module can also be configured to determine, based on a flag at the photograph level, whether to proceed with the intra-prognostic analysis in the block to be processed by using the two or more different intra-prognostic modes obtained through decoding. In an exemplary mode, the decoding module can be further configured to determine, based on a flag below a photography level and above a CU level, whether to proceed with the intra-prognostic realization in the block to be processed by using the two or more different intra-prognostic modes obtained through decoding. One version of the disclosure also provides an encoder, which includes: a forecasting module, a combination module, and a processing module. The forecasting module can be configured to perform intra-forecasting on a block to be processed by using two or more different intra-forecasting modes, to obtain two or more forecasting blocks corresponding to the different intra-forecasting modes. The combination module can be configured to obtain a target forecast block from the block to be processed based on weighting matrices and the two or more forecast blocks obtained. The processing module can be configured to: test all or some of the possible combinations of forecasting modes and weight matrix derivation modes, calculate loss costs and select a combination with a low loss cost; use two or more different intra-forecasting modes and weight matrices in combination as the two or more different intra-forecasting modes and weight matrices for the intra-forecast; and write the two or more different determined intra-forecasting modes, weight matrix derivation modes and other information into a bit stream according to the syntax. In one exemplary mode, the processing module can also be configured to set a flag. The flag may be a photograph-level flag to indicate to a decoder whether to proceed with obtaining the two or more forecast blocks corresponding to the different intra-forecast modes; and / or, the flag may be a flag below a photograph level and above a CU level, to indicate to a decoder whether to proceed with obtaining the two or more forecast blocks corresponding to the different intra-forecast modes for a specified area. In an exemplary mode, the processing module can also be configured to: establish a mutually exclusive forecasting mode with the intra-forecast; and avoid using the mutually exclusive forecasting mode if the intra-forecast is performed on the block to be processed; or avoid performing the intra-forecast if the mutually exclusive forecasting mode is used for the block to be processed. In an exemplary mode, the combination module can also be configured to store intra-prognostic mode information used in intra-prognostic testing. Figure 14 is a schematic flowchart of another intra-prognostic method according to one disclosure modality. As shown in Figure 14, the method includes the following operations. In 1400, intra-prognostic testing is performed in a block that will be processed using two or more different intra-prognostic modes, respectively. In 1401, during forecasting using each intra-forecasting mode, when a predefined number of pixels is forecasted, a predefined number of forecast pixels is obtained from the block to be processed based on weighting matrices and the pixels corresponding to the intra-forecasting mode that has been forecasted. In 1402, a target forecast block of the block to be processed is obtained based on the predefined number of forecast pixels. The difference between the intra-prognostic method provided in this mode and the mode shown in FIG. 7 is that the processed target in the mode shown in FIG. 7 is a block, while the processed target in the mode shown in FIG. 14 is a pixel. Similarly, in the mode shown in FIG. 14, intra-prognostication is performed on the block to be processed using two or more different intra-prognostic modes. When the predefined number of pixels is predicted, the pixels corresponding to each predicted intra-prognostic mode are combined based on weighting matrices to obtain the predefined number of prediction pixels for the block to be processed. Finally, the resulting predefined number of prediction pixels is combined to obtain the prediction block for the block to be processed.In the dissemination methods, multiple forecast blocks are determined using multiple intra-forecasting modes, thus achieving the forecasting of complex textures. This improves the quality of the intra-forecasting and increases compression performance. Furthermore, the specific implementation of