WO2022252222A1

WO2022252222A1 - Encoding method and encoding device

Info

Publication number: WO2022252222A1
Application number: PCT/CN2021/098383
Authority: WO
Inventors: 郑萧桢; 缪泽翔; 李蔚然; 郭泽
Original assignee: 深圳市大疆创新科技有限公司
Priority date: 2021-06-04
Filing date: 2021-06-04
Publication date: 2022-12-08
Also published as: CN116918331A

Abstract

The present application provides an encoding method and an encoding device. The method comprises: obtaining image complexity information of each of n1 tiles in a current frame, the image complexity information comprising a transform coefficient obtained after a pixel value of each tile is subjected to photo core transform processing, and n1 being a positive integer greater than or equal to 2; and updating Quantization Parameters (QPs) of the n1 tiles according to the image complexity information. By updating or calculating, according to the image complexity information, QPs of respective tiles comprised in an image to be encoded, the solution provided in the present application can ensure that encoding efficiency is improved and hardware resource consumption is reduced without sacrificing code rate control accuracy, and can also ensure the flexibility of the QPs to avoid the problem of an uncontrollable output code rate.

Description

Encoding method and encoding device

technical field

The present application relates to the field of encoding and decoding, and more specifically, relates to an encoding method and an encoding device.

Background technique

Joint Photographic Experts Group Extended Range (JPEG XR) is a continuous-tone still image compression algorithm and file format.

Since the final bit rate of the image compressed by the JPEG XR encoder depends on the degree of quantization, the degree of quantization depends on the specified quantization parameter (Quantization Parameter, QP). Among the current rate control algorithms for JPEG XR, some algorithms require multiple encodings. This approach will bring additional complexity and lead to encoding delay, which will affect the running speed of the encoder and is not conducive to real-time encoding. Other algorithms use fixed quantization parameters. However, using fixed quantization parameters will lead to uncontrollable output code rate.

Therefore, how to improve coding efficiency and ensure the flexibility of QP is a problem to be solved.

Contents of the invention

The embodiment of the present application provides a coding method and a coding device, which can ensure the improvement of coding efficiency and reduce the consumption of hardware resources without sacrificing the accuracy of code rate control, and at the same time can ensure the flexibility of quantization parameters to avoid the problem of uncontrollable output code rate .

In the first aspect, an encoding method is provided, including: acquiring image complexity information of each of the n1 tiles in the current frame, the image complexity information including the pixel value of each tile For the transformation coefficients obtained after image kernel transformation processing (PCT processing), n1 is a positive integer greater than or equal to 2; the quantization parameters of the n1 tiles are updated according to the image complexity information.

In the solution provided by the embodiment of the present application, since the image complexity information is related to the transformation coefficient of the tile in the current frame, and the transformation coefficient is the coefficient obtained after performing PCT processing on the pixel value of the tile, by The degree information is updated or the QP of each tile included in the image to be encoded can be guaranteed to improve the encoding efficiency and reduce the consumption of hardware resources without sacrificing the bit rate control accuracy, and at the same time, it can ensure the flexibility of quantization parameters to avoid output The code rate is uncontrollable.

In a second aspect, an encoding method is provided, including: obtaining image complexity information of the current frame, the image complexity information including transformation obtained after image kernel transformation processing (PCT processing) is performed on the pixel values of the current frame coefficient; determine the initial quantization parameter (initial QP) of the current frame according to the image complexity information; update the initial QP of the target frame according to the initial QP of the current frame, and the target frame is the first x of the current frame frame and/or next y frames, where x and y are positive integers greater than or equal to 1.

In the solution provided by the embodiment of the present application, since the image complexity information is related to the transformation coefficient of the current frame, and the transformation coefficient is the coefficient obtained after performing PCT processing on the pixel values in the current frame, it is determined according to the image complexity information The initial QP of the current frame and the initial QP of the target frame are updated according to the initial QP, which can ensure that the encoding efficiency is improved and the hardware resource consumption is reduced without sacrificing the accuracy of the code rate control, and at the same time, the flexibility of quantization parameters can be guaranteed to avoid output The code rate is uncontrollable.

In a third aspect, an encoding device is provided, including: a complexity calculation module, configured to obtain image complexity information of each of the n1 tiles in the current frame, the image complexity information including the The transformation coefficient obtained after the pixel value of each tile undergoes image kernel transformation processing (PCT processing), n1 is a positive integer greater than or equal to 2; the code rate control module is used to update the n1 according to the image complexity information Quantization parameters for tiles.

For the beneficial effects of the third aspect, reference may be made to the beneficial effects of the first aspect, which will not be repeated here.

In a fourth aspect, an encoding device is provided, including: a complexity calculation module, configured to obtain image complexity information of a current frame, where the image complexity information includes performing image kernel transformation processing on pixel values of the current frame ( Transform coefficients obtained after PCT processing); a code rate control module, configured to determine the initial quantization parameter (initial QP) of the current frame according to the image complexity information; the code rate control module is also used to: according to the described image complexity information The initial QP of the current frame updates the initial QP of the target frame, where the target frame is the previous x frames and/or the last y frames of the current frame, and x and y are positive integers greater than or equal to 1.

For the beneficial effects of the fourth aspect, reference may be made to the beneficial effects of the second aspect, which will not be repeated here.

In a fifth aspect, an encoding device is provided, including: a processor, configured to: acquire image complexity information of each of the n1 tiles in the current frame, where the image complexity information includes The pixel values of a tile are transformed coefficients obtained after image kernel transformation processing (PCT processing), n1 is a positive integer greater than or equal to 2; the quantization parameters of the n1 tiles are updated according to the image complexity information.

For the beneficial effects of the fifth aspect, reference may be made to the beneficial effects of the first aspect, which will not be repeated here.

In a sixth aspect, an encoding device is provided, including: a processor, configured to: obtain image complexity information of a current frame, where the image complexity information includes performing image kernel transformation processing (PCT) on pixel values of the current frame transform coefficient obtained after processing); determine the initial quantization parameter (initial QP) of the current frame according to the image complexity information; update the initial QP of the target frame according to the initial QP of the current frame, and the target frame is the The previous x frame and/or the next y frame of the current frame, where x and y are positive integers greater than or equal to 1.

For the beneficial effects of the sixth aspect, reference may be made to the beneficial effects of the second aspect, which will not be repeated here.

In a seventh aspect, an encoding device is provided, including a processor and a memory. The memory is used to store a computer program, and the processor is used to call and run the computer program stored in the memory to execute the method in the first aspect or the second aspect or each implementation thereof.

In an eighth aspect, a chip is provided for implementing the method in the above first aspect or the second aspect or each implementation manner thereof.

Specifically, the chip includes: a processor, configured to call and run a computer program from the memory, so that the device installed with the chip executes the method in the above first aspect or the second aspect or each implementation thereof.

A ninth aspect provides a computer-readable storage medium for storing a computer program, and the computer program includes any possible implementation manners for executing the first aspect to the second aspect or the first aspect to the second aspect. method directive.

In a tenth aspect, a computer program product is provided, including computer program instructions, the computer program instructions causing a computer to execute the methods in the implementation manners of the first aspect to the second aspect or the first aspect to the second aspect.

Description of drawings

The drawings used in the embodiments will be briefly introduced below.

FIG. 1 is a structural diagram of a technical solution applying an embodiment of the present application.

Fig. 2 is a schematic diagram of a video coding framework 2 according to an embodiment of the present application.

Fig. 3 is a schematic diagram of JPEG XR processing images according to five levels from large to small when processing images provided by the embodiment of the present application.

Fig. 4 is a schematic structural diagram of a JPEG XR encoder provided by an embodiment of the present application.

FIG. 5 is a schematic diagram of forming transform coefficients based on macroblocks according to an embodiment of the present application.

Fig. 6 is a schematic diagram of an encoding method provided by an embodiment of the present application.

Fig. 7a is a schematic diagram of division of an image to be coded provided by an embodiment of the present application.

Fig. 7b is a schematic diagram of division of an image to be coded according to another embodiment of the present application.

Fig. 7c is a schematic diagram of division of an image to be coded according to yet another embodiment of the present application.

Fig. 7d is a schematic diagram of division of an image to be encoded according to yet another embodiment of the present application.

Fig. 7e is a schematic diagram of division of an image to be encoded according to yet another embodiment of the present application.

Fig. 7f is a schematic diagram of division of an image to be coded according to yet another embodiment of the present application.

FIG. 8 is a schematic diagram of an implemented matrix position conversion function provided by an embodiment of the present application.

FIG. 9 is a schematic diagram of a mapping relationship of blocks provided by an embodiment of the present application.

FIG. 10 is a schematic diagram of an encoding method provided by another embodiment of the present application.

Fig. 11 is a schematic diagram of an encoding method provided by another embodiment of the present application.

Fig. 12 is a schematic structural diagram of an encoding device provided by an embodiment of the present application.

Fig. 13a is a schematic structural diagram of a JPEG XR encoder provided by another embodiment of the present application.

Fig. 13b is a schematic structural diagram of a JPEG XR encoder provided by another embodiment of the present application.

Fig. 14 is a schematic structural diagram of an encoding device provided by another embodiment of the present application.

FIG. 15 is a schematic structural diagram of an encoding device provided by another embodiment of the present application.

FIG. 16 is a schematic structural diagram of an encoding device provided in yet another embodiment of the present application.

FIG. 17 is a schematic structural diagram of a chip provided by an embodiment of the present application.

Detailed ways

The technical solutions in the embodiments of the present application are described below.

Unless otherwise specified, all technical and scientific terms used in the embodiments of the present application have the same meaning as commonly understood by those skilled in the technical field of the present application. The terms used in the present application are only for the purpose of describing specific embodiments, and are not intended to limit the scope of the present application.

As shown in FIG. 1 , the system 100 may receive data to be processed 102 , process the data to be processed 102 , and generate processed data 108 . For example, the system 100 may receive data to be encoded and encode the data to be encoded to generate encoded data, or the system 100 may receive data to be decoded and decode the data to be decoded to generate decoded data. In some embodiments, the components in system 100 may be implemented by one or more processors, which may be processors in computing devices, or processors in mobile devices (such as drones). The processor may be any type of processor, which is not limited in this embodiment of the present invention. In some possible designs, the processor may include an encoder, decoder, or codec, among others. One or more memories may also be included in system 100 . The memory may be used to store instructions and data, for example, computer-executable instructions for implementing the technical solutions of the embodiments of the present invention, data to be processed 102, processed data 108, and the like. The storage may be any type of storage, which is not limited in this embodiment of the present invention.

