CN101500159B - Method and apparatus for image entropy encoding, entropy decoding - Google Patents

Abstract

The invention provides a method for encoding an image entropy and decoding an entropy and a device thereof, mainly comprising the following steps: according to a code table of entropy encoding and decoding corresponding to the first kind of block mapping, parameters of a code table of entropy encoding and decoding corresponding to the second kind of block mapping are mapped and obtained; and the parameters of the code table of entropy encoding and decoding are utilized for carrying out the entropy encoding and decoding of the image block data under the second kind of block mapping. In the embodiment of the invention, as only the code table corresponding to the first kind of block mapping is stored, and then is converted and processed according to the code table corresponding to the first kind of block mapping, the code table corresponding to the second kind of block mapping is obtained; therefore, under the condition of occupying less storing resource occupation, the invention can realize the reusing of the code table in different mapping modes, further ensures the efficiency of entropy encoding and entropy decoding.

Description

Method and device for image entropy encoding and entropy decoding

technical field

The present invention relates to the technical field of encoding and decoding, in particular to an image entropy encoding and entropy decoding method and device. the

Background technique

In the process of video and image coding and decoding, in order to reduce the data volume of video and image data during transmission or storage, it is usually necessary to adopt corresponding compression coding and decoding technologies. In the process of video and image compression and encoding, transformation technology is needed to concentrate an image, image content and information in a region in a specific region after transformation, so that the corresponding content can be effectively compressed in the video compression algorithm. Specifically, an image or an area in the image may be divided into several small blocks or sub-areas (called sub-blocks, or image blocks), and then transformed in units of sub-blocks. Wherein, the size of the sub-block may be, but not limited to, 4×4 or 8×8, and the corresponding 4 and 8 are both in units of image pixels. In video and image coding and decoding, a 16×16 image block is usually used as a codec processing unit, and this unit becomes a macroblock. A macroblock may be further divided into several sub-blocks with smaller sizes. In existing documents, sub-blocks and macroblocks may be collectively referred to as image blocks. the

In the process of sub-block division, specifically, a specific area can be divided according to different sub-block sizes, and then different transformation processes are performed on sub-blocks of different sizes, which is called adaptive block size transformation technology. The transformation mentioned here refers to a data processing method with certain mathematical characteristics, such as DCT (discrete cosine transform), wavelet transform, integer transform, etc. These transformations are usually expressed in matrix form in video and image coding and decoding. the

When there are multiple transformation modes in the adaptive block size transformation technology, some macroblocks can use one transformation mode alone. The transformation mode refers to the transformation and the corresponding codec processing method for transformation. All sub-blocks in a ×16 macroblock are divided into 8×8 sizes, and all sub-blocks adopt 8×8 transformation and encoding and decoding techniques corresponding to the transformation (that is, adopt a transformation mode). At the same time, sub-blocks in some macroblocks can use different transformation modes, for example: if a 16×16 macroblock contains 4×4, 4×8, 8×4 and 8×8 sub-block types, Then the transform modes corresponding to 4×4, 4×8, 8×4 and 8×8 can be used respectively. the

Among them, if the data blocks adopt different transformations, the distribution rules of the data after different transformations will be different. Therefore, VLC (Variable Length Coding) is performed on the data blocks after different transformations. ) and VLD (Variable Length Coding, variable length decoding)) processes, it is necessary to use the appropriate code table for encoding or decoding operations according to the distribution law (that is, statistical characteristics) of the data blocks. the

When using variable-length coding in the video codec standard, a set of code tables will be set for the coding elements. The code tables are divided into coded image brightness component elements, image chrominance component elements, inter-frame coding, and intra-frame coding according to the coding elements. The coding element refers to the run length (run) and the quantized coefficient value (level) generated during the encoding process, where run refers to the number of coefficients whose value is 0 before a non-zero coefficient in the image block scanning during the encoding or decoding process number, level refers to the numerical size of the quantization coefficient. In variable-length encoding, run and level appear as a pair of data. the

In variable-length coding, after the coder obtains a pair of run and level, it searches for the corresponding index value in the code table according to the values of run and level, and the index value is called the transformation coefficient (trans_coefficient). In variable length encoding, the value of trans_coefficient will be specified within a certain range. If the value of trans_coefficient cannot be determined because the level or run value is too large during encoding, the escape coefficient difference (escape_level_diff) will also be used. Escape_level_diff is usually between trans_coefficient and Used when there is no connection between run and level pairs. During the encoding process, the encoding system will take into account factors such as the values of trans_coefficient and escape_level_diff (if present) and whether the current encoding element is an image block luminance component element, an image block chrominance component element, whether it is an inter-frame coding element, or an intra-frame coding element. Determines the code table to use for the next encoded element. If the encoding of the current image block ends, the encoding system will write an end-of-block (EOB) after the encoding of the last encoding element of the current image block is completed, and the value of the EOB is determined by the current code table. the

In variable-length decoding, according to the values of trans_coefficient and escape_level_diff (if present) parsed from the code stream, look up the pair of run and level corresponding to trans_coefficient and escape_level_diff in the corresponding code table. During the decoding process, the decoding system will take into account factors such as the values of trans_coefficient and escape_level_diff (if present) and whether the current decoding element is an image block luminance component element, an image block chrominance component element, whether it is an inter-frame coding element, or an intra-frame coding element. Determines the code table to use for the next decoded element. If the syntax element analyzed by the decoding system is the corresponding value of the EOB in the current code table, it means that the decoding of the current image block is completed, that is, the decoding of the next image block needs to be started. the

Usually, in order to save the memory resources occupied by the encoder and decoder, usually only one code table corresponding to the transformed data block is set, and data blocks of other sizes can be transformed accordingly, and then the set corresponding code table can be used Perform encoding or decoding operations. the