the weighting matrices in the mode shown in FIG. 14 is the same as that described in the mode shown in FIG. 7. The various weighting matrices ensure the prediction of more complex textures. In this way, the quality of the intra-prognostication is improved, and compression performance is increased. Moreover, the intra-prognostication method provided in the disclosure modes is applicable to a wider range of scenarios. The following describes a decoding method that uses the intra-forecasting method provided in the disclosure methods as applied in AVS3. Since AVS3 uses AWP technology, the disclosure intra-forecasting method is referred to as Spatial Angle Weighted Forecasting (SAWP) in this method. Some terms from the AVS3 standard are used in this method. For example, the forecast sample matrix in this method is the forecast block; that is, "block" can be understood as the sample matrix. As another example, the array in this method is the matrix. This method is implemented by applying SAWP to the luma component as an example, but the disclosure methods are not limited to the luma component and can also be used for the chroma component and any other component in any format. In this mode, one encoder end can set a sequence-level flag to determine whether the SAWP is performed on a current sequence that will be decoded at the decoder end. A sequence_header is defined as in Table 2. qq Lcnn / eznz / E / YiAi TABLE 2 Definition of Sequence_header Descriptor sequence_header() { ...... sawp_enable_flag u(l) ...... In Table 2, sawp_enable_flag is a flag that enables SAWP, which is a binary variable. For example, a value of 1 indicates that SAWP can be performed; a value of 0 indicates that SAWP cannot be performed. In this mode, the encoder end can set a flag at the photo level to determine whether the SAWP is performed on a current photo that will be decoded at the decoder end. For example, the SAWP can be configured to be used for intra-photos (e.g., photo I) but not for inter-photos (e.g., photo B and photo P). Alternatively, the SAWP can be configured to be used for intra-photos but not for inter-photos. Another example is that the SAWP can be configured to be used for some inter-photos but not for others. In this mode, the encoder end can set a flag below a photo level and above a CU level (e.g., tile, cut, patch, or LCU) to determine if SAWP is used in an area at the decoder end. It should be noted that the flags mentioned above may not be set in this mode. The decoder decodes the current CU and the flag that enables SAWP from the current CU if the current CU uses intraprognostic; otherwise, the decoder does not need to decode the flag that enables SAWP from the current CU. Since the information related to DT and IPF is mutually exclusive with SAWP, if the current CU uses SAWP, there is no need to process the information related to DT and IPF. TABLE 3 if (! DirectFIag &&(mode == 'PRED_No_Constraint'))____________________________________________ intra_cu_flag______________________________________________________________________________ J_________________________________________________ J_________________________________________________ PartSize = 'SIZE_2Mx2N'_______________________________________________________________________ If(SawpEnableFlag&& IntraCuFlag&&width>=SawpMinSize&& height>=SawpMinSize&&width*SawpMaxRatio>=height&&height*SawpMaxRatio>=width){______ sawpflag_____________________________________________________________________ J_________________________________________________ if (DtEnableFlag && IntraCuFlag&&!SawpFlag) {____________________________________________________ dtsplitflag qq Lcnn / eznz / E / YiAi SawpMinSize is a minimum length and width value, and SawpMaxRatio is a maximum length-to-width ratio. In Table 3, sawp_flag indicates a SAWP flag, i.e., a binary variable. For example, a value of 1 indicates that the SAWP is performed, and a value of 0 indicates that the SAWP is not performed. The value of SawpFlag is equal to the value of sawp_flag. If there is no sawp_flag in the bit stream, the value of SawpFlag is 0. The value of SawpEnableFlag in Table 3 is equal to the value of sawp_enable_flag in Table 2. If there is no sawp_enable_flag in the bit stream, the value of SawpEnableFlag is 0. In this modality, assuming SAWP is used for current UC, the weight matrix derivation modes and two intra-prognostic modes need to be decoded (two intra-prognostic modes are used as an example in this modality). In this modality, for example, the weight matrix derivation mode reuses