Data to be encoded may include text, images, graphic objects, animation sequences, audio, video, or any other data that requires encoding. In some cases, the data to be encoded may include sensory data from sensors such as vision sensors (e.g., cameras, infrared sensors), microphones, near-field sensors (e.g., ultrasonic sensors, radar), position sensors, temperature sensors, touch sensors, etc. In some cases, the data to be encoded may include information from the user, for example, biometric information, which may include facial features, fingerprint scans, retinal scans, voice recordings, DNA samples, and the like.

Fig. 2 is a schematic diagram of a video coding framework 2 according to an embodiment of the present application. As shown in FIG. 2 , after the video to be encoded is received, starting from the first frame of the video to be encoded, each frame in the video to be encoded is sequentially encoded. Among them, the current coded frame mainly undergoes: Prediction, Transform, Quantization and Entropy Coding, etc., and finally outputs the code stream of the current coded frame. Correspondingly, the decoding process usually decodes the received code stream according to the reverse process of the above process, so as to recover the video frame information before decoding.

Specifically, as shown in FIG. 2 , the video encoding framework 2 includes an encoding control module 201 for performing decision-making control actions and parameter selection during the encoding process. For example, as shown in FIG. 2, the encoding control module 201 controls parameters used in transformation, quantization, inverse quantization, and inverse transformation, controls the selection of intra-frame mode or inter-frame mode, and parameter control of motion estimation and filtering, And the control parameters of the encoding control module 201 will also be input into the entropy encoding module, and encoded to form a part of the encoded code stream.

When coding the current coded frame starts, the coded frame is divided 202 , specifically, it is first divided into slices and then divided into blocks. Optionally, in an example, the coded frame is divided into a plurality of non-overlapping largest CTUs, and each CTU can also be iteratively divided into a series of smaller coded Unit (Coding Unit, CU), in some examples, CU can also contain associated prediction unit (Prediction Unit, PU) and transformation unit (Transform Unit, TU), where PU is the basic unit of prediction, and TU is transformation and the basic unit of quantification. In some examples, a PU and a TU are respectively obtained by dividing a CU into one or more blocks, and one PU includes multiple prediction blocks (Prediction Block, PB) and related syntax elements. In some examples, the PU and the TU may be the same, or they may be obtained by dividing the CU through different methods. In some examples, at least two of CU, PU, and TU are the same. For example, CU, PU, and TU are not distinguished, and prediction, quantization, and transformation are all performed in units of CU. For the convenience of description, hereinafter, a CTU, a CU or other formed data units are all referred to as coding blocks.

It should be understood that, in this embodiment of the present application, the data unit targeted by video encoding may be a frame, a slice, a coding tree unit, a coding unit, a coding block, or a group of any of the above. In different embodiments, the size of the data unit may vary.

Specifically, as shown in FIG. 2 , after the coded frame is divided into multiple coded blocks, a prediction process is performed to remove redundant information in the spatial domain and time domain of the current coded frame. Currently, commonly used predictive coding methods include intra-frame prediction and inter-frame prediction. Intra-frame prediction only uses the reconstructed information in this frame image to predict the current coding block, while inter-frame prediction uses the information in other frame images (also called reference frames) that have been reconstructed before to predict the current coding block. Make predictions. Specifically, in the embodiment of the present application, the encoding control module 201 is used to decide to select intra prediction or inter prediction.

When the intra-frame prediction mode is selected, the process of intra-frame prediction 203 includes obtaining the reconstructed block of the coded adjacent block around the current coded block as a reference block, based on the pixel value of the reference block, using the prediction mode method to calculate the prediction value to generate a prediction block , subtracting the corresponding pixel values of the current coding block and the prediction block to obtain the residual of the current coding block, and the residual of the current coding block undergoes transformation 204 , quantization 205 and entropy coding 210 to form a code stream of the current coding block. Further, all the coded blocks of the current coded frame form a part of the coded code stream of the coded frame after undergoing the above coded process. In addition, the control and reference data generated in intra-frame prediction 203 are also encoded by entropy encoding 210 to form a part of the encoded code stream.

Specifically, the transform 204 is used to remove the correlation of the residual of the image block, so as to improve the coding efficiency. For the transformation of the residual data of the current coding block, two-dimensional discrete cosine transform (Discrete Cosine Transform, DCT) transformation and two-dimensional discrete sine transform (Discrete Sine Transform, DST) transformation are usually used. Multiply with an N×M transformation matrix and its transpose matrix respectively, and obtain the transformation coefficient of the current coding block after multiplication.

After the transform coefficients are generated, quantization 205 is used to further improve the compression efficiency. The quantized coefficients can be obtained after the transform coefficients are quantized, and then the quantized coefficients are subjected to entropy encoding 210 to obtain the residual code stream of the current encoding block, wherein the entropy encoding method includes But not limited to Context Adaptive Binary Arithmetic Coding (CABAC) entropy coding. Finally, the bit stream obtained by entropy encoding and the encoded encoding mode information are stored or sent to the decoding end. At the encoding end, inverse quantization 206 is performed on the quantized result, and inverse transformation 207 is performed on the dequantized result. After the inverse transformation 207, the reconstructed pixels are obtained using the inverse transformation result and the motion compensation result. Afterwards, the reconstructed pixels are filtered (ie loop filtered) 211 . After 211, the filtered reconstructed image (belonging to the reconstructed video frame) is output. Subsequently, the reconstructed image can be used as a reference frame image of other frame images for inter-frame prediction. In this embodiment of the present application, the reconstructed image may also be referred to as a reconstructed image or a reconstructed image.

Specifically, the coded adjacent blocks in the process of intra prediction 203 are: the coded neighboring blocks before the current coded block is coded, and the residual generated during the coded process of the neighboring blocks is transformed 204, quantized 205, After inverse quantization 206 and inverse transformation 207, the reconstructed block is added to the prediction block of the adjacent block. Correspondingly, inverse quantization 206 and inverse transformation 207 are inverse processes of quantization 206 and transformation 204, and are used to recover residual data before quantization and transformation.

As shown in FIG. 2, when the inter prediction mode is selected, the inter prediction process includes motion estimation (Motion Estimation, ME) 208 and motion compensation (Motion Compensation, MC) 209. Specifically, the encoder can perform motion estimation 208 according to the reference frame images in the reconstructed video frame, and search for the image block most similar to the current encoding block in one or more reference frame images according to a certain matching criterion as the prediction block, The relative displacement between the prediction block and the current coding block is the motion vector (Motion Vector, MV) of the current coding block. And subtracting the original value of the pixel of the coding block from the pixel value of the corresponding prediction block to obtain the residual of the coding block. After transformation 204 , quantization 205 and entropy coding 210 , the residual of the current coding block forms a part of the coded code stream of the coded frame. For the decoding end, motion compensation 209 may be performed based on the determined motion vector and the predicted block to obtain the current coding block.

Wherein, as shown in FIG. 2 , the reconstructed video frame is obtained after filtering 211 . A reconstructed video frame includes one or more reconstructed images. Filtering 211 is used to reduce compression distortion such as block effect and ringing effect generated during the encoding process. The reconstructed video frame is used to provide reference frames for inter-frame prediction during the encoding process. During the decoding process, the reconstructed video frame is output after post-processing for the final decoded video.

Specifically, the inter-frame prediction mode may include an advanced motion vector prediction (Advanced Motion Vector Prediction, AMVP) mode, a merge (Merge) mode or a skip (skip) mode.

For the AMVP mode, the motion vector prediction (Motion Vector Prediction, MVP) can be determined first. After the MVP is obtained, the starting point of the motion estimation can be determined according to the MVP, and the motion search is performed near the starting point. After the search is completed, the optimal MV, the position of the reference block in the reference image is determined by the MV, the reference block is subtracted from the current block to obtain the residual block, the MV is subtracted from the MVP to obtain the motion vector difference (Motion Vector Difference, MVD), and the MVD and MVP The index is transmitted to the decoder through the code stream.

For the Merge mode, the MVP can be determined first, and the MVP can be directly determined as the MV of the current block. Among them, in order to obtain the MVP, an MVP candidate list (merge candidate list) can be constructed first. In the MVP candidate list, at least one candidate MVP can be included, and each candidate MVP can have an index corresponding to it. After the MVP is selected, the MVP index can be written into the code stream, and the decoder can find the MVP corresponding to the index from the MVP candidate list according to the index, so as to decode the image block.

It should be understood that the above process is only a specific implementation of the Merge mode. The Merge pattern can also have other implementations.

For example, Skip mode is a special case of Merge mode. After the MV is obtained according to the Merge mode, if the encoder determines that the current block is basically the same as the reference block, then there is no need to transmit the residual data, only the index of the MVP, and a flag that can indicate that the current block can be directly Obtained from the reference block.

That is to say, the feature of the Merge mode is: MV=MVP (MVD=0); and the Skip mode has one more feature, namely: the reconstructed value rec=predicted value pred (residual value resi=0).

The Merge mode can be applied to geometric forecasting techniques. In the geometric prediction technology, the image block to be coded can be divided into multiple sub-image blocks shaped as polygons, and the motion vector can be determined for each sub-image block from the motion information candidate list, and based on the The motion vector determines the prediction sub-block corresponding to each sub-image block, and constructs the prediction block of the current image block based on the prediction sub-block corresponding to each sub-image block, so as to realize the coding of the current image block.

For the decoding end, perform operations corresponding to the encoding end. First, entropy decoding, inverse quantization and inverse transformation are used to obtain residual information, and it is determined whether the current image block uses intra prediction or inter prediction according to the decoded code stream. If it is intra-frame prediction, use the reconstructed image block in the current frame to construct prediction information according to the intra-frame prediction method; if it is inter-frame prediction, you need to parse out the motion information, and use the parsed motion information in the reconstructed image The reference block is determined to obtain the prediction information; next, the prediction information and the residual information are superimposed, and the reconstruction information can be obtained after filtering.

As mentioned above, encoding video based on the video encoding framework 2 shown in FIG. 2 can save space or traffic occupied by video image storage and transmission. In general, the uncompressed original image data collected by the camera occupies a large storage space. Take the image with a resolution of 3840×2160 and a storage format of YUV4:2:210-bit as an example. Storing the image requires about 20M bytes of storage space. Usually, an 8G memory card can only store 500 uncompressed photos of the above specifications. 20M bytes of traffic is required. Therefore, in order to save space or traffic occupied by image storage and transmission, image data needs to be encoded and compressed.