Taking 4×4 blocks and 8×8 blocks as examples, in the process of encoding or decoding operations, the corresponding implementation schemes for selecting and using the code table can be specifically as follows:

Option One

Assuming that there is currently only a code table corresponding to the 4×4 block, for the data block after the 8×8 transformation (that is, the 8×8 block), the way to use the code table can be as follows: first, the 8×8 block is sampled The method is divided into 4×4 data blocks; after that, for the obtained 4 4×4 data blocks, code tables corresponding to 4×4 blocks are respectively used for encoding or decoding operations. the

Option II

Assuming that there are only code tables corresponding to 8×8 blocks at present, for the data blocks obtained after 4×4 transformation (that is, 4×4 blocks), the way to use the code tables can be: combine multiple 4×4 blocks with a certain Then, for the obtained 8×8 block, the code table corresponding to the corresponding 8×8 block can be used for encoding or decoding.

In the process of realizing the present invention, the inventors found that, in the above-mentioned prior art, no matter whether it is to divide the 8×8 blocks and then adopt the corresponding code table, or to string 4×4 blocks into 8×8 blocks and use The implementation schemes of the corresponding code tables all have the problem that the currently used code table cannot well adapt to the data distribution law of the data block of the corresponding size, thereby reducing the corresponding encoding and decoding efficiency. the

Contents of the invention

Embodiments of the present invention provide a method and device for image entropy encoding and entropy decoding, so as to reduce the occupation of storage resources and improve encoding and decoding efficiency. the

An image entropy coding method, comprising:

According to the entropy encoding code table corresponding to the first type of block transformation, the entropy encoding code table parameters corresponding to the second type of block transformation are obtained by mapping, and the entropy encoding is performed on the second type of block transformation image block data by using the entropy encoding code table parameters;

Wherein, the mapping process includes performing a mapping operation according to the jump parameter used to determine the code table sequence number after the jump of the entropy coding code table; or, adjusting the entropy coding code table corresponding to the first type of block transformation according to the mapping parameter The element value in is used to obtain the entropy encoding code table parameter corresponding to the second type of block transformation after mapping. the

A method of entropy decoding, comprising:

According to the entropy decoding code table corresponding to the first type of block transformation, map to obtain the entropy decoding code table parameters corresponding to the second type of block transformation, and use the entropy decoding code table parameters to perform entropy decoding on the image block data under the second type of block transformation;

Wherein, the mapping process includes performing a mapping operation according to the jump parameter used to determine the code table sequence number after the jump of the entropy coding code table; or, adjusting the entropy coding code table corresponding to the first type of block transformation according to the mapping parameter The element value in is used to obtain the entropy encoding code table parameter corresponding to the second type of block transformation after mapping. the

An entropy encoding device, comprising:

The first storage unit is used to save the entropy encoding code table corresponding to the first block transformation;

The first code table processing unit is used to perform mapping processing on the entropy coding code table corresponding to the first type of block transformation stored in the first storage unit, and obtain the entropy coding code table parameters corresponding to the second type of block transformation;

An encoding unit, configured to use the entropy encoding code table parameters corresponding to the second type of block transformation obtained by the first code table processing unit to perform entropy encoding operations on the data of the image block of the second type of block transformation;

Wherein, the first code table processing unit performing the mapping process includes performing a mapping operation according to the jump parameter used to determine the code table serial number after the entropy coding code table jumps; or, adjusting the first type of block transformation according to the mapping parameter Corresponding to the element values in the entropy encoding code table, the entropy encoding code table parameters corresponding to the second type of block transformation after mapping are obtained. the

An entropy encoding device, comprising:

The second storage unit is used to save the entropy decoding code table corresponding to the first block transformation;

The second code table processing unit is used to obtain the entropy decoding code table parameters corresponding to the second block transformation according to the entropy decoding code table mapping corresponding to the first block transformation stored in the second storage unit;

The decoding unit is used to use the entropy encoding code table parameters corresponding to the second type of block transformation obtained by the second code table processing unit to perform entropy decoding operations on the data of the image block of the second type of block transformation;

Wherein, the said second code table processing unit performing the mapping process includes performing a mapping operation according to the jump parameter used to determine the code table serial number after the entropy coding code table jumps; or, adjusting the first type according to the mapping parameter The element values in the entropy encoding code table corresponding to the block transformation are obtained to obtain the entropy encoding code table parameters corresponding to the second type of block transformation after mapping. the

It can be seen from the technical solutions provided by the above-mentioned embodiments of the present invention that in the embodiments of the present invention, since only the code table corresponding to the first type of block transformation is saved, and for the code table corresponding to the second type of block transformation, it is Obtained according to the stored code table conversion processing corresponding to the first type of block conversion, therefore, it can realize the different conversion modes under the condition of occupying less storage resources (only need to store the code table corresponding to the first type of block conversion) Multiplexing of the code table, and guarantee the efficiency of entropy coding or entropy decoding. the

Description of drawings

Fig. 1 is the processing schematic diagram that the embodiment of the present invention provides;

Fig. 2 is a schematic structural diagram of an entropy encoding device provided by an embodiment of the present invention;

FIG. 3 is a schematic structural diagram of an entropy decoding device provided by an embodiment of the present invention. the

Detailed ways

In the embodiment of the present invention, when only the code table corresponding to one type of block transformation is saved, the corresponding code table corresponding to another type of block transformation can be obtained according to the mapping process. the