the weight matrix derivation mode of AWP, and the decoding of the two intra-prognostic modes of SAWP reuses the decoding of the intra-prognostic modes in the related technique. qq Lcnn / eznz / E / YiAi In Table 4, sawpjdx indicates an index for the SAWP mode, which is used to determine a weight matrix for SAWP, and a value of Sawpldx is equal to a value of sawpjdx. If there is no sawpjdx in the bit stream, the value of Sawpldx is 0. In Table 4, intrajuma_pred_mode0 indicates a first luma forecasting mode of the SAWP, which is used to determine the first intra-prognostic mode for the luma block of the SAWP; intraJuma_pred_model indicates a second luma forecasting mode of the SAWP, which is used to determine the second intra-prognostic mode for the luma block of the SAWP. In one modality, the method of parsing syntactically for sawpjdx may be the same as for awpjdx in the related technique, the method of parsing syntactically for intrajuma_pred_mode0 may be the same as that for intrajuma_pred_mode in the related technique, and the method of parsing syntactically for intrajuma_pred_model may be the same as that for intrajuma_pred_mode in the related technique. In one modality, the method of parsing syntactically for intrajuma_pred_model may also include the following. If both intrajuma_pred_mode0 and intrajuma_pred_model use the MPM, intrajuma_pred_model does not need to be decoded to determine whether it is the first or second intra-prognostic mode of the MPM. That is, the second intra-prognostic mode is determined based on the information about the first decoded intra-prognostic mode. Since the MPM in AVS3 has only two intra-prognostic modes, if intrajuma_pred_mode0 uses one of the intra-prognostic modes, intrajuma_pred_model uses the other intra-prognostic mode by default. The binarization method for intra_luma_pred_modeO is shown in Table 5. The value of 0 or 1 for intra_luma_pred_modeO indicates whether MPM is used. Specifically, if the first symbol in the binary symbol string is 1, it indicates that MPM is used, and if the first binary symbol is 0, it indicates that MPM is not used. If the first binary symbol indicates MPM, the second binary symbol in the binary symbol string indicates that MPM is used. qq Lcnn / eznz / E / YiAi TABLE 5 Intra_luma_pred_modeO Binary symbol string 0 10 1 11 2 0...... ...... ...... The binarization method for intra_luma_pred_model is shown in Table 6. The value of intra_luma_pred_model (0 or 1) indicates whether MPM is used. Specifically, if the first binary symbol in the binary symbol string is 1, the second binary symbol is not needed. If the value of intra_luma_pred_modeO is 1, the value of intra_luma_pred_model is 0. If the value of intra_luma_pred_modeO is 0, the value of intra_luma_pred_model is 1. TABLE 6 Intraju ma_pred_mode 1 Binary symbol string 0 1 (and intra luma pred modeO is 1) 1 1 (and intra luma pred modeO is 0) 2 0...... ...... ...... Another modality for syntactically analyzing the bitstream structure and decoding is described by taking the intra-prognostic method provided in the disclosure modalities that is applied in AVS3 as an example. A decoder decodes the current CU, and if the current CU uses intra prognostic, it decodes the flag that enables DT and IPF of the current CU, as well as the unique prognostic mode intra_luma_pred_mode for each prognostic unit in the current intra prognostic method. If the current CU does not use DT or IPF, the flag that enables SAWP in the current CU is decoded. If the current CU uses SAWP, the weight matrix derivation mode and an intra-prognostic mode, intra_luma_pred_model, are decoded, and the decoded intra_luma_pred_mode is used as intra_luma_pred_mode0. TABLE 7 dt split flag intra luma pred mode ipf flag If(SawpEnableFlag&&! DtSplitFIag &&!IpfFlag &&IntraCuFlag&&width>=SawpMinSize&& height>=SawpM¡nS¡ze&&w¡dth*SawpMaxRat¡o>=he¡ght&&he¡ght*SawpMaxRat¡o>=w¡dt hK_______________________________________________________________ sawpflag_________________________________________________________________}____________________________________________________________________________________________________ if ((SawpFlag) {__________________________________________________________________________ sawpidx__________________________________________________ intra_luma_pred_model________________________________________} qq Lcnn / eznz / E / YiAi IntraLumaPredModeO and IntraLumaPredModel are determined based on intra_luma_pred_modeO and intra_luma_pred_model, respectively, in addition to determining the intra-predictive