Joint Photographic Experts Group Extended Range (JPEG XR) is a continuous tone still image compression algorithm and file format, also known as HD Photo or network media image (Windows Media Photo), developed by Microsoft (microsoft ) development, is part of the network media (windows media) family. It supports lossy as well as lossless data compression, and is the preferred image format for Microsoft's XML Paper Specification (XPS) documents. Among them, XML is Extensible Markup Language (Extensible Markup Language). Currently supported software includes .NET framework (3.0 or later), operating system (windows vista/windows 7), Internet Explorer (IE) 9, animation player (flashplayer) 11, etc.

JPEG XR is an image codec that can realize high dynamic range image encoding, and only requires integer operations during compression and decompression. It can support monochrome, Red Green Blue (RGB), Cyan Magenta Yellow Black (CMYK), 16-bit unsigned integer or 32-bit fixed-point or floating-point multi-channel color format images, and it can also support RGBE Radiance images Format. It can optionally embed the International Color Consortium (ICC) color profile for color consistency across devices. The alpha channel can indicate the degree of transparency, and supports Exchangeable Image File (EXIF) and Extensible Metadata Platform (XMP) metadata formats. This format also supports multiple images in one file. It supports only partial decoding of the image, and there is no need to decode the entire image for some specific operations such as cropping, downsampling, horizontal and vertical flipping, or rotation.

As shown in Figure 3, it is a schematic diagram of processing images in five levels from large to small when JPEG XR processes images. Wherein, the figure includes an image (image), a tile (tile), a macro block (macro block), a block (block), and a pixel (pixel). One of the images can consist of one or more tiles. If the tile is on the right and bottom edge of the image, it will be padded to an integer number of macroblocks (16×16). Each macroblock may contain 16 4x4 blocks, and each block may contain 4x4 pixels. JPEG XR performs a two-stage transform on each 4×4 block and the recombined low-pass block in the 16×16 macroblock.

As shown in Figure 4, it is a schematic structural diagram of a JPEG XR encoder provided by an embodiment of the present application. The JPEG XR encoder may include five modules: a filtering module 410, a transform module 420, a quantization module 430, a prediction module 440, and an entropy encoding module 450. The functions of these five modules are similar to those of the modules involved in the above-mentioned FIG. 2 similar. Specifically, the filtering module 410 can reduce the block effect of the decoded and reconstructed image through smoothing between adjacent pixels; the transform module 420 can convert the image information from the spatial domain to the frequency domain, and remove part of the redundant information in the spatial domain; the quantization module 430 can convert the frequency domain The coefficients are shrunk to reduce the magnitude of the coefficients that need to be coded. The degree of reduction of the coefficients depends on the size of the specified quantization parameter (Quantization Parameter, QP); the prediction module 440 can remove the adjacent blocks through the prediction between the coefficients of the adjacent blocks The correlation between partial coefficients; the entropy encoding module 450 can encode the finally obtained coefficients into a binary code stream.

From the functional analysis of the above five modules and the description of these modules in Figure 2 above, it can be seen that the size of the final code stream (that is, the code rate) mainly depends on the degree of quantization, prediction efficiency and entropy coding performance. Among them, quantization degree is decisive.

The following section first introduces the transformation module and quantization module of JPEG XR.

1. Transformation module

The transformation of JPEG XR is based on an integer transformation, and each macroblock can participate in two stages of transformation. Transforms can all be performed on a 4x4 block basis. As shown in Figure 5, the first-stage transformation can be applied to 16 blocks within a macroblock, resulting in 16 low-pass coefficients (Low Pass Coefficient, LP Coefficient) and 240 high-pass (High Pass Coefficient, HP Coefficient) coefficients, namely Each of these 16 blocks produces one LP coefficient and 15 HP coefficients. The second-stage transformation can be applied to the recombined block of 16 LP coefficients obtained in the first stage, and finally generate 1 DC coefficient (Direct Current Coefficient, DC Coefficient) and 15 LP coefficients through re-transformation.

2. Quantization module

Quantization in JPEG XR is highly flexible, since quantization parameters may vary among tiles, macroblocks and DC coefficients, LP coefficients, HP coefficients. The quantization parameter range of JPEG XR is an integer from 0 to 255. When the quantization parameter is 0 and 1, it is lossless compression, and when the quantization parameter is 255, it is the most lossy compression. The mapping relationship from quantization parameters to scaling factors (Scale Factor, SF) is shown in the following formula (1):

The quantized coefficients are obtained by dividing the original coefficients by the corresponding scaling factor for the chosen quantization parameter and then rounding them to integers.

It can be seen that the final size of the code rate of the compressed image using the JPEG XR encoder depends on the degree of quantization, and the degree of quantization depends on the specified quantization parameters. Among the current rate control algorithms for JPEG XR, some algorithms require multiple encodings. This approach will bring additional complexity and lead to encoding delay, which will affect the running speed of the encoder and is not conducive to real-time encoding. Other algorithms use fixed quantization parameters. However, using fixed quantization parameters will lead to uncontrollable output code rate.

In view of the above problems, this application proposes a coding method, which can improve the coding efficiency and reduce the consumption of hardware resources without sacrificing the accuracy of code rate control, and at the same time can ensure the flexibility of quantization parameters to avoid uncontrollable output code rate The problem.

As shown in FIG. 6 , it is an encoding method 600 provided in the embodiment of the present application, and the encoding method 600 may include steps 610-620.

610. Obtain the image complexity information of each of the n1 tiles in the current frame, where the image complexity information includes performing image core transformation (Photo Core Transform, PCT) on the pixel value of each tile Transform coefficient obtained after processing, n1 is a positive integer greater than or equal to 2.

In the embodiment of the present application, when dividing the image to be encoded, the image to be encoded may be divided according to a fixed width or a fixed height, or may not be divided according to a fixed width or a fixed height.

In addition, in the embodiment of the present application, the coded image may be divided horizontally, or the image to be coded may be divided vertically.

It should be understood that the horizontal division in the embodiment of the present application may refer to the division of the image to be coded in the horizontal direction, and the vertical division may refer to the division of the image to be coded in the vertical direction.

For example, as shown in FIG. 7a , it is a schematic diagram of division of an image to be encoded provided by an embodiment of the present application. The image to be encoded can be divided vertically according to the preset fixed width, assuming that the preset fixed width is 384, as can be seen from Figure 7a, the image to be encoded can be divided into 3 tiles, and the width of these 3 tiles Both are the same, both are 384.

As shown in FIG. 7 b , it is a schematic diagram of division of an image to be encoded provided by another embodiment of the present application. Similarly, the image to be encoded can be divided vertically according to the preset fixed width, assuming that the preset fixed width is 384, as can be seen from Figure 7b, the image to be encoded can be divided into 3 tiles, and tile 1 and Tile 2 has the same width, both 384, while Tile 3 has a width less than 384.

It can be understood that, according to the above-mentioned division methods in FIG. 7a and FIG. 7b, the divided tiles (ie, tile 1, tile 2, and tile 3) have the same height, which is the height of the image to be encoded.

As shown in FIG. 7c, it is a schematic diagram of division of an image to be coded provided by another embodiment of the present application. The image to be coded can be horizontally divided according to the preset fixed height, assuming the preset fixed height is 384, as can be seen from Figure 7c, the image to be coded can be divided into 3 tiles, and the height of these 3 tiles Both are the same, both are 384.

As shown in FIG. 7d , it is a schematic diagram of division of an image to be encoded provided by yet another embodiment of the present application. The image to be coded can be horizontally divided according to the preset fixed height, assuming that the preset fixed height is 384, as can be seen from Figure 7d, the image to be coded can be divided into 3 tiles, and tile 1 and tile 2 The heights of tiles are the same, both are 384, and the height of tile 3 is less than 384.

It can be understood that, according to the above division manners in FIG. 7c and FIG. 7d , the widths of the divided tiles are the same, which is the width of the image to be encoded.

In some embodiments, the image to be coded may not be divided according to a fixed width or a fixed height, but may be divided according to multiple preset widths or multiple heights.

As shown in FIG. 7e , it is a schematic diagram of division of an image to be coded provided by yet another embodiment of the present application. The image to be coded can be divided vertically according to multiple preset fixed widths. Assuming that the preset fixed widths include 160 and 384, it can be seen from Figure 7e that the width of tile 1 is 160, and the width of tile 2 is 384, tile 3 has a width of 224.

As shown in FIG. 7f , it is a schematic diagram of division of an image to be coded provided by yet another embodiment of the present application. The image to be coded can be horizontally divided according to multiple preset fixed heights. Assuming that the multiple preset fixed heights are 160 and 384, it can be seen from Figure 7f that the height of tile 1 is 160, and the height of tile 2 The height is 384 and the height of tile 3 is 224.

It should be understood that the above numerical values are for illustration only, and may also be other numerical values, which are not specifically limited in the present application.

It should be noted that, in some embodiments, when the image to be coded is divided according to a fixed width or a fixed height, the setting of the fixed width or fixed height is related to the number of pixels included in the PCT-processed block, specifically See below for a description.

In some embodiments, the calculation of the image complexity information may also be replaced by operators including but not limited to Hadamard transform or mean square error.

520. Update quantization parameters of the n1 tiles according to the image complexity information.

The image complexity information in the embodiment of the present application may refer to the transform coefficients of the tiles included in the image to be encoded, and the QP of the tiles is updated according to the transform coefficients of the tiles.

As pointed out above, the image complexity information includes the transformation coefficients obtained after PCT processing is performed on the pixel values of each tile, and the PCT processing involved will be described below.

The PCT processing may include the following flow, wherein, the 4*4 blocks included in each macroblock may be transformed according to the following flow.

_2×2T_h(a, b, c, d, flag), _T_odd(a, b, c, d), _T_odd_odd(a, b, c, d) and _FwdPermute(a, b, c, d, e, f, g, h, i, j, k, l, m, n, o, p) represent 4 different calculations, and the letters in brackets represent the input and output of the calculation.