As shown in Figure 1, the corresponding processing process may include: when it is necessary to obtain a code table corresponding to another block transformation, obtain the stored code table corresponding to the first type of block transformation, and obtain the stored code table corresponding to the first type of block transformation. After the code table, the entropy encoding code table parameters or entropy decoding code table parameters corresponding to the second type of block transformation can be obtained according to the entropy encoding code table or entropy decoding code table mapping process corresponding to the first type of block transformation. The corresponding first kind of block transformation and the second kind of block transformation are transformations under different block sizes, and the corresponding different block sizes can be 4×16, 16×4, 16×8, 8×16, 8×8, 8× 4, 4×8 or 4×4, etc. the

During the above processing, the corresponding entropy encoding code table parameters or entropy decoding code table parameters include at least one of the following:

Intra-frame luminance code table number and type under variable-length coding, inter-frame luminance code table number and type, intra-frame chroma code table number and type, inter-frame chroma code table number and type, run program run, Quantization coefficient value level, transform coefficient trans_coefficient, escape coefficient difference escape_level_diff and increase image block end flag EOB;

The data of the corresponding image block may include the value of the run length and the value of the quantization coefficient value level. the

Wherein, the corresponding mapping process may specifically be determined according to the corresponding statistical characteristics of the second type of block transformation. Specifically, it is possible to analyze the distribution law of data after different transformation processes, and set a corresponding mapping processing method for the code table corresponding to the data obtained through a certain transformation, so that a saved block transformation code table can be processed. Another data distribution rule corresponding to block transformation. the

In the above implementation solution, the corresponding mapping process may be a rule for adjusting the jump sequence between code tables, or may be a rule for adjusting elements in the code table. the

If the rules for adjusting the jump order between code tables are used for mapping processing, it can include:

At the coding end, the jump parameters corresponding to the second type of block transformation can be saved in advance, or the corresponding jumps corresponding to the current coding sequence or coding image or coding strip can be determined according to the coding characteristics of the current coding sequence or coding image or coding strip. In this way, the jump order of the code tables in the code table corresponding to the corresponding first block conversion can be adjusted according to the jump parameters corresponding to the saved second block conversion, and the adjusted jump order The code table is used as an entropy coding code table parameter corresponding to the second type of block transformation to perform an entropy coding operation of the image under the second type of block transformation. Wherein, the corresponding jump parameter may specifically be determined according to the statistical characteristics of the corresponding second type of block transformation; for example, the code table corresponding to the second type of block transformation may be obtained according to the statistical characteristics of the second type of block transformation, and the Correspondence between the code tables corresponding to the two types of block transformations and the code tables corresponding to the first type of block transformations determine corresponding jump parameters. the

In the above process, the corresponding jump parameter can be the serial number of the code table after the jump, or, it can also be the difference between the serial number of the code table after the jump and the current code serial number; Coding properties of a picture or coded slice may include: pixel values, residual values, pixel values or mathematical statistical characteristic values of a sequence or picture or slice, or pixel values and residual values of a sequence or picture or slice The mathematical statistical eigenvalue of the difference. the

Correspondingly, at the decoding end, the jump parameters corresponding to the second type of block transformation can also be saved in advance, and the corresponding current decoding sequence or decoding image can also be determined from the encoded code stream according to the current decoding sequence or decoding image or decoding strip Or decode the jump parameter corresponding to the slice, and then determine the corresponding entropy decoding code table parameter corresponding to the second type of block conversion according to the obtained jump parameter and the code table corresponding to the first type of block conversion, for use in An entropy decoding operation is performed on the decoded sequence or decoded image or decoded slice of the current image of the second type of block transformation. the

If the rules for adjusting the elements in the code table are used for mapping processing, it can include:

At the coding end, pre-save the mapping parameters corresponding to the second type of block transformation, or determine the corresponding mapping parameters corresponding to the current coding sequence or coding image or coding strip according to the coding characteristics of the current coding sequence or coding image or coding strip; and The element values in the code table corresponding to the first type of block transformation can be adjusted according to the corresponding mapping parameters corresponding to the second type of block transformation, and the entropy encoding code table parameters corresponding to the second type of block transformation can be obtained to facilitate the second type of block transformation The entropy encoding operation of the image below. Wherein, the corresponding mapping parameters can specifically be determined according to the statistical characteristics of the second type of block transformation; specifically, the code table corresponding to the second type of block transformation can be obtained according to the statistical characteristics of the second type of block transformation, and according to the second type of block transformation Correspondence between element values contained in the code table corresponding to the block transformation and corresponding element values contained in the code table corresponding to the first type of block transformation determine corresponding mapping parameters. Among them, the coding characteristics of the corresponding coded sequence or coded image or coded strip may include: the pixel value of the sequence or picture or strip, the residual value, the mathematical statistical characteristic value of the pixel value or the residual value, or the sequence or picture or the mathematical statistical eigenvalues of the pixel values and residual values of the strips. the

Correspondingly, at the decoding end, the mapping parameters corresponding to the second type of block transformation can also be saved in advance, or the current decoding sequence or decoding image or decoding strip can be obtained from the encoded code stream according to the current decoding sequence or decoding image or decoding strip Take the corresponding mapping parameters, and adjust the data elements in the entropy decoding code table corresponding to the first block transformation according to the mapping parameters, and use the adjustment result as the entropy decoding code table parameters corresponding to the second block transformation (that is, according to the acquired The mapping parameters and the code table corresponding to the first type of block transformation determine the corresponding code table corresponding to the second type of block transformation), so as to perform entropy decoding on the decoded sequence or decoded image or decoded slice of the image under the current second type of block transformation. Alternatively, at the decoding end, the data element corresponding to the second type of block transformation received from the entropy encoded code stream can also be adjusted according to the mapping parameters, and the corresponding entropy decoding code can be transformed according to the adjusted data element value and the first type of block transformation The table obtains entropy decoding code table parameters corresponding to the second type of block transformation, so as to perform entropy decoding operations. the