sample matrices predMatrixO and predMatrixl. The SawpWeightArrayY weighting matrix is ​​determined based on Sawpidx, and an input index is Sawpidx. A new forecast sample matrix, predMatrixSawp, is determined based on the two intra-forecast sample matrices, predMatrixO and predMatrixl, and the determined weighting matrix, SawpWeightArrayY. The following operation is included. The value of the predMatrixSawp[x][y] element in the predMatrixSawp forecast sample matrix for SAWP mode is ((predMatrixO[x][y] * AwpWeightArrayY[x][y] + predMatrixl[x][y] * ( 8 - AwpWeightArrayY[x][y] ) + 4 ) >> 3). After the SAWP forecast block—that is, after the sample forecast matrix predMatrixSawp is determined for SAWP mode—subsequent processing may further include: decoding, inverse transformation, and inverse quantization of the quantization coefficients to determine a residual block; combining the residual block and the forecast block into a reconstruction block; and subsequent loop filtering. The specific implementation is not used to limit the scope of disclosure protection, and the details are not repeated here. The method for storing SAWP's intra-predictive mode in this mode may be similar to the method for storing AWP's motion information, but the input index is replaced with Sawpidx and the output intra-predictive reference mode (interPredAwpRefMode) is replaced with sawpRefMode. If the sawpRefMode of a 4x4 block is 0, IntraLumaPredModeO is stored; otherwise, if the sawpRefMode of the 4x4 block is 1, IntraLumaPredModel is stored. Since the first version of AVS3 only supports 34 intra-prognostic modes, the 34th mode (which has an index number of 33 if the index starts at 0) is the PCM mode. More intra-prognostic modes are added in the second version of AVS3, expanding to 66. To be compatible with the first version, the second version does not change the original decoding method for intra_luma_pred_mode, but proposes that if intra_luma_pred_mode is greater than 1, another flag needs to be added, as shown in Table 8; that is, the intra-prognostic mode expansion flag luma eipm_pu_flag. TABLE 8 intra_luma_pred_mode______________________________________________ if (EipmEnableFlag && intra_luma_pred_mode > 1) {_____________________ eipm_pu_flag___________________________________________________________ J___________________________________ The luma intra-prognostic mode expansion flag, eipm_pu_flag, is a binary variable. A value of 1 indicates that the angular intra-prognostic mode should be used. A value of 0 indicates that the luma intra-prognostic mode should not be used. The value of EipmPuFlag is equal to the value of eipm_pu_flag. If there is no eipm_pu_flag in the bit stream, the value of EipmPuFlag is 0. Therefore, with reference to the description of the text of the second version of AVS3, the syntax of all of intra_luma_pred_mode, intra_luma_pred_modeO, and intra_luma_pred_model in the above modalities must be followed by eipm_pu_flag, eipm_pu_flag0, and eipm_pu_flagl as in Table 8. Furthermore, IntraLumaPredModeO is determined based on intra_luma_pred_modeO and eipm_pu_flagO, and IntraLumaPredModel is determined based on intraLuma_pred_model and eipm_pu_flagl. In disclosure modes, modes that do not use PCM modes, or modes that only use the first 33 or 34 modes of the first version, there is no need to transmit eipm_pu_flag. Those skilled in the art should know that the modules or operations of the disclosure can be implemented by means of a universal computing device, and these modules or operations can be concentrated on a single computing device or distributed over a network consisting of a plurality of computing devices, and can optionally be implemented by programmable code executable by the computing devices, so that these modules or operations can be stored on a storage device for execution with the computing devices, and in some circumstances, the operations shown or described can be performed in sequences different from those described here, or the modules or operations can be made on a single integrated circuit module. Therefore, the disclosure is not limited to any specific combination of hardware and software. The foregoing are merely preferred forms of disclosure and are not intended to limit disclosure, and various changes and modifications to the disclosure may be made by those skilled in the art. Any modifications, equivalent replacements, improvements, and the like made within the spirit and principle of the disclosure will fall within the scope of protection of the disclosure.