PCT4×4 (a, b, c, d, e, f, g, h, i, j, k, l, m, n, o, p)

_2×2T_h(a, d, m, p, 0)

_2×2T_h(f, g, j, k, 0)

_2×2T_h(b,c,n,o,0)

_2×2T_h(e, h, i, l, 0)

_2×2T_h(a, b, e, f, 1)

_T_odd(c,d,g,h)

_T_odd(i, m, j, n)

_T_odd_odd (k, l, o, p)

_FwdPermute(a, b, c, d, e, f, g, h, i, j, k, l, m, n, o, p)

1), the calculation process of _2×2T_h(a, b, c, d, flag) is shown in formula (2) - formula (9)

a=a+d (2)

b=b-c (3)

t1＝((a-b+flag)＞＞1) (4)

t2=c (5)

c=t1-d (6)

d=t1-t2 (7)

a=a-d (8)

b=b+c (9)

Wherein, t1 and t2 in the above formula are temporary values, and the >> symbol represents a right shift operation.

2), the calculation process of _T_odd(a, b, c, d) is shown in formula (10) - formula (21)

b=b-c (10)

a=a+d (11)

c＝c+((b+1)＞＞1) (12)

d=((a+1)>>1)-d (13)

b＝b-((a*3+4)＞＞3) (14)

a＝a+((b*3+4)＞＞3) (15)

d＝d-((c*3+4)＞＞3) (16)

c＝c+((d*3+4)＞＞3) (17)

d=d+(b>>1) (18)

c＝c-((a+1)＞＞1) (19)

b=b-d (20)

a=a+c (21)

3), the calculation process of _T_odd_odd(a, b, c, d) is shown in formula (22) - formula (36)

b＝-1*b (22)

c＝-1*c (23)

d=d+a (24)

c=c-b (25)

t1＝d＞＞1 (26)

t2＝c＞＞1 (27)

a＝a-t1 (28)

b=b+t2 (29)

a＝a+((b*3+4)＞＞3) (30)

b＝b-((a*3+4)＞＞2) (31)

a＝a+((b*3+3)＞＞3) (32)

b=b-t2 (33)

a＝a+t1 (34)

c=c+b (35)

d=d-a (36)

4), the calculation process of _FwdPermute(a, b, c, d, e, f, g, h, i, j, k, l, m, n, o, p) is shown in Figure 8

This process realizes a matrix position conversion function, that is, the position of the letter of the left figure shown in FIG. 8 is converted to the position of the letter of the right figure shown in FIG. 8 .

Exemplarily, the above-mentioned PCT processing flow is described by taking simple numerical values as an example, assuming that the above-mentioned a, b, c, d, e, f, g, h, i, j, k, l, m, n, o , the values of p are respectively 1, 2, 3, 4, 2, 3, 5, 4, 2, 3, 4, 0, 3, 2, 1, 1, and PCT processing is performed on these values according to the above procedure.

1), carry out _2×2T_h (a, b, c, d, flag) processing

①. The process of _2×2T_h(a, d, m, p, 0) is as follows

a=a+p=1+1=2

d=d-m=4-3=1

t1=((a-d+flag)>>1)=0

t2=m=3

m=t1-p=-1

p=t1-t2=0-3=-3

a=a-p=2-(-3)=5

d=d+m=1+(-1)=0

②. The process of _2×2T_h(f, g, j, k, 0) is as follows

f=f+k=3+4=7

g=g-j=5-3=2

t1=((f-g+flag)>>1)=2

t2=j=3

j=t1-k=2-4=-2

k=t1-t2=2-3=-1

f=f-k=7-(-1)=8

g=g+j=2-(-2)=0

③. The process of _2×2T_h(b, c, n, o, 0) is as follows

b=b+o=2+1=3

c=c-n=3-2=1

t1=((b-c+flag)>>1)=1

t2=n=2

n=t1-o=1-1=0

o=t1-t2=1-2=-1

b=b-o=3-(-1)=4

c=c+n=1+0=1

④ The process of _2×2T_h(e, h, i, l, 0) is as follows

e=e+l=2+0=2

h=h-i=4-2=2

t1=((e-h+flag)>>1)=0

t2=i=2

i=t1-l=0-0=0

l=t1-t2=0-2=-2

e=e-l=2-(-2)=4

h=h+i=2+0=2

⑤ The process of _2×2T_h(a, b, e, f, 1) is as follows

a=a+f=5+8=13

b=b-e=4-4=0

t1=((a-b+flag)>>1)=7

t2=e=4

e=t1-f=7-8=-1

f=t1-t2=7-4=3

a=a-f=13-3=10

b=b+e=0+(-1)=-1

Based on this, the values of a, b, c, d, e, f, g, h, i, j, k, l, m, n, o, p are updated to 10, -1, 1, 0, - 1, 3, 0, 2, 0, -1, -1, -2, -1, 0, -1, -3.

2), carry out the processing of _T_odd (a, b, c, d)

①. The process of _T_odd(c, d, g, h) is as follows

It can be seen from the above that the values of c, d, g, and h are 1, 0, 0, and 2, respectively.

d=d-g=0-0=0

c=c+h=1+2=3

g=g+((d+1)>>1)=0

h=((c+1)>>1)-h=0

d=d-((c*3+4)>>3)=0

c=c+((d*3+4)>>3)=3

h=h-((g*3+4)>>3)=0

g=g+((h*3+4)>>3)=0

h=h+(d>>1)=0

g=g-((c+1)>>1)=-2

d=d-h=0-0=0

c=c+g=3+(-2)=1

Therefore, the updated values of c, d, g, and h are 1, 0, -2, 0 respectively.

②. The process of _T_odd(i, m, j, n) is as follows

It can be seen from the above that the values of i, m, j, and n are 0, -1, -1, and 0, respectively.

m=m-j=-1-(-1)=0

i=i+n=0+0=0

j=j+((m+1)>>1)=-1

n=((i+1)>>1)-n=0

m=m-((i*3+4)>>3)=0

i=i+((m*3+4)>>3)=0

n=n-((j*3+4)>>3)=0

j=j+((n*3+4)>>3)=-1

n=n+(m>>1)=0

j=j-((i+1)>>1)=-1

m=m-n=0

i=i+j=-1

Therefore, the updated values of i, m, j, and n are -1, 0, -1, 0 respectively.

3), carry out _T_odd_odd (k, l, o, p) processing

It can be seen from the above that the values of k, l, o, p are -1, 2, -1, -3 respectively.

l=-1*l=2

o=-1*o=1

p=p+k=(-3)+(-1)=-4

o=o-l=1-2=-1

t1=p>>1=2

t2=o>>1=0

k=k-t1=(-1)-2=-3

l=l+t2=2+0=2

k=k+((l*3+4)>>3)=-2

l=l-((k*3+4)>>2)=1

k=k+((l*3+3)>>3)=-2

l=l-t2=1-0=1

k=k+t1=(-2)+2=0

o=o+l=(-1)+1=0

p=p-k=(-4)-0=-4

Therefore, the updated values of k, l, o, p are 0, 1, 0, -4 respectively.

Based on this, the updated values of a, b, c, d, e, f, g, h, i, j, k, l, m, n, o, p are 10, -1, 1, 0 respectively , -1, 3, -2, 0, -1, -1, 0, 1, 0, 0, 0, -4.

4), carry out _FwdPermute (a, b, c, d, e, f, g, h, i, j, k, l, m, n, o, p) processing

As shown in Figure 8, matrix transformation processing is performed on the input values, and finally updated a, b, c, d, e, f, g, h, i, j, k, l, m, n, o, The values of p are 10, -1, -1, 0, 1, -4, 0, 0, -1, 0, 3, -1, 0, 1, -2, 0.

It should be understood that the above process is only illustrated with simple numerical values. In an actual encoding process, the pixel values in the image to be encoded may be much larger than the numerical values listed above, but the calculation process is essentially the same.

Optionally, in some embodiments, the updating the QP of the n1 tiles according to the image complexity information includes: calculating the i-th of the n1 tiles according to the image complexity information The cumulative value of the target number of bytes of the tile, i is a positive integer less than or equal to n1; the QP of the i-th tile is updated according to the cumulative value of the target number of bytes of the i-th tile.

The cumulative value of the target number of bytes of the i-th tile in this embodiment of the present application may refer to the target number of bytes of all tiles in the previous i-th tile (including the i-th tile).

For example, the cumulative value of the target byte count of the second tile may refer to the sum of the target byte count of the first tile and the target byte count of the second tile; the target byte count of the fifth tile The cumulative value of the number of sections can refer to the target byte count of the first tile, the target byte count of the second tile, the target byte count of the third tile, and the target byte count of the fourth tile and the sum of the target byte count of the fifth tile; and so on, the cumulative value of the target byte count of the n1th tile can refer to the target byte count of the first tile, the second tile The sum of the target number of bytes of the slice..., the target number of bytes of the n1-1th tile, and the target number of bytes of the n1th tile.

In the embodiment of the present application, the QP of the i-th tile may be updated according to the accumulated value of the target byte number of the i-th tile. For example, the QP of the first tile can be updated according to the cumulative value of the target byte count of the first tile (that is, the target byte count of the first tile), and according to the target byte count of the third tile The cumulative value of updates the QP of the third tile.

It should be noted that, for the first tile, there may be an initial QP, and then the initial QP of the first tile is updated according to the target number of bytes of the first tile.

As mentioned above, the image complexity information in the embodiment of the present application includes the transformation coefficient obtained after PCT processing is performed on the pixel value of each tile. This is because, since the image complexity information calculated by using the transformation form of the JPEG XR transformation module has a strong correlation with the size of the final code stream, the image complexity information used in the embodiment of this application can be processed by the above-mentioned PCT Calculated by accumulating transform coefficients.

Wherein, the acquisition of the image complexity information of each of the n1 tiles in the current frame includes:

After the blocks in the tile are mapped, preset transformation is performed to generate preset transformation parameters, and the image complexity information of the tile is generated according to the preset transformation parameters. The preset transformation includes PCT transformation

It is worth noting that, in order to reduce the boundary conditions of the hardware, in some embodiments, if the width of the image to be encoded is not an integer multiple of the fixed width, the complexity of pixels in the tile whose width on the rightmost side is less than the fixed width may not be calculated , or, if the height of the image to be encoded is not an integer multiple of the fixed height, then the complexity of the pixels in the tile whose lowermost height is less than the fixed height may not be calculated.

In some other embodiments, if the width of the image to be encoded is not an integer multiple of the fixed width, the image complexity of the rightmost tile of the image to be encoded can be calculated through the proportional relationship, or, if the height of the image to be encoded is not fixed Integer multiples of the height, the image complexity of the bottom tile of the image to be encoded can be calculated through the proportional relationship.