That is to say, the above-mentioned jump parameters used to indicate the rules of the jump sequence between code tables or the mapping parameters used to indicate the element mapping processing method in the code table may be pre-reserved in the entropy encoding and entropy decoding systems. It is also possible to update the jump order rules between code tables or the element processing rules in the code table according to the characteristics of the entropy coding sequence or image or strip at the entropy coding end, and write the above rules into the entropy coding information such as the sequence, image or strip, In order to facilitate the decoding end to obtain corresponding processing rules; correspondingly, the corresponding jump parameters or mapping parameters can be obtained by parsing from encoding elements such as sequences, images, or slices in the code stream at the decoding end. the

In the process of entropy encoding, the above-mentioned embodiments of the present invention can specifically be used to perform entropy encoding operations on any one or more of the following currently encoded data:

(1) The current entropy coding data is the image brightness component;

(2) The current entropy coded data is the image chrominance component;

(3) The current entropy encoded data is intra-frame entropy encoded data;

(4) The current entropy encoded data is inter-frame entropy encoded data. the

In the process of entropy decoding, the above-mentioned embodiments of the present invention can be specifically used to perform entropy decoding operations on any one or more of the following current entropy decoding data:

(1) The current entropy decoding data is the image brightness component;

(2) The current entropy decoding data is the image chrominance component;

(3) The current entropy decoding data is intra-frame entropy coding data;

(4) The current entropy-decoded data is inter-frame entropy-coded data. the

To facilitate the understanding of the embodiments of the present invention, the implementation process of the embodiments of the present invention will be described in detail below using specific application examples. the

Specific embodiment one

In this embodiment, the rules for adjusting the jump order between code tables are used as conversion rules, that is, the corresponding code tables are obtained by using code table jump implementation. the

Assume that the transformed blocks include 4×4 blocks and 8×8 blocks. Generally, the data distribution of 4×4 blocks obtained after 4×4 transformation is steeper than that of 8×8 blocks. When using 8×8 When the VLC and VLD code tables corresponding to a block encode or decode a 4×4 block, the data distribution rule of the 4×4 block can be adapted to by changing the jumping condition of the code table. The image block that undergoes 4×4 transformation (encoding end) or 4×4 inverse transformation (decoding end) is called a 4×4 transformation block; after 8×8 transformation (encoding end) or 8×8 inverse transformation (coding end ) is called an 8×8 transform block. the

Assume that in the process of entropy encoding or entropy decoding, there are 19 code tables corresponding to VLC and VLD code tables for the corresponding 8×8 block, which can specifically include:

(1) There are 7 code tables for intra-frame encoding of luma blocks, counted as: VLC0_Intra, VLC1_Intra, VLC2_Intra, VLC3_Intra, VLC4_Intra, VLC5_Intra, VLC6_Intra;

(2) 7 pieces of luminance block inter-frame coding, counted as: VLC0_Inter, VLC1_Inter, VLC2_Inter, VLC3_Inter, VLC4_Inter, VLC5_Inter, VLC6_Inter;

(3) Five chrominance block code tables, counted as: VLC0_Chroma, VLC1_Chroma, VLC2_Chroma, VLC3_Chroma, VLC4_Chroma. the

Among them, when using the above-mentioned tables to perform VLC encoding or VLD decoding on 8×8 data blocks, the jump rule of the code table is to increase one by one, that is, when the coding elements meet certain conditions, the code table will switch to the sequence number of the current code table. Tables that are next to each other but whose sequence number is one higher, for example, switch from VLC5_Intra to VLC6_Intra, etc. the

When the above-mentioned tables are used to perform VLC encoding or VLD decoding on the 4×4 transformed data blocks (that is, 4×4 blocks), the data distribution of the 4×4 blocks is different from the 8×8 obtained through 8×8 transformation. If the distribution rules of the data of the 8 blocks are inconsistent, the code table jump rules suitable for the 4×4 blocks can be used to select and apply each code table to adapt to the distribution rules of the data of the 4×4 blocks. Considering that the corresponding data change range in the data distribution of 4×4 blocks is larger than that of 8×8 blocks in general, the code table jumping methods that can be used can be as shown in Table 1 to Table 3 :

Table 1

Table 2

table 3

In the above three jump tables, when the current element is coded or decoded using a certain code table, if the current coded or decoded element satisfies the predetermined condition, the code table used for the next coded or decoded element will jump to the code table according to the current code The table corresponds to a jump to a determined code table, specifically jumping to the code table pointed to by the sum of the current code table serial number plus the corresponding jump value. For example, when performing inter-frame encoding and decoding on a luma block, the current encoding or decoding element uses the VLC3_Inter code table in each code table corresponding to the inter-frame encoding of the luma block (that is, each code table in Table 2). At this time, if The value of the jump parameter N4 is 2, then when the current encoding or decoding element satisfies the predetermined condition, then because the sequence number of the VLC3_Inter code table is 3, and then the N4 value is added, the sequence number of the code table after the jump is 5, namely The code table used by the corresponding next encoding or decoding element will jump to the VLC5_Inter code table whose sequence number is 5. the

In the above three jump tables, if the VLC5_Intra or VLC5_Inter or VLC3_Chroma code table is used for the previous encoding or decoding element, it is the penultimate code table under the current encoding conditions (such as luma block intra-frame coding) , the code table that can be used for the next encoding or decoding element can only be the last code table, so the corresponding jump parameter value is set to 1. the

In the above three jump tables, if the previous encoding or decoding element uses the VLC6_Intra or VLC6_Inter or VLC4_Chroma code table because it is the last code table under the current coding conditions (such as luma block intra-frame coding), so jump The transfer parameter value is unavailable and set to N/A; that is, the sum of the jump parameter value of the code table and the serial number of the current code table cannot exceed the serial number value of the last code table under the current encoding or decoding conditions, otherwise, it will not be determined The code table used after the jump. the