Claims

1. An intra-prognostic method, applicable to a decoder, comprising: performing intra-prognostication on a block to be processed by using two or more different intra-prognostic modes, to obtain two or more prognostic blocks corresponding to the different intra-prognostic modes; and obtaining a target prognostic block from the block to be processed based on weighting matrices and the two or more prognostic blocks obtained.

2. The intra-prognostic method according to claim 1, further characterized in that, prior to obtaining the two or more prognostic blocks corresponding to the different intra-prognostic modes, the method further comprises: syntactically analyzing a bit stream to obtain the two or more different intra-prognostic modes, the block to be processed, and the weighting matrices.

3. The intra-prognostic method according to claim 1, further characterized in that, prior to obtaining the two or more prognostic blocks corresponding to the different intra-prognostic modes, the method further comprises: syntactically parsing a bit stream to obtain a flag, wherein the flag is a photograph-level flag to indicate whether to proceed with obtaining the two or more prognostic blocks corresponding to the different intra-prognostic modes; and / or the flag is a flag below a photograph level and above a coding unit (CU) level and used to indicate whether to proceed with obtaining the two or more prognostic blocks corresponding to the different intra-prognostic modes for a specified area.

4. The intra-prognostic method according to any of claims 1 to 3, further characterized in that the two or more different intra-prognostic modes comprise: first intra-prognostic modes, or a first intra-prognostic mode and a second intra-prognostic mode; and the first intra-prognostic mode is an intra-prognostic mode that generates a prognostic block independently of other intra-prognostic modes, and the second intra-prognostic mode is an intra-prognostic mode that depends on a basic intra-prognostic mode to determine a prognostic block.

5. The intra-prognostic method according to claim 4, further characterized in that the two or more different intra-prediction modes comprise at least one of the first intra-prognostic modes.

6. The intra-prognostic method according to claim 5, further characterized in that the intra-prognostic modes comprise two different intra-prognostic modes; and the two different intra-prognostic modes are both the first intra-prognostic mode.

7. The intra-prognostic method according to claim 5, further characterized in that the intra-prognostic modes comprise two different intra-prognostic modes; and the two different intra-prognostic modes comprise a first intra-prognostic mode and a second intra-prognostic mode.

8. The intra-prognostic method according to claim 5, further characterized in that the first intra-prognostic modes comprise: a direct current (DC) mode, a planar mode, a flat mode, a bilinear mode, and an angular prognostic mode.

9. The intra-prognostic method according to claim 5, further characterized in that the second intra-prognostic mode comprises: enhancing a subpixel interpolation of a reference pixel, and filtering a prognostic pixel.

10. The intra-prognostic method according to claim 6, further characterized in that it additionally comprises: performing the intra-prognostic in the target prognostic block by using the second intra-prognostic mode, and using a prognostic outcome as the target prognostic block.

11. The intra-forecasting method according to any of claims 1 to 3, further characterized in that the intra-forecasting modes comprise two different intra-forecasting modes, the forecast blocks comprise a first forecast block and a second forecast block, and the weighting matrices comprise a first weighting matrix and a second weighting matrix; and obtaining the target forecast block from the block to be processed based on the weighting matrices and the two or more forecast blocks obtained comprises: calculating a first product of the first forecast block and the first weighting matrix and calculating a second product of the second forecast block and the second weighting matrix; calculating a sum of the first product, the second product and a predefined value; and obtaining the target forecast block based on the calculated sum.

12. The intra-forecasting method according to claim 11, QQI Í'nn / Q7n7 / B / YIAI, further characterized in that the second weighting matrix is ​​a difference between a maximum weight and the first weighting matrix; and the method further comprises: right-shifting the calculated sum by a predefined number of bits, to obtain a forecast block from a block to be processed.

13. The intra-prognostic method according to claim 12, further characterized in that the maximum weight is 8, the predefined value is 4, and the predefined number of bits is 3.

14. The intra-prognostic method according to claim 11, further characterized in that at least one of all possible weighting matrices comprises at least two different weights.

15. The intra-prognostic method according to claim 14, further characterized in that at least one of all possible weighting matrices comprises at least two different weights, and at least one of all possible weighting matrices comprises only the same weights.

16. The intra-prognostic method according to claim 14, further characterized in that for at least one weighting matrix comprising at least two different weights, each point in the block to be processed is weighted by prognostic values ​​derived from two intra-prognostic modes based on a minimum weight and a maximum weight.

17. The intra-forecasting method according to claim 14, further characterized in that only one of all possible weighting matrices comprises only two types of weights, wherein one type of weight indicates that a forecast value of a corresponding point is derived exclusively from a value of the corresponding point in the first forecast block, and the other type of weight indicates that a forecast value of a corresponding point is derived exclusively from a value of the corresponding point in the second forecast block.

18. The intra-forecasting method according to claim 14, further characterized in that one of all possible weighting matrices comprises a plurality of weight types, wherein a maximum weight and a minimum weight respectively indicate that the forecast values ​​of the corresponding points are derived exclusively from a value of the corresponding point in the first forecast block and a value of the corresponding point in the second forecast block; and a weight other than the maximum and minimum weights indicates that a forecast value of a corresponding point is derived from a weighted average of values ​​of the corresponding point in the first forecast block and the second forecast block.

19. The intra-prognostic method according to claim 14, further characterized in that among all possible weighting matrices, when a weighting matrix comprises only two types of weights, the positions in which the weights change form a straight line; and when a weighting matrix comprises a plurality of types of weights, the positions in a combination area in which the weights are equal form a straight line.