As mentioned above, the input processed by PCT can include 4*4 pixel blocks. Considering the saving of hardware line cache resources in System On Chip (SOC) implementation, 2x8 blocks can be mapped to 4x4 blocks for input . This operation can save 2 lines of cache resources.

As shown in FIG. 9 , it is a schematic diagram of a block mapping relationship provided by the embodiment of the present application. Wherein, (a) in FIG. 9 is a schematic diagram of a block before mapping, and (b) in FIG. 9 is a schematic diagram of a block after mapping.

This mapping relationship can be understood as moving the last four pixels of the first row to the position below the second row, that is, the position of the third row, and moving the last four pixels of the second row to the bottom of the third row, that is, the position of the fourth row. row position.

It can be understood that the above operations only involve the change of the position of the pixel, but not the change of the value of the pixel.

In addition, it is also pointed out above that in some embodiments, when the image to be coded is divided according to a fixed width or a fixed height, the setting of the fixed width or fixed height is related to the number of pixels included in the PCT-processed block.

Since the above mapping relationship is to map a 2x8 block to a 4x4 block for input, that is, the block before mapping includes 2 rows and 8 columns of pixels. Therefore, when setting the above fixed width, it can be set to a value that is an integer multiple of 8. In this way, when mapping the rightmost block of the tile, it can just be mapped as a 4*4 block.

Similarly, when setting the above-mentioned fixed height, it can be set to a value that is an integer multiple of 2, so that when mapping the bottom block of the tile, it can just be mapped to a 4*4 block.

In the above description, when the image to be coded is divided vertically according to a fixed width or horizontally divided according to a fixed height, the width of the rightmost tile may not meet the fixed width, or the height of the bottommost tile may not be Satisfying a fixed height, when calculating the image complexity value of the rightmost tile or the bottommost tile, it can be calculated in the following way.

Taking the vertical division of the image to be coded according to a fixed width as an example, if the image to be coded is divided into n1 tiles, the width of the rightmost tile (that is, the n1th tile) may be less than 384. Therefore, the image complexity value of the 1st to n1-1th tiles can be calculated first, and then the image complexity value of the n1th tile can be calculated according to the proportional relationship.

After mapping all 2x8 blocks in tiles 1 to n1-1 and completing PCT processing, the absolute values of all DC coefficients in the obtained tiles and coefficients other than DC coefficients (called LPHP coefficients) can be absolute The values are accumulated as the image complexity value of the tile, as shown in formula (37) and formula (38), where i=1, 2, 3, . . . , n1-1.

complexityDC _i ＝∑ _{2x8 blocks in tile} DC coefficient (37)

complexityLPHP _i ＝∑ _{2x8 blocks in tile} LPHP coefficient (38)

Among them, complexityDC _i represents the image complexity value of the DC coefficient of the i-th tile, ∑ _{2x8 blocks in tile} DC coefficient represents the sum of the DC coefficients of all blocks in the i-th tile, and complexityLPHP _i represents the i-th tile The image complexity value of the LPHP coefficient, ∑ _{2x8 blocks in tile} LPHP coefficient represents the sum of the LPHP coefficients of all blocks in the i-th tile.

For the rightmost tile whose width is less than 384, its image complexity value can be calculated by formula (39) and formula (40):

complexityDC _n1 =complexityDC _n1-1 *tilewidth _n1 /384 (39)

complexityLPHP _n1 =complexityLPHP _n1-1 *tilewidth _n1 /384 (40)

Among them, complexityDC _n1 represents the image complexity value of the DC coefficient of the rightmost tile, complexityDC _n1-1 represents the image complexity value of the DC coefficient of the tile adjacent to the left side of the rightmost tile, and complexitLPHP _n1 Indicates the image complexity value of the LPHP coefficient of the rightmost tile, complexitLPHP _n1-1 indicates the image complexity value of the LPHP coefficient of the tile adjacent to the left side of the rightmost tile, and tilewidth _n1 indicates the rightmost The actual width of the tile.

As shown in FIG. 7 b above, by vertically dividing the image to be coded according to a fixed width 384 , three tiles can be obtained, namely tile 1 , tile 2 and tile 3 . Among them, the widths of tile 1 and tile 2 are both 384, and the width of tile 3 is smaller than 384.

The image complexity values of tile 1 and tile 2 can be obtained according to the above formulas (37) and (38), and the image complexity value of tile 3 can be obtained according to the above formulas (39) and (40). Wherein, the DC coefficient and the LPHP coefficient in the tile have been described above, and for the sake of brevity, details will not be repeated here.

Suppose the sum of the absolute values of the DC coefficients in tile 1 is 100, the sum of the absolute values of the LPHP coefficients is 80; the sum of the absolute values of the DC coefficients in tile 2 is 146, and the sum of the absolute values of the LPHP coefficients is 100 . which is:

(1), tile 1

complexityDC ₁ ＝∑ _{2x8 blocks in tile} DC coefficient＝100

complexityLPHP ₁ ＝∑ _{2x8 blocks in tile} LPHP coefficient＝80

Then the image complexity value of the DC coefficient of tile 1 is 100, and the image complexity value of the LPHP coefficient of tile 1 is 80.

(2), tile 2

complexityDC ₂ ＝∑ _{2x8 blocks in tile} DC coefficient＝146

complexityLPHP ₂ ＝∑ _{2x8 blocks in tile} LPHP coefficient＝100

Then the image complexity value of the DC coefficient of tile 2 is 146, and the image complexity value of the LPHP coefficient of tile 2 is 100.

(3), tile 3

complexitDC ₃ = complexitDC ₂ * tilewidth ₃ /384 = 146*100/384 = 38

complexitLPHP ₃ = complexitLPHP ₂ * tilewidth ₃ /384 = 146*100/384 = 26

Then the image complexity value of the DC coefficient of tile 3 is 38, and the image complexity value of the LPHP coefficient of tile 3 is 26.

After the image complexity value of each tile is obtained, the target number of bytes of each tile can be calculated according to its image complexity value, and then the i-th tile can be updated based on the cumulative value of the target number of bytes of the i-th tile QP of tiles. For calculating the target number of bytes of each tile according to its image complexity value, please refer to the following content.

In the solution provided by the embodiment of the present application, by updating the QP of the i-th tile according to the cumulative value of the target number of bytes of the i-th tile included in the image to be encoded, it can be ensured that without sacrificing the accuracy of code rate control, Improve coding efficiency and reduce hardware resource consumption, while ensuring the flexibility of quantization parameters to avoid the problem of uncontrollable output code rate.

Optionally, in some embodiments, the updating the QP of the i-th tile according to the cumulative value of the target number of bytes of the i-th tile includes: according to the absolute value of the first difference and The first threshold updates the QP of the ith tile, and the first difference is the cumulative value of the target number of bytes of the ith tile and the actual number of encoded bytes of the ith tile The difference between the accumulated values of .

Similarly, the cumulative value of the actual number of encoded bytes of the i-th tile in this embodiment of the present application may refer to the actual number of encoded bytes of all tiles in the previous i-th tile (including the i-th tile).

For example, the cumulative value of the actual number of encoded bytes of the second tile may refer to the sum of the actual number of encoded bytes of the first tile and the actual number of encoded bytes of the second tile; The actual number of encoded bytes can refer to the actual number of encoded bytes of the first tile, the actual number of encoded bytes of the second tile, the actual number of encoded bytes of the third tile, the actual number of encoded bytes of the fourth tile The sum of the actual number of encoded bytes and the actual number of encoded bytes of the fifth tile; and so on, the cumulative value of the actual number of encoded bytes of the n1th tile can refer to the actual encoding of the first tile The sum of the number of bytes, the actual number of encoded bytes of the second tile..., the actual number of encoded bytes of the n1-1th tile, and the actual number of encoded bytes of the n1th tile.

In the embodiment of the present application, the actual number of encoded bytes of the i-th tile can be obtained by the encoder during the process of encoding the image to be encoded.

The first threshold in the embodiment of the present application may be fixed or continuously adjusted, which is not specifically limited in the present application.

In the embodiment of this application, the encoder can encode in units of tiles. Before the start of image encoding, the initial QP ₀ of the first tile can be calculated according to the original size of the image and the target number of bytes of the current frame. For the QP of the current tile, the QP of the current tile may be updated according to the difference between the cumulative value of the target byte number of the current tile and the cumulative value of the actual coded byte number.

The calculation of the target byte count of the current frame and the target byte count of each tile is as follows.

a. The calculation of the target number of bytes of the current frame can be calculated by formula (41)

Wherein, targetByte represents the target number of bytes of the image (being the current frame in the embodiment of the present application), width represents the width of the image, height represents the height of the image, bitdepth represents the bit depth of the image, and m represents the number of all pixels The ratio of the number to the number of brightness pixels (for example, when encoding a YUV422 format image, the value of n is 2), and compressRatio indicates the image compression ratio.

b. The calculation of the target number of bytes of each tile can be calculated by formula (42)

Among them, n represents the number of tiles that the current image is divided into, i represents the i-th tile, t arg etByte _i represents the target byte number of the i-th tile, and complexityLPHP _i represents the number of bytes corresponding to the i-th tile. Image complexity of LPHP coefficients,

Represents the sum of the image complexity of the LPHP coefficients of all tiles in the current frame.

It should be noted that not all image complexity information is used in this embodiment, which is guidance information obtained based on prior data. For example, the calculation of the Qp of each tile according to the image complexity information in the following is calculated based on the image complexity value of the LPHP coefficient.

Continuing to take the above figure 7b as an example, assuming that the height of the image is 100, the bit depth is 8, the compression ratio is 200, and the encoding format is YUV422, then the target number of bytes of the current frame is:

The target byte count for each tile is:

After obtaining the target number of bytes for each tile, you can first calculate the difference between the cumulative value of the target number of bytes for each tile and the cumulative value of the actual number of encoded bytes, and then update or Calculate the QP of each tile.

In this embodiment of the present application, the first threshold may include multiple thresholds, for example, may include two thresholds, and the threshold may be calculated by using formula (43) and formula (44).

Wherein, targetByte represents the target number of bytes of the image (that is, the current frame in the embodiment of the present application), n represents the number of tiles that represent the current image to be encoded, and threshold1 and threshold2 represent threshold 1 and threshold 2. alpha and beta are preset parameters, and one setting method can be alpha=4, beta=16.