In the above three jump tables, each jump parameter value is determined according to the distribution law of the data of the 4*4 blocks. If it is necessary to obtain code tables of other kinds of block transformations, the corresponding jump parameters need to be set according to the promulgation rules of the data of this kind of block transformations. the

The above jump table can be pre-stored in the encoding and decoding system, and the jump table can also be updated at the encoding end according to the characteristics and contents of the current encoding sequence, image, and strip, and the updated jump table can be stored in the sequence header or image header or strip header. The decoding end extracts the updated jump table from the sequence header or picture header or slice header in the code stream to decode the current sequence or picture or slice. the

Specific embodiment two

In this embodiment, the encoding end maps the elements in the entropy encoding code table to obtain the entropy encoding parameters corresponding to the second type of block transformation; the decoding end maps the image block data using the second type of block transformation according to the mapping rule , using the entropy decoding code table corresponding to the first type of block transformation to decode the mapped data. In the embodiment of the present invention, the entropy encoding and entropy decoding code tables corresponding to the second transformation can also be generated at the entropy encoding and entropy decoding ends respectively according to the above mapping rules, so as to encode and decode the image block data using the second transformation. the

Taking VLC encoding and VLD decoding code table as an example, the elements in the corresponding code table include: index value, run, level, EOB. In this embodiment, the index value in the code table is trans_coefficient. If both 4×4 and 8×8 transforms are used in the current encoding. Since there are a total of 64 pixel values to be encoded in the 8×8 transform, there may be at most 64 pairs of run and level pairs; and there are a total of 16 pixel values to be encoded in the 4×4 transform, and the run and level pairs may be at most There are 16 pairs. Usually, the value of EOB is related to the logarithm of the pair of run and level under the coding conditions corresponding to the current code table. If there are more logarithms of run and level pairs under the current encoding conditions in a statistical sense, the value of EOB will be set to a larger value, otherwise EOB will be set to a smaller value. Since the 4×4 transform block itself has fewer elements to be encoded than the 8×8 transform block, when encoding the 4×4 transform block, the corresponding EOB value should be smaller than the EOB value corresponding to the 8×8 transform block, otherwise it will Reduce coding efficiency. the

To sum up, if the current encoding system stores an 8×8 code table, the value of EOB should not be set too large when using this code table to encode a 4×4 transform block. Therefore, the value corresponding to EOB in the original code table is M, and the corresponding EOB value in the code table can be converted to a smaller value N when using the 8×8 transformed code table to encode a 4×4 transform block, while the original code table In the index value from N to M-1, the values of the index values from N to M-1 are sequentially increased by 1 when encoding 4×4 transform block data, but the corresponding run and level pairs of data remain unchanged. That is, the index value corresponding to the original run and level pair is converted from N to M-1 to N+1 to M. In this way, since the bit overhead of the EOB is greatly reduced when encoding the 4×4 transform block data, an ideal coding gain can be brought to the encoding of the 4×4 transform block. the

At the decoding end, when decoding 4×4 transform block data, if the index value is between N+1 and M, the index value is decremented by 1; if the index value is equal to N, the index value is assigned to M. After the above operations, search the converted index value in the current code table to obtain the corresponding pair of run and level. the

During the above processing, the conversion relationship between the corresponding index value and the encoding or decoding element can be obtained according to the statistical characteristics of the data. For example, in the VLC and VLD code tables, if the EOB value is 8, a coded or decoded image block has 7 pairs of run values and level values in a statistical sense under current coding or decoding conditions. When encoding the 4×4 transformed 4×4 image block data, the value corresponding to the code table EOB can be converted to '0', that is, the EOB value is changed from '8' when encoding the 4×4 transformed block If it is 0, write the value '0' into the code stream. When decoding the 4×4 transform block, if the element with the value '0' is analyzed, its value will be converted to '8'. According to the current code table, this element can be known is EOB, the decoding of the current image block ends. the

Assuming that the value corresponding to the EOB of the VLC3_Intra code table of Embodiment 1 is 8, the EOB value can be mapped to 0 when encoding or decoding the 4×4 image block data after 4×4 transformation, and the value is an index of 0 to 7 The values are mapped to 1 to 8 in turn, and the corresponding run value and level value remain unchanged, as shown in the following table, where Table 4 shows the EOB value and index value in the VLC3_Intra code table of the original 8×8 image block The relationship between run and level, Table 5 shows the relationship between the EOB value and index value in the code table of the 4×4 image block after the mapping process, such as run and level data:

Table 4

table 5

The above conversion method for the EOB value of the VLC3_Intra code table can be extended to the remaining 18 code tables in the embodiment. That is, it is assumed that CurrentVLCTable represents the current code table, and trans_coefficient is the index value. If the current encoding block is a 4×4 transformation block, convert the EOB value in CurrentVLCTable to '0', and write the value '0' into the code stream when encoding EOB; trans_coefficient in CurrentVLCTable is 0 to the corresponding value of EOB in CurrentVLCTable The value of the encoding element corresponding to -1 is incremented by 1 when the code stream is written. When decoding, if the current block is a 4×4 transform block, the mapping operation is performed as follows: if trans_coefficient is between 1 and the corresponding value of EOB in CurrentVLCTable (inclusive), then trans_coefficient is reduced by 1; if trans_coefficient is equal to 0, then trans_coefficient is assigned a value of The value of EOB in CurrentVLCTable. the

It can be seen that the code table corresponding to the image block of the desired size can be obtained after converting the elements in the saved code table through the above-mentioned mapping conversion method, and the code table is obtained through certain processing rules without storing in the encoding and In the decoding system, system resources are effectively saved. the