20. The intra-prognostic method according to claim 14, further characterized in that among all possible weighting matrices, when a weighting matrix comprises only two types of weights, the positions at which the weights change form a curved line; and when a weighting matrix comprises a plurality of types of weights, the positions in a combination area at which the weights are equal form a curved line.

21. The intra-prognostic method according to claim 1, further characterized in that a width and a height of the block to be processed or the target prognostic block comprises: the width being greater than or equal to a first threshold TH1 and the height being greater than or equal to a second threshold TH2; or the width being less than or equal to a fourth threshold TH4 and the height being less than or equal to a fifth threshold TH5.

22. The intra-prognostic method according to claim 21, further characterized in that the first TH1 threshold is equal to the second TH2 threshold, and the first TH1 threshold is 8.

23. The intra-prognostic method according to claim 1 or 21, further characterized in that a width and a height of the block to be processed or the target prognostic block comprises: a width-to-height ratio that is less than or equal to a predefined ratio threshold THR, and a height-to-width ratio that is less than or equal to the ratio threshold THR.

24. The intra-prognostic method according to claim 23, further characterized in that the THR ratio threshold is 4.

25. The intra-prognostic method according to any of claims 1 to 3, further characterized in that it additionally comprises: storing information from the intra-prognostic mode used in the intra-prognostic method.

26. The intra-prognostic method according to claim 25, further characterized in that the intra-prognostic modes comprise two different intra-prognostic modes; and storing the intra-prognostic modes used in the intra-prognostic method comprises: storing different intra-prognostic modes in at least two minimum units.

27. The intra-prognostic method according to claim 25, further characterized in that storing the intra-prognostic modes used in the intra-prognostic method comprises: selecting the same intra-prognostic mode to be stored in all the minimum units corresponding to the entire block to be processed.

28. The intra-prognostic method according to claim 27, further characterized in that the intra-prognostic modes comprise two different intra-prognostic modes; and selecting the same intra-prognostic mode to be stored in all the minimal units corresponding to the entire block to be decoded comprises: determining, based on the weight matrix derivation modes obtained through bitstream syntactic analysis, whether all of the minimal units of the block to be processed store one of the two intra-prognostic modes, or store the other of the two intra-prognostic modes; or determining, based on the mode numbers of the weight matrix derivation modes, whether all of the minimal units of the block to be processed store one of the two intra-prognostic modes, or store the other of the two intra-prognostic modes.

29. The intra-prognostic method according to claim 2, further characterized in that the intra-prognostic modes comprise two different intra-prognostic modes; and one of the two intra-prognostic modes is determined based on information about the other decoded intra-prognostic mode.

30. The intra-prognostic method according to claim 29, further characterized in that the first intra-prognostic mode and the second intra-prognostic mode are both the most probable modes (MPMs).

31. A decoder, comprising: a decoding module, configured to decode a received bit stream, to obtain two or more different intra-forecasting modes, a block to be processed, and weighting matrices; a forecasting module, configured to perform intra-forecasting on the block to be processed by using the two or more different intra-forecasting modes, to obtain two or more forecasting blocks corresponding to the different intra-forecasting modes; and a combination module, configured to obtain a target forecasting block from the block to be processed based on the weighting matrices and the two or more forecasting blocks obtained.

32. The decoder according to claim 31, further characterized in that the decoding module is additionally configured to: determine, based on a photograph-level flag, whether to proceed with intra-prognostic realization in the block to be processed by using the two or more different intra-prognostic modes obtained through decoding.

33. The decoder according to claim 31, further characterized in that the decoding module is additionally configured to determine, based on a flag below a photograph level and above the level of an encoding unit (CU), whether to proceed with the realization of intra-prognostic analysis in the block to be processed by using the two or more different intra-prognostic analysis modes obtained through decoding.

34. The decoder according to claim 31, further characterized in that it additionally comprises a processing module, configured to: establish a mutually exclusive forecasting mode with the intra-forecast; and avoid using the mutually exclusive forecasting mode when the intra-forecast is performed on the block to be processed; or avoid performing the intra-forecast when the mutually exclusive forecasting mode is used for the block to be processed.

35. An intra-prognostic method, applicable to a coder and comprising: performing intra-prognostication on a block to be processed by using two or more different intra-prognostic modes, to obtain two or more prognostic blocks corresponding to the different intra-prognostic modes; and obtaining a target prognostic block from the block to be processed based on weighting matrices and the two or more prognostic blocks obtained.