In addition, it is also pointed out above that the initial QP ₀ of the first tile can be calculated first, and then the QP of each tile can be updated or calculated based on the initial QP ₀ .

Because image complexity can represent the amount of information in the original image content, and the target number of bytes represents the amount of information after compression, it is reasonable and effective to calculate the parameter QP that characterizes the degree of compression by combining complexity and target number of bytes to establish a mathematical model. At the same time, in order to narrow the complexity and the range of the target number of bytes, the two will be converted to the logarithmic domain for calculation when calculating QP.

The initial QP calculation can be calculated by formula (45) to formula (49). The formula (45) in this embodiment normalizes the number of bytes to the data volume bpp of each pixel, and converts it to the logarithmic domain.

log(bpp)=log(t arg etByte/(width*height)) (45)

Among them, log in the formula means the logarithm with the natural logarithm as the base, and log10 means the logarithm with the base 10. Equation (46) and Equation (47) transform the complexity into the logarithmic domain. Since compression is a nonlinear process, the complexity will be further converted into bpp multiplicative relational operator Equation (46) in the process of establishing the mathematical model and the additive relationship operator formula (47) to improve the fit of the mathematical model.

Wherein, paramA, paramB, paramC, and paramD are preset parameters, and a typical configuration from prior knowledge may be paramA=0.02213, paramB=-24.32, paramC=30.45, and paramD=-76.78.

In this embodiment, QP is obtained by formula (48) established by multiplicative operator, additive operator, log(bpp) and variable parameters x and y of complexity.

QP＝x*log(bpp)+y+z (48)

Among them, z in the formula is a variable parameter, which can be updated later, and the initial value can be set to 0.

QP ₀ = Cilp3(min QP, max QP, QP) (49)

Wherein, QP ₀ in the formula is the final determined initial QP, minQP and maxQP are the minimum value and maximum value of QP respectively, minQP may not be less than 0, and maxQP may not be greater than 255. A typical set of configurations might be min QP=5, max QP=150.

It should be noted that the above formula (49) means that QP ₀ can be any value in the brackets. If the QP calculated by the above formula (18) is between minQP and maxQP, then QP ₀ is QP; if the QP calculated by the above formula (18) is less than minQP, then QP ₀ is minQP; if by the above formula (18 ) calculated QP is greater than maxQP, then QP ₀ is maxQP.

Continue taking the above-mentioned Fig. 7b as an example, as mentioned above, the width and height of the image to be encoded are 868 and 100 respectively, then log(bpp)=log(t arg etByte/(width*height))=log(868(868* 100)) = 0.01

QP=x*log(bpp)+y+z=147.47

_QP0 = Cilp3(5, 150, 147.47)

Then it can be obtained that the determined initial QP ₀ is 147.47.

In the subsequent encoding process, the QP of tile 1 in the current frame can be updated based on the determined initial QP ₀ , then the QP of tile 2 can be calculated based on the updated QP of tile 1, and finally the tile can be calculated based on the QP of tile 2 QP for slice 3.

In the solution provided by the embodiment of the present application, by updating the QP of the tile included in the image to be encoded according to the absolute value of the first difference and the first threshold, it can further ensure that the coding efficiency is improved without sacrificing the accuracy of the code rate control And reduce the consumption of hardware resources, while ensuring the flexibility of quantization parameters to avoid the problem of uncontrollable output code rate.

It has been explained above that the encoding end can update the QP of the i-th tile according to the absolute value of the first difference and the first threshold, which will be described in detail below.

Optionally, in some embodiments, the updating the QP of the ith tile according to the absolute value of the first difference and the first threshold includes: if the first difference is positive, setting The difference between the QP of the i-1 tile and the first offset QP is used as the QP of the i-th tile; if the first difference is negative, the i-1 tile’s The sum of QP and the first offset QP is used as the QP of the ith tile; wherein, the first offset QP is obtained based on the absolute value of the first difference and the first threshold .

In the embodiment of the present application, the QP of the ith tile of the image to be encoded may be updated in combination with the sign of the first difference.

If the i-th tile is currently encoded, the cumulative value of the target number of bytes and the cumulative value of the actual number of encoded bytes of the i-th tile can be calculated according to formula (50) and formula (51).

Among them, accTarBytes _i represents the cumulative value of the target number of bytes of the i-th tile,

Indicates the sum of the target bytes of all tiles in the first i tile (including the i-th tile), accActBytes _i indicates the cumulative value of the actual coded bytes of the i-th tile,

Indicates the sum of the actual number of encoded bytes of all tiles in the first i tile (including the i-th tile).

The first difference is calculated according to formula (50) and formula (51), as shown in formula (52), and the offset value of QP is obtained through formula (53) and formula (54).

deltaBytes _i =accTarBytes _i -accActBytes _i (52)

QPoffset _i ＝|deltaBytes _i |＞threhold1? offsetA: offsetB (53)

offsetB＝|deltaBytes _i |＞threhold2?1:0 (54)

Among them, deltaBytes _i represents the difference between the cumulative value of the target number of bytes of the i-th tile and the cumulative value of the actual number of encoded bytes, that is, the first difference, and QPoffset _i represents the offset of the QP of the i-th tile value, offsetA is a preset parameter and can be set to 2.

The meaning expressed by formula (53) is: if the absolute value |deltaBytes _i | , then the offset value QPoffset _i of the QP of the i-th tile currently encoded can be offsetA, otherwise it can be offsetB.

Similarly, the meaning expressed by formula (54) is: if the absolute value |deltaBytes _i | If the threshold threshold2 is set, the QP offset value QPoffset _i of the QP of the i-th tile currently encoded can take a value of 1, otherwise it takes a value of 0.

Further update the QP used by the i-th tile, the range of QP can still be between [min QP, max QP]. The QP of the ith tile can be updated according to equation (55).

QP _i ＝QP _i-1 -sign(deltaBytes _i )*QPoffset _i (55)

QP _i =Cilp3(min QP, max QP, QP _i ) (56)

Among them, QP _i represents the QP of the i-th tile, QP _i-1 represents the QP of the i-1-th tile, and sign(deltaBytes _i ) represents the sign of deltaBytes _i .

Continuing to take the above-mentioned FIG. 7b as an example for description, the QPs of tile 1, tile 2, and tile 3 are updated or calculated respectively.

a. Calculation of QP of tile 1

First, the preset threshold 1 (threhold1) and the preset threshold 2 (threhold2) can be calculated respectively according to the above formula (43) and formula (44). Suppose alpha=4, beta=16, then:

Secondly, as mentioned above, the target number of bytes of tile 1 is 337, assuming that the actual number of encoded bytes of tile 1 is 300, then the target number of bytes of tile 1 and the actual number of encoded bytes can be obtained by formula (52). The difference in the number of sections is deltaBytes ₁ =accTarBytes ₁ -accActBytes ₁ =337-300=37.

Calculate the QP offset value of tile 1 according to formula (53) and formula (54). Since the difference between the target number of bytes of tile 1 and the actual number of encoded bytes is 37, which is greater than the preset threshold 2, therefore, The QP of tile 1 has an offset value of 1.

Then QP1 of tile 1 can be updated according to the offset value of QP of tile 1 and the absolute value of the first difference: QP ₁ =QP ₀ -sign(deltaBytes ₁ )*QPoffset ₁ =147.47-1=146.47

b. Calculation of QP of tile 2

As mentioned above, the target byte count of tile 2 is 421, assuming that the actual encoded byte count of tile 2 is 470, the cumulative value of the target byte count of tile 2 is the same as the target byte count of tile 1 The sum of the target bytes of tile 2 is 421+337=758; the cumulative value of the actual encoded bytes of tile 2 is the actual encoded bytes of tile 1 and the actual encoded bytes of tile 2 The sum of the numbers is 300+470=770.

Then, the difference between the cumulative target byte count and the actual coded byte count of tile 2 can be obtained by formula (52) as deltaBytes ₂ =accTarBytes ₂ -accActBytes ₂ =758-770=-12.

Calculate the QP offset value of tile 2 according to formula (53) and formula (54). Since the difference between the cumulative value of the target byte number of tile 2 and the cumulative value of the actual coded byte number is -12, it The absolute value is 12, which is smaller than the preset threshold value 2. Therefore, the offset value of the QP of tile 2 is 0.

Then QP2 of tile 2 can be updated according to the offset value of QP of tile 2 and the absolute value of the first difference: QP ₂ =QP ₁ -sign(deltaBytes ₂ )*QPoffset ₂ =146.47-0=146.47

c. Calculation of QP of tile 3

As mentioned above, the target byte count of tile 3 is 109.5, assuming that the actual encoded byte count of tile 3 is 50, the cumulative value of the target byte count of tile 3 is the same as the target byte count of tile 1 The sum of the target byte count of tile 2 and the target byte count of tile 3 is 421+337+109.5=867.5; the cumulative value of the actual coded byte count of tile 3 is the actual coded word of tile 1 The sum of the number of sections, the actual number of encoded bytes of tile 2 and the actual number of encoded bytes of tile 3 is 300+470+50=820.

Then, the difference between the cumulative target byte count and the actual coded byte count of tile 3 can be obtained by formula (52) as deltaBytes ₃ =accTarBytes ₃ -accActBytes ₃ =867.5-820=47.5.

Calculate the QP offset value of tile 3 according to formula (53) and formula (54), because the difference between the cumulative value of the target byte number of tile 3 and the cumulative value of the actual coded byte number is 47.5, which is greater than the expected Set the threshold to 2, therefore, the offset value of the QP of tile 3 is 1.

Then QP3 of tile 3 can be updated according to the offset value of QP of tile 3 and the absolute value of the first difference: QP ₃ =QP ₂ -sign(deltaBytes ₃ )*QPoffset ₃ =146.47-1=145.47

Optionally, in some embodiments, the target number of bytes of the i-th tile is related to first information; the first information is at least one of the following information: the target number of bytes of the current frame , the transformation coefficient of the ith tile or the transformation coefficient of the current frame.

Optionally, in some embodiments, the target byte count of the current frame is related to second information; the second information is at least one of the following information: width of the current frame, width of the current frame The height of the current frame, the bit depth of the current frame, the coding format of the current frame or the image compression ratio of the current frame.