The mapping rules of the above-mentioned code table elements under different transformations can be pre-stored in the encoding and decoding system, and the mapping rules can also be updated at the encoding end according to the characteristics and contents of the current encoding sequence, image, and strip, and the updated mapping rules Stored in the sequence header or picture header or slice header. The decoding end extracts the updated mapping rule from the sequence header or picture header or slice header in the code stream to decode the current sequence or picture or slice. the

The embodiment of the present invention also provides an entropy encoding device, its specific implementation structure is shown in Figure 2, and may include the following processing units:

(1) The first storage unit is used to store the entropy encoding code table corresponding to the first block transformation;

(2) The first code table processing unit is configured to perform mapping processing on the entropy coding code table corresponding to the first type of block transformation stored in the first storage unit to obtain the parameters of the entropy coding code table corresponding to the second type of block transformation ;

Wherein, the specific mapping processing that the first code table processing unit can adopt includes any of the following implementations:

Mode 1: perform the mapping operation according to the jump parameter used to determine the code table serial number after the jump, wherein the jump parameter is the code table serial number after the jump, or the jump parameter is the jump The difference between the code table serial number after and the current code serial number;

Mode 2: Adjust the element values in the entropy coding code table corresponding to the first block transformation according to the mapping parameters corresponding to the second block transformation, and obtain the entropy coding code table parameters corresponding to the second block transformation;

(3) a coding unit, used to use the entropy coding code table parameters corresponding to the second block transformation obtained by the first code table processing unit to perform entropy coding operations on the data of the image block of the second block transformation;

The encoding unit is configured to perform an entropy encoding operation on the data of the current or next image block transformed by the second type of block processed by the second type of transformation according to the entropy encoding code table parameters corresponding to the second type of block transformation. the

Optionally, the device may also include a first jump parameter acquisition unit, configured to acquire a preset jump parameter; or, it may also be used to acquire a coding sequence based on a current coding sequence or a coding image or a coding strip. The jump parameters determined by the characteristics, and notify the encoding unit to write the jump parameters into the encoded code stream. the

Optionally, the device may also include a first mapping parameter acquisition unit, configured to acquire preset mapping parameters; or, may also be used to acquire a mapping parameter determined according to the encoding characteristics of the current encoding sequence or encoding image or encoding slice. The mapping parameters of the current coding sequence or coded image or coded slice, and notify the coding unit to write the mapping parameters into the code stream. the

The embodiment of the present invention also provides an entropy decoding device, its specific implementation structure is shown in Figure 3, and may include the following processing units:

(1) The second storage unit is used to store the entropy decoding code table corresponding to the first block transformation;

(2) The second code table processing unit is used to obtain the entropy decoding code table parameters corresponding to the second block transformation according to the entropy decoding code table mapping corresponding to the first block transformation preserved by the second storage unit;

The processing of the mapping adopted by the second code table processing unit can include any of the following methods:

Mode 1: perform the mapping operation according to the jump parameter used to determine the code table serial number after the jump, wherein the jump parameter is the code table serial number after the jump, or the jump parameter is the jump The difference between the code table serial number after and the current code serial number;

Mode 2: Adjust the data elements corresponding to the second type of block transformation according to the mapping parameters, and obtain the entropy decoding code corresponding to the second type of block transformation according to the adjusted data element value and the entropy decoding code table corresponding to the first type of block transformation Table parameters; or, the data elements in the entropy decoding code table corresponding to the first block transformation can also be adjusted according to the mapping parameters, and the adjustment result is used as the entropy decoding code table parameter corresponding to the second block transformation;

(3) A decoding unit, configured to use the entropy coding code table parameters corresponding to the second type of block transformation obtained by the second code table processing unit to perform entropy decoding operations on the data of the image block of the second type of block transformation. the

The decoding unit is configured to perform an entropy coding operation on the current or next image under the second type of block transformation using the second type of block transformation according to the entropy decoding code table parameters corresponding to the second type of block transformation. the

Optionally, the device may also include a second jump parameter acquisition unit, configured to acquire a preset jump parameter; Mapping parameters of the current decoded sequence or decoded image or decoded slice obtained in the bitstream. the

Optionally, the device may further include a second mapping parameter acquisition unit, configured to acquire preset mapping parameters; or, configured to obtain the coded code stream received according to the current decoding sequence or decoded image or decoded slice Get the mapping parameters of the currently decoded sequence or decoded image or decoded slice. the

In summary, the implementation of the embodiment of the present invention can obtain a suitable code table according to the data statistical characteristics of the image block of the current size, thereby effectively reducing the entropy caused by the ABT (Adapt Block Transform, Adaptive Block Transform) technology Encoding code table overhead. Moreover, in the embodiment of the present invention, it is only necessary to save the code table corresponding to an image block of a certain size to obtain the code table of an image block of another size, so that the occupation of hardware resources can be effectively reduced, and at the same time, the corresponding encoding and decoding can be improved. efficiency. the

The above is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of changes or modifications within the technical scope disclosed in the present invention. Replacement should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be determined by the protection scope of the claims. the

Claims (30)

1. An image entropy coding method, comprising:

mapping to obtain entropy coding code table parameters corresponding to a second block transformation according to an entropy coding code table corresponding to the first block transformation, and performing entropy coding on the second block transformation image block data by using the entropy coding code table parameters;

the mapping processing comprises the step of mapping according to a jump parameter for determining the code table serial number after the jump of the entropy coding code table; or adjusting element values in the entropy coding code table corresponding to the first block transform according to the mapping parameters to obtain entropy coding code table parameters corresponding to the second block transform after mapping; the mapping parameter and the skip parameter are determined according to the statistical properties of the corresponding second block transform.