36. The intra-prognostic method according to claim 35, further characterized in that, prior to obtaining the two or more prognostic blocks corresponding to the different intra-prognostic modes, the method further comprises: attempting all or some of the possible combinations of prognostic modes and weight matrix derivation modes, calculating the loss costs of the combinations, and selecting a combination with a low loss cost; and using two or more different intra-prognostic modes and weight matrices in the combination as the two or more different intra-prognostic modes and weight matrices qq Lcnn / eznz / E / YiAi for the intra-prognostic.

37. The intra-prognostic method according to claim 36, further characterized in that prior to the attempt, the method further comprises: analyzing a texture of the block to be processed; and the attempt further comprises: determining the intra-prognostic modes to be attempted based on a texture analysis result.

38. The intra-forecasting method according to claim 36, further characterized in that, after the attempt, the method further comprises: in the condition that a loss cost of the selected combination is less than or equal to the costs of other forecasting modes, determining the intra-forecasting modes in the selected combination as the forecasting modes for the block to be processed; or in the condition that the loss cost of the selected combination is greater than the costs of other forecasting modes, selecting part of the other forecasting modes as the forecasting modes for the block to be processed.

39. The intra-forecast method according to any of claims 36 to 38, further characterized in that a method for calculating loss costs comprises any or any combination of the following: sum of absolute differences (SAD), sum of absolute transformed differences (SATD), or velocity distortion optimization (RDO).

40. The intra-forecasting method according to any of claims 36 to 38, further characterized in that selecting the combination with the lowest loss cost comprises: performing a first selection operation with the SATD and / or the SAD, to determine candidate combinations in all or part of the possible combinations of forecasting modes and weighting matrix derivation modes; and performing a second selection operation with the RDO, to determine a combination with a minimum loss cost in the candidate combinations.

41. The intra-prognostic method according to claim 36, further characterized in that when an angular intra-prognostic mode causes a high cost, the angular intra-prognostic mode and a predefined number of intra-prognostic modes adjacent to the angular intra-prognostic mode are removed from the candidate combinations.

42. The intra-prognostic method according to claim 35, further characterized in that, prior to obtaining the two or more prognostic blocks corresponding to the different intra-prognostic modes, the method further comprises setting a flag, wherein the flag is a photograph-level flag to indicate to a decoder whether to proceed with obtaining the two or more prognostic blocks corresponding to the different intra-prognostic modes; and / or, the flag is a flag below a photograph level and above the level of an encoding unit (CU) and used to indicate to the decoder whether to proceed with obtaining the two or more prognostic blocks corresponding to the different intra-prognostic modes in a specified area; and writing the flag to the bit stream.

43. The intra-prognostic method according to claim 35, further characterized in that, before obtaining the two or more prognostic blocks corresponding to the different intra-prognostic modes, the method further comprises: obtaining a mutually exclusive prognostic mode with the intra-prognostic method; and determining to use the intra-prognostic method, and proceeding with obtaining the two or more prognostic blocks corresponding to the different intra-prognostic modes; or determining to use the mutually exclusive prognostic mode.

44. An encoder, comprising: a forecasting module, configured to perform intra-forecasting on a block to be processed by using two or more different intra-forecasting modes, to obtain two or more forecast blocks corresponding to the different intra-forecasting modes; a combination module, configured to obtain a target forecast block from the block to be processed based on weighting matrices and the two or more forecast blocks obtained; and a processing module, configured to: test all or some of the possible combinations of forecasting modes and weighting matrix derivation modes, calculate loss costs and select a combination with a low loss cost; use two or more different intra-forecasting modes and weighting matrices in the combination as the two or more different intra-forecasting modes and weighting matrices for the intra-forecast;and write in a bit stream, according to the syntax, information about the two or more different intra-prognostic modes and determined weight matrix derivation modes.; 45. The encoder according to claim 44, further characterized in that the processing module is additionally configured to set a flag, wherein the flag is a photograph-level flag to indicate to a decoder whether to proceed with obtaining the two or more forecast blocks corresponding to the different intra-prognostic modes; and / or QQI Qnn / Q7n7 / B / YIAI the flag is a flag below a photograph level and above an encoding unit (CU) level and used to indicate to the decoder whether to proceed with obtaining the two or more forecast blocks corresponding to the different intra-prognostic modes in a specified area.