In this embodiment of the present application, the target number of bytes of the i-th tile may be related to the target number of bytes of the current frame, the transformation coefficient of the i-th tile, or the transformation coefficient of the current frame. For example, the target number of bytes of the i-th tile can be calculated by the above formula (42). For specific content, please refer to the description of the above formula (42). For the sake of brevity, details will not be repeated here.

In this embodiment of the present application, the target number of bytes of the current frame may be related to the width of the current frame, the height of the current frame, the bit depth of the current frame, the encoding format of the current frame, or the image compression ratio of the current frame. For example, the target number of bytes of the current frame can be calculated by the above-mentioned formula (41), and the specific content can refer to the description of the above-mentioned formula (41), for the sake of brevity, no more details are given here.

Based on this, the calculation of the QP of n1 tiles in the current frame is described above, and the calculation of the QP of the tiles in the target frame will be described below, please refer to the following for details.

Optionally, in some embodiments, the method further includes: updating the n2 tiles in the target frame according to the ratio of the absolute value of the second difference to the target byte number of the current frame and the size of the second threshold The QP of the slice.

Wherein, the second difference is the difference between the target number of bytes of the current frame and the actual number of encoded bytes of the current frame, and the target frame is the first x frames and/or the last x frames of the current frame y frame, x and y are positive integers greater than or equal to 1, and n2 is a positive integer greater than or equal to 2.

In the embodiment of the present application, the QPs of n2 tiles in the target frame may be updated according to the ratio of the absolute value of the second difference to the target byte number of the current frame and the size of the second threshold.

In the embodiment of the present application, if the current encoding method is forward prediction, the target frame may be the previous x frames of the current frame; if the current encoding method is backward prediction, the target frame may be the last y frames of the current frame; if If the current encoding method is bidirectional prediction, the target frame may be the previous x frame and the next y frame of the current frame.

Similarly, similar to the division method of the above-mentioned current frame, in the embodiment of the present application, when dividing the image to be encoded (target frame), the image to be encoded may be divided according to a fixed width or a fixed height, or may not be divided according to a fixed width Or divide the image to be coded with a fixed height; there is no limit.

In the embodiment of the present application, the coded image may be divided horizontally, or the image to be coded may be divided vertically.

It should be understood that n2 in this embodiment of the present application may be the same as or different from n1 above, which is not specifically limited in this application.

In the solution provided by the embodiment of the present application, the ratio of the second difference (the difference between the target number of bytes of the current frame and the actual number of encoded bytes) to the target number of bytes of the current frame is used to update the number of tiles included in the target frame. QP can further improve the coding efficiency, and at the same time, it can ensure the flexibility of quantization parameters to avoid the problem of uncontrollable output code rate.

Optionally, in some embodiments, updating the QP of n2 tiles in the target frame according to the ratio of the absolute value of the second difference to the target byte number of the current frame and the size of a second threshold, include:

If the second difference is a positive value, the difference between the first parameter and the offset parameter used to calculate the QP of the n1 tiles is used as the updated first parameter; according to the updated first parameter Calculate the QP of the n2 tiles;

or,

If the second difference is a negative value, the sum of the first parameter and the offset parameter used to calculate the QP of the n1 tiles is used as the updated first parameter; calculated according to the updated first parameter QP of the n2 tiles.

In the embodiment of the present application, to update the QP of the n2 tiles of the target frame, the initial QP0 of the first tile of the target frame can be obtained first, and the parameter z in the above formula (48) can be updated in the following way, and then based on the update The last parameter z (namely the first parameter in this application) calculates the initial QP ₀ of the target frame.

First, the actual number of encoded bytes of the current frame can be calculated by formula (57).

Among them, accActBytes _n1 represents the actual number of encoded bytes of the current frame, and n1 represents the number of tiles divided by the current frame.

Calculate the difference between the target byte number of the current frame and the actual coded byte number according to formula (58).

deltaBytes _n1 = t arg etByte _n1 -accActBytes _n1 (58)

Among them, deltaBytes _n1 indicates the difference between the target number of bytes of the current frame and the actual number of encoded bytes, targ etByte _n1 indicates the target number of bytes of the current frame, and accActBytes _n1 indicates the actual number of encoded bytes of the current frame.

The updated parameter z is determined according to Equation (59)-Equation (61).

Wherein, delta z represents the difference before and after the update of parameter z, and deltaThres0, deltaThres1, deltaThres2 in the above-mentioned formula (59) - formula (61) are preset threshold coefficients, for example, can be preset as deltaThres0=0.02, deltaThres1=0.05 , deltaThres2=0.1; tmpOffset1 and tmpOffset2 are temporary variables.

deltaOffset1, deltaOffset2, and deltaOffset3 in formula (60) and formula (61) are preset offset values, for example, deltaOffset1=1, deltaOffset2=2, deltaOffset3=3 can be preset.

After getting delta z, the updated parameter z can be calculated based on formula (62).

z _new = z-sign(deltaBytes _n1 )*delta z (62)

Among them, z _new represents the value of the parameter z after the update, z represents the value of the parameter z before the update, and sign(deltaBytes _n ) represents the sign of deltaBytes _n .

Still taking the above-mentioned FIG. 7b as an example for description, the above-mentioned parameter z is updated.

The actual encoded byte count of the current frame can be obtained by the above formula (57):

As mentioned above, the target byte count for this current frame is 867.5.

Then it can be obtained through the above formula (58), that the difference between the target byte number and the actual coded byte number of the current frame is: deltaBytes ₃ =t arg etByte ₃ -accActBytes ₃ =867.5-820=47.5.

After obtaining the difference between the target number of bytes of the current frame and the actual number of encoded bytes, it can be calculated by formula (59), the ratio of the absolute value of the above difference to the target number of bytes of the current frame is 47.5/ 867.5=0.05, because the ratio 0.05 is greater than the difference preset threshold coefficient 0 (deltaThres0=0.02), then delta z=tmpOffset1, further, can be calculated according to formula (60), the ratio 0.05 is equal to the difference preset threshold coefficient 1 (deltaThres1=0.05), then delta z=tmpOffset2; further, it can be calculated according to formula (61), the ratio 0.05 is less than the difference preset threshold coefficient 2 (deltaThres2=0.1), then delta z=deltaOffset2, if deltaOffset2 =2, then the difference between the parameter z before and after updating is 2.

The updated parameter z _new =z-sign(deltaBytes ₃ )*delta z=0−2=−2 is calculated by the above formula (62). When subsequently calculating the initial QP ₀ of the first tile of the target frame, the updated parameter z _new can be used for calculation.

In other words, in the process of encoding the target frame, the value of the parameter z in the above formula (48) is calculated with the updated parameter z _new , that is, -2 is used to calculate the initial QP of the first tile of the target frame ₀ . Subsequently, the initial QP ₁ of the first tile can be updated based on the image complexity information of the first tile of the target frame, and then the QP ₂ of the second tile can be calculated based on the updated initial QP ₁ of the first tile , and then calculate the QP ₃ of the third tile based on the QP ₂ of the second tile, ..., and so on, until the QP calculation of all tiles of the target frame is completed.

In the solution provided by the embodiment of the present application, the updated first parameter is determined in combination with the sign of the second difference value, and the QP of the tile included in the target frame is updated based on the updated first parameter, which can further improve coding efficiency.

FIG. 10 is a schematic diagram of an encoding method 1000 provided by another embodiment of the present application. The encoding method 1000 may include steps 1010-1060. The encoding method provided by the embodiment of the present application is outlined below with reference to FIG. 10 .

1010. Calculate the target byte count of the current frame and the target byte count of each tile.

1020. Determine whether to encode the first tile currently.

If the first tile is currently being encoded, then step 1030 is performed; if the first tile is not currently being encoded, then step 1040 is performed.

1030. Calculate the initial QP of the first tile.

1040. Perform intra-frame QP update of the current frame.

1050. Determine whether the encoding of the last tile of the current frame is completed.

If yes, execute step 1060 , if not, return to execute step 1020 .

1060, frame-level parameter update.

For the content of the above steps 1010-1060, please refer to the content of Fig. 6-Fig.

In the above content, only single-channel encoding of the image to be encoded is involved. In some embodiments, multi-channel encoding can be performed on the image to be encoded. In this case, the image of each tile in the current frame is acquired Complexity information can obtain image complexity information of multiple components, see below for details.

Optionally, in some embodiments, if multi-channel encoding is performed on the current frame, the acquiring the image complexity information of each of the n1 tiles in the current frame includes: respectively acquiring the n1 The image complexity information of the components of each tile in the tiles, the components include the brightness component and/or at least one chrominance component of the current frame; update the n1 tiles according to the image complexity information The QP includes: updating the QP of the n1 tiles according to the image complexity information of at least one of the components.

The multi-channel encoding in this embodiment of the present application may include YUV encoding or RGB encoding, without limitation.

YUV refers to a pixel format in which luma parameters and chrominance parameters are represented separately. The format of YUV encoding can include YUV444, YUV422, YUV420 and YUV411. The following takes the YUV encoding format as YUV422 as an example for illustration.

Y component: For the image complexity information of the tiles of the Y component, please refer to the above content.

UV component: Since the format of YUV encoding is YUV422, the pixel of the UV component can be half of the pixel value of the image to be encoded, and the image complexity information of the tile is obtained based on this half of the pixel value.

In the embodiment of the present application, the image complexity information of the Y component and the UV component of each tile in the n1 tiles can be respectively obtained, and the average of the image complexity information of the Y component and the UV component can be based on The QP of n1 tiles can be updated by the value; the QP of n1 tiles can also be updated according to the image complexity information of the Y component; the QP of n1 tiles can also be updated according to the image complexity information of the UV component; limit.

It should be understood that updating the QP of n1 tiles according to the image complexity information of at least one of the components is not limited to the ones listed above, and can also be in other ways, for example, the image complexity information of the Y component and the image of the UV component The root mean square value of the complexity information and the like are not specifically limited in this application.

In the solution provided by the embodiment of the present application, the accuracy of the quantization parameter can be improved by updating the QP of the tile according to the obtained image complexity information of at least one component of the image complexity information of the components of each tile.

Optionally, in some embodiments, the acquiring the image complexity information of each of the n1 tiles in the current frame includes: acquiring multiple transformation parameters of each tile; One parameter is selected from the plurality of transformation parameters as the image complexity information of each tile.