2. The method as claimed in claim 1, wherein when the mapping operation is performed using a skip parameter according to the code table number used for determining the skip of the entropy coding code table, the skip parameter is the code table number of the entropy coding code table after the skip, or the skip parameter is a difference value between the code table number of the entropy coding code table after the skip and the current code table number.

3. The method of claim 2, wherein entropy encoding the second block-transformed image block data using the entropy encoding code table parameters comprises:

and after the coding code table corresponding to the second block transform is determined in the coding code table corresponding to the first block transform according to the code table serial number jumped by the entropy coding code table, entropy coding is carried out on the data of the current or next second block transform image block according to the coding code table corresponding to the second block transform.

4. The method as claimed in claim 2, wherein the step of obtaining the skipped code table sequence number of the entropy coding code table further comprises:

if the code table corresponding to the serial number of the current code table is the penultimate code table under the current coding condition, the serial number of the code table after the entropy coding code table jumps is the serial number of the code table corresponding to the last code table under the current coding condition;

or,

and if the code table corresponding to the current code table serial number is the last code table under the current coding condition, not performing entropy coding code table skip operation.

5. The method of claim 2, wherein the hop parameter is preset; or, determining the jump parameter according to the coding characteristic of the current coding sequence or the coding image or the coding strip, and writing the jump parameter into the coding code stream.

6. The method of claim 1, wherein the step of adjusting the values of the elements in the entropy coding code table corresponding to the first block transform comprises;

and carrying out mapping adjustment processing on element values in a preset numerical value range in a coding code table corresponding to the first block transformation.

7. The method of claim 1, wherein the adjusting the values of the elements in the entropy coding code table corresponding to the first block transform comprises:

if the element value in the code table is the end mark of the current image block, mapping the element value M to N, wherein M and N are respectively an integer value, adding 1 to the element value between N and M-1, and keeping the index relationship between the element value and other elements in the code table unchanged.

8. The method of claim 1, wherein the mapping parameter is preset; or determining the mapping parameters of the current coding sequence or the coding image or the coding strip according to the coding characteristics of the current coding sequence or the coding image or the coding strip, and writing the mapping parameters into the code stream.

9. The method according to claim 5 or 8, wherein the coding characteristics of the coded sequence or coded picture or coded slice comprise: pixel values, residual values, mathematical statistical characteristic values of pixel values or residual values of a sequence or image or strip, or pixel values and mathematical statistical characteristic values of residual values of a sequence or image or strip.

10. The method according to any of claims 1 to 8, wherein said entropy encoding code table parameters comprises: at least one of the sequence number and type of an intra-frame brightness coding code table under variable length coding, the sequence number and type of an inter-frame brightness coding code table, the sequence number and type of an intra-frame chroma coding code table, the sequence number and type of an inter-frame chroma coding code table, run, quantization coefficient value level, transform coefficient trans _ coefficient, escape coefficient difference value escape _ level _ diff and an increase image block end mark EOB; the data of the image block comprises values of run and values of quantized coefficient value level.

11. An entropy decoding method, comprising:

mapping according to the entropy decoding code table corresponding to the first block transformation to obtain entropy decoding code table parameters corresponding to the second block transformation, and performing entropy decoding on the image block data under the second block transformation by using the entropy decoding code table parameters;

the mapping processing comprises mapping operation according to a jump parameter for determining a code table sequence number after the jump of the entropy decoding code table; or adjusting the element value in the entropy decoding code table corresponding to the first block transform according to the mapping parameter to obtain the entropy decoding code table parameter corresponding to the second block transform after mapping; the mapping parameter and the skip parameter are determined according to the statistical properties of the corresponding second block transform.

12. The method as claimed in claim 11, wherein when the mapping operation is performed by using a jump parameter according to the code table sequence number used for determining the jump of the entropy-decoded code table, the jump parameter is the code table sequence number after the jump of the entropy-decoded code table, or the jump parameter is a difference value between the code table sequence number after the jump of the entropy-decoded code table and the current code table sequence number.

13. The method of claim 12, wherein entropy decoding the block data of the second block transform using the entropy decoding code table parameters comprises:

and after the decoding code table corresponding to the second block transform is determined in the decoding code table corresponding to the first block transform according to the code table serial number after the entropy decoding code table is jumped, performing entropy decoding on the data of the current or next image block using the second block transform according to the decoding code table corresponding to the second block transform.

14. The method as claimed in claim 12, wherein the obtaining the skipped code table sequence number further comprises:

if the code table corresponding to the current code table serial number is the penultimate code table under the current decoding condition, the code table serial number after the entropy decoding code table jumps is the code table serial number corresponding to the last code table under the current decoding condition;

or,

and if the code table corresponding to the current code table serial number is the last code table under the current decoding condition, skipping the entropy decoding code table.

15. The method of claim 12, wherein the hop parameter is preset; or, the decoding sequence or the decoding image or the decoding stripe is obtained by analyzing from the received coding code stream according to the current decoding sequence or the decoding image or the decoding stripe.

16. The method of claim 11, wherein the mapping process comprises:

and adjusting the data elements corresponding to the second block transform according to the mapping parameters, and obtaining the entropy decoding code table parameters corresponding to the second block transform according to the adjusted data element values and the entropy decoding code table corresponding to the first block transform.

17. The method of claim 11 or 16, wherein the adjustment of the data element value comprises; and carrying out mapping adjustment processing on element values in a preset numerical range.

18. The method according to claim 11 or 16, wherein said entropy decoding the data of the second image block under block transform using the entropy decoding code table parameters comprises:

if the data element value obtained by analyzing the code stream is N, the data element value is mapped to a value M of an image block end mark in the current entropy decoding code table, the data element value between the data element value N +1 and M in the code stream is subtracted by 1, the index relation between the data element value and other elements in the current entropy decoding code table is kept unchanged, and N and M are integer values.