46. ​​The encoder according to claim 44, further characterized in that the processing module is additionally configured to: establish a mutually exclusive forecasting mode with the intra-forecast; and avoid using the mutually exclusive forecasting mode when the intra-forecast is performed on the block to be processed; or avoid performing the intra-forecast when the mutually exclusive forecasting mode is used for the block to be processed.

47. The encoder according to claim 44, further characterized in that the combination module is additionally configured to: store intra-prognostic mode information used in intra-prognostic.

48. The decoder according to claim 44, further characterized in that it additionally comprises a processing module, configured to: establish a mutually exclusive forecasting mode with the intra-forecast; and avoid using the mutually exclusive forecasting mode when the intra-forecast is performed on the block to be processed; or avoid performing the intra-forecast when the mutually exclusive forecasting mode is used for the block to be processed.

49. An intra-forecasting method comprising: performing intra-forecasting on a block to be processed by using two or more different intra-forecasting modes; during forecasting by using each intra-forecasting mode, when forecasting a predefined number of pixels, obtaining a predefined number of forecast pixels from the block to be processed based on a weighting matrix and the pixels corresponding to the intra-forecasting mode that has been forecasted; and obtaining a target forecast block from the block to be processed based on the obtained predefined number of forecast pixels.

50. An intra-prognostic device, comprising: a prognostic module, configured to perform intra-prognostication on a block to be processed by using two or more different intra-prognostic modes obtained through decoding, to obtain two or more prognostic blocks corresponding to the different intra-prognostic modes; and a combination module, configured to obtain a target prognostic block from the block to be processed based on weighting matrices and the two or more prognostic blocks obtained.

51. The intra-prognostic device according to claim 50, further characterized in that the device is arranged in an encoder or a decoder.

52. The intra-prognostic device according to claim 51, further characterized in that the device is arranged in the decoder, and the device further comprises a decoding module, wherein the decoding module is configured to decode a received bit stream and obtain the two or more different intra-prognostic modes, the block to be processed, and the weighting matrices.

53. The intra-prognostic device according to claim 52, further characterized in that the decoding module is additionally configured to: determine, based on a flag at the photograph level, whether to proceed with the intra-prognostic realization in the block to be processed by using the two or more different intra-prognostic modes obtained through decoding.

54. The intra-prognostic device according to claim 52, further characterized in that the decoding module is further configured to determine, based on a flag below a photograph level and above the level of an encoding unit (CU), whether to proceed with performing intra-prognostic analysis on the block to be processed by using the two or more different intra-prognostic modes obtained through decoding.

55. The intra-forecasting device according to claim 51, further characterized in that the device is arranged in the encoder, and the device further comprises a processing module, wherein the processing module is configured to: test all or some of the possible combinations of forecasting modes and weight matrix derivation modes, calculate loss costs and select a combination with a low loss cost; use two or more different intra-forecasting modes and weight matrices in the combination as the two or more different intra-forecasting modes and weight matrices for the intra-forecasting; and write to a bit stream, according to the syntax, information about the two or more different intra-forecasting modes and weight matrix derivation modes determined.

56. The intra-prognostic device according to claim 55, further characterized in that the processing module is additionally configured to set a flag, wherein the flag is a photograph-level flag to indicate to a decoder whether to proceed with obtaining the two or more prognostic blocks corresponding to the different qq Lcnn / cznz / E / YiAi intra-prognostic modes; and / or the flag is a flag below a photograph level and above an encoding unit (CU) level and used to indicate to the decoder whether to proceed with obtaining the two or more prognostic blocks corresponding to the different intra-prognostic modes in a specified area.

57. The intra-prognostic device according to claim 55, further characterized in that the processing module is additionally configured to: establish a mutually exclusive prognostic mode with the intra-prognostic; and avoid using the mutually exclusive prognostic mode when the intra-prognostic is performed on the block to be processed; or avoid performing the intra-prognostic when the mutually exclusive prognostic mode is used for the block to be processed.

58. The intra-prognostic device according to claim 50, further characterized in that the combination module is additionally configured to: store intra-prognostic mode information used in the intra-prognostic.

Images