19. The method according to claim 11 or 16, wherein the mapping parameter is preset; or, the decoding sequence or the decoding image or the decoding stripe is obtained by analyzing from the received coding code stream according to the current decoding sequence or the decoding image or the decoding stripe.

20. The method according to any of claims 11 to 19, wherein entropy decoding the code table parameters comprises: at least one of the sequence number and the type of an intra-frame brightness decoding code table, the sequence number and the type of an inter-frame brightness decoding code table, the sequence number and the type of an intra-frame chroma decoding code table, the sequence number and the type of an inter-frame chroma decoding code table, a run, a quantization coefficient value level, a transform coefficient trans _ coefficient, a transfer coefficient difference value escape _ level _ diff and an increase image block end mark EOB under variable length decoding; the data of the image block comprises values of run and values of quantized coefficient value level.

21. An entropy encoding apparatus, characterized by comprising:

the first storage unit is used for storing an entropy coding code table corresponding to the first block transform;

the first code table processing unit is used for carrying out mapping processing according to the entropy coding code table corresponding to the first block transform stored in the first storage unit to obtain entropy coding code table parameters corresponding to the second block transform;

an encoding unit, configured to perform entropy encoding operation on the data of the image block of the second block transform by using the entropy encoding code table parameters corresponding to the second block transform obtained by the first code table processing unit;

the first code table processing unit performs mapping processing, including mapping operation according to a jump parameter for determining a code table sequence number after the entropy coding code table jumps; or adjusting element values in the entropy coding code table corresponding to the first block transform according to the mapping parameters to obtain entropy coding code table parameters corresponding to the second block transform after mapping; the mapping parameter and the skip parameter are determined according to the statistical properties of the corresponding second block transform.

22. The apparatus of claim 21, wherein the first code table processing unit is configured to, during the mapping process, determine the skip parameter as a skipped code table sequence number, or determine the skip parameter as a difference between the skipped code table sequence number and a current code table sequence number; and the coding unit is used for performing entropy coding operation on the data of the current or next image block of the second block transform processed by the second transform according to the entropy coding code table parameters corresponding to the second block transform.

23. The apparatus of claim 22, further comprising a first skip parameter obtaining unit for obtaining a preset skip parameter; or, the method is used for acquiring the jump parameter determined according to the coding characteristic of the current coding sequence, the coding image or the coding strip, and informing the coding unit to write the jump parameter into the coding code stream.

24. The apparatus of claim 21, wherein the first code table processing unit performs mapping processing comprising: adjusting element values in an entropy coding code table corresponding to the first block transformation according to mapping parameters corresponding to the second block transformation to obtain entropy coding code table parameters corresponding to the second block transformation; and the coding unit is used for performing entropy coding operation on the data of the image block of the current or next second block transform according to the entropy coding code table parameters corresponding to the second block transform obtained by adjustment.

25. The apparatus according to claim 24, further comprising a first mapping parameter obtaining unit, configured to obtain a preset mapping parameter; or the mapping unit is used for acquiring the mapping parameters of the current coding sequence, the coding image or the coding strip determined according to the coding characteristics of the current coding sequence, the coding image or the coding strip, and informing the coding unit to write the mapping parameters into the code stream.

26. An entropy decoding apparatus, comprising:

the second storage unit is used for storing the entropy decoding code table corresponding to the first block transform;

the second code table processing unit is used for obtaining the entropy decoding code table parameters corresponding to the second block transformation according to the entropy decoding code table mapping corresponding to the first block transformation stored in the second storage unit;

a decoding unit, configured to perform entropy decoding operation on the data of the image block of the second block transform by using the entropy decoding code table parameters corresponding to the second block transform obtained by the second code table processing unit;

the second code table processing unit performs mapping processing, including mapping operation according to a jump parameter for determining a code table sequence number after the jump of the entropy decoding code table; or adjusting the element value in the entropy decoding code table corresponding to the first block transform according to the mapping parameter to obtain the entropy decoding code table parameter corresponding to the second block transform after mapping; the mapping parameter and the skip parameter are determined according to the statistical properties of the corresponding second block transform.

27. The apparatus as claimed in claim 26, wherein the second code table processing unit is configured to, during the mapping process, determine the skip parameter as a skipped code table sequence number, or determine the skip parameter as a difference between the skipped code table sequence number and a current code table sequence number; and the decoding unit is used for carrying out entropy decoding operation on the image under the second block transform which is currently or next processed by the second transform according to the entropy decoding code table parameters corresponding to the second block transform.

28. The apparatus of claim 27, further comprising a second skip parameter obtaining unit for obtaining a preset skip parameter; or, the skip parameter is used for acquiring the current decoding sequence or decoding image or decoding strip from the received coding code stream according to the current decoding sequence or decoding image or decoding strip.

29. The apparatus of claim 26, wherein the processing of the mapping employed by the second code table processing unit comprises: adjusting the data elements corresponding to the second block transform according to the mapping parameters, and obtaining entropy decoding code table parameters corresponding to the second block transform according to the adjusted data element values and the entropy decoding code table corresponding to the first block transform; and the decoding unit is used for carrying out entropy decoding operation on the data of the current or next image block of the second block transform processed by the second transform according to the entropy decoding code table parameters corresponding to the second block transform obtained by adjustment.

30. The apparatus according to claim 29, further comprising a second mapping parameter obtaining unit, configured to obtain a preset mapping parameter; or, the mapping parameter is used for obtaining the current decoding sequence or decoding image or decoding strip from the received coding code stream according to the current decoding sequence or decoding image or decoding strip.

Images