In the embodiment of the present application, one parameter may be selected from the acquired transformation parameters to calculate the image complexity information of each tile. In the content described above, the image complexity information of each tile is calculated based on the LPHP coefficient of the tile. In some embodiments, the image complexity information of each tile can also be calculated according to the DC coefficient of the tile. degree information, or, the image complexity information of each tile may also be calculated according to the average value of the LPHP coefficient and the DC coefficient of the tile, which is not specifically limited in the present application.

Optionally, in some embodiments, the method 600 may also include:

If the updated QP of the n1 tiles is less than a third threshold, use the third threshold as the updated QP of the n1 tiles; or,

If the updated QP of the n1 tiles is greater than a fourth threshold, use the fourth threshold as the updated QP of the n1 tiles;

Wherein, the third threshold is the minimum value of the QP used for encoding, and the fourth threshold is the maximum value of the QP used for encoding.

In the embodiment of the present application, in the process of encoding the image to be encoded, the minimum value (ie, the third threshold) and the maximum value (ie, the fourth threshold) of QP may be set. The third threshold and/or the fourth threshold in this embodiment of the present application may be fixed or continuously adjusted, and are not limited.

If the QP of the tile calculated by the above formula is less than the minimum value, then the QP of the tile can be calculated based on the set minimum value; if the QP of the tile calculated by the above formula is greater than the maximum value, then the follow-up can be based on the set maximum value Calculate the QP of the tile.

Exemplarily, as mentioned above, the minimum and maximum values of QP can be set to 5 and 150 respectively. If the QP of a tile calculated by the above formula is any value from 5 to 150, it can be based on the calculated QP. The next tile is encoded; if the QP of the tile calculated by the above formula is 3, the next tile can be encoded based on 5; if the QP of the tile calculated by the above formula is 160, it can be encoded based on 150 The next tile is encoded.

Optionally, in some embodiments, the encoding method is applied in the JPEG XR encoding format.

As mentioned above, the JPEG XR encoding format is a continuous-tone still image compression algorithm and file format that can support lossy data compression as well as lossless data compression.

The JPEG XR encoding format has certain advantages over the JPEG encoding format.

First, JPEG uses 8-bit encoding, enabling 256 colors, while JPEG XR can use 16-bit or more, providing better results and more editing flexibility.

Secondly, the JPEG XR encoding format uses a more efficient compression algorithm. In the case of the same size as a JPEG file, the image quality can be twice that of the latter, or half the size of the latter for the same quality. And unlike JPEG, JPEG XR's highest quality compression allows no loss of information.

FIG. 11 is a schematic diagram of an encoding method 1100 provided by still another embodiment of the present application. The encoding method 1100 may include steps 1110-1130.

1110. Acquire image complexity information of the current frame, where the image complexity information includes transformation coefficients obtained after image kernel transformation processing (PCT processing) is performed on pixel values of the current frame.

In the embodiment of the present application, for the process of PCT processing, reference may be made to the descriptions of the foregoing formulas (2) to (36), and for the sake of brevity, details are not repeated here.

1120. Determine an initial quantization parameter (initial QP) of the current frame according to the image complexity information.

In the embodiment of the present application, the initial QP of the current frame can be determined according to the image complexity information, and the initial QP can be described according to the above formula (45) to formula (49). For the sake of brevity, details are not repeated here.

It should be noted that since the encoding method 1100 is performed in units of frames, the initial QP in the embodiment of the present application is the initial QP of the current frame; while the encoding method 600 above is performed in units of tiles Therefore, the initial QP in the above embodiment is the initial QP of the first tile in the current frame.

1130. Update the initial QP of the target frame according to the initial QP of the current frame, where the target frame is x frames before and/or y frames after the current frame, and x and y are positive integers greater than or equal to 1.

Optionally, in some embodiments, the updating the initial QP of the target frame according to the initial QP of the current frame includes: according to the difference between the absolute value of the second difference and the target byte number of the current frame The ratio and the size of the second threshold update the initial QP in the target frame; wherein, the second difference is the difference between the target byte number of the current frame and the actual coded byte number of the current frame.

The second threshold in the embodiment of the present application may be a fixed value, or may be continuously adjusted, which is not specifically limited in the present application.

Optionally, in some embodiments, the updating the initial QP of the target frame according to the ratio of the absolute value of the second difference to the target byte number of the current frame and the size of a second threshold includes :

If the second difference is a positive value, the difference between the first parameter and the offset parameter used to calculate the initial QP of the current frame is used as the updated first parameter;

calculating the initial QP of the target frame according to the updated first parameter;

or,

If the second difference is a negative value, the sum of the first parameter and the offset parameter used to calculate the initial QP of the previous frame is used as the updated first parameter;

Calculate the initial QP of the target frame according to the updated first parameter.

The first parameter in the embodiment of the present application can be the parameter z in the above formula (48). In the process of calculating the initial QP of the current frame above, the initial value of the parameter z can be set to 0, and the subsequent calculation of the target frame When the initial QP of , the parameter z in the formula (48) can be updated according to the image complexity information of the current frame.

Specifically, the difference between the target number of bytes of the current frame and the actual number of encoded bytes can be calculated first, based on the ratio of the absolute value of the difference to the target number of bytes of the current frame and the size update formula of the second threshold (48 ), and then update the initial QP of the target frame based on the updated parameter z and the image complexity information of the target frame. For details, please refer to the above formula (57) ~ formula (62) and formula (45) ~(49), for the sake of brevity, no more details here.

In the solution provided by the embodiment of the present application, the updated first parameter is determined in combination with the sign of the second difference value, and the QP of the target frame is updated based on the updated first parameter, which can further improve coding efficiency.

In the above content, only single-channel encoding of the image to be encoded is involved. In some embodiments, multi-channel encoding can be performed on the image to be encoded. In this case, obtaining the image complexity information of the current frame can obtain multiple The image complexity information of each component, see below for details.

Optionally, in some embodiments, if multi-channel encoding is performed on the current frame, the acquiring the image complexity information of the current frame includes: respectively acquiring the image complexity information of the components of the current frame, the The components include a luminance component and/or at least one chrominance component of the current frame; updating the initial QP of the target frame according to the initial QP of the current frame includes: updating according to the image complexity information of at least one of the components The initial QP of the target frame.

Assuming that the image complexity information of the Y component and the UV component of the current frame are respectively obtained, the initial QP of the target frame can be updated based on the average value of the image complexity information of the Y component and the image complexity information of the UV component; The image complexity information updates the initial QP of the target frame; the initial QP of the target frame can also be updated according to the image complexity information of the UV component; not limited.

It should be understood that updating the initial QP of the target frame according to the image complexity information of at least one of the components is not limited to the ones listed above, and can also be in other ways, for example, the image complexity information of the Y component and the image complexity of the UV component The root mean square value of degree information, etc., is not specifically limited in this application.

In the solution provided by the embodiment of the present application, the accuracy of quantization parameters can be improved by updating the initial QP of the target frame according to the acquired image complexity information of at least one component in the image complexity information of the components of the current frame.

Optionally, in some embodiments, the acquiring the image complexity information of the current frame includes: acquiring multiple transformation parameters of the current frame; selecting a parameter from the multiple transformation parameters as the current frame image complexity information.

In the embodiment of the present application, one parameter may be selected from the acquired transformation parameters to calculate the image complexity information of each tile. In the content described above, the image complexity information of the current frame can be calculated based on the LPHP coefficient of the current frame. In some embodiments, the image complexity information of the current frame can also be calculated according to the DC coefficient of the current frame, or , the image complexity information of the current frame may also be calculated according to the average value of the LPHP coefficient and the DC coefficient of the current frame, which is not specifically limited in the present application.

Optionally, in some embodiments, the method may further include: if the updated initial QP of the target frame is smaller than a third threshold, using the third threshold as the updated target frame or, if the updated initial QP of the target frame is greater than a fourth threshold, use the fourth threshold as the updated initial QP of the target frame; wherein, the third The threshold is the minimum value of the QP used for encoding, and the fourth threshold is the maximum value of the QP used for encoding.

In the embodiment of the present application, during the process of encoding the image to be encoded, the minimum value (ie, the third threshold) and the maximum value (ie, the fourth threshold) of the initial QP may be set.

If the initial QP of the current frame calculated by the above formula is less than the minimum value, then the initial QP of the target frame can be calculated based on the set minimum value; if the initial QP of the current frame calculated by the above formula is greater than the maximum value, then the follow-up can be based on the set The maximum value of calculates the initial QP of the target frame.

Exemplarily, as mentioned above, the minimum and maximum values of the initial QP can be set to 5 and 150 respectively. If the initial QP of the current frame calculated by the above formula is any value from 5 to 150, it can be obtained based on the calculation The initial QP of the current frame is used to encode the target frame; if the initial QP of the current frame calculated by the above formula is 3, the target frame can be encoded based on 5; if the initial QP of the current frame calculated by the above formula is 160, then the target frame can be encoded based on 150 Encode the target frame.

The method embodiment of this application is described in detail above with reference to Figure 1-Figure 11, and the device embodiment of this application is described below in conjunction with Figure 12-Figure 17. The device embodiment and the method embodiment correspond to each other, so the details are not described in detail For details, please refer to the previous method embodiments.

FIG. 12 is a schematic structural diagram of an encoding device 1200 provided by an embodiment of the present application. The encoding device 1200 may include a complexity calculation module 1210 and a code rate control module 1220 .

A complexity calculation module 1210, configured to acquire image complexity information of each of the n1 tiles in the current frame, the image complexity information including image kernel transformation processing of pixel values of each tile The transformation coefficient obtained after (PCT processing), n1 is a positive integer greater than or equal to 2.

A code rate control module 1220, configured to update the quantization parameters (QP) of the n1 tiles according to the image complexity information.

Figure 13a is a schematic structural diagram of a JPEG XR encoder provided by an embodiment of the present application. The schematic diagram may include a filter module 410 , a transform module 420 , a quantization module 430 , a prediction module 440 , an entropy encoding module 450 , a complexity calculation module 460 and a code rate control module 470 .

Among them, the five modules of the filtering module 410 , the transform module 420 , the quantization module 430 , the prediction module 440 , and the entropy coding module 450 are similar to the functions of the modules mentioned above in FIG. 2 .

The complexity calculation module 460 and the code rate control module 470 can be the complexity calculation module 1210 and the code rate control module 1220 in the embodiment of the present application, and can realize the update of the QP of the tile in the embodiment of the present application.

Among them, the complexity calculation module 460 can obtain image complexity information and output the information to the code rate control module 470, and the code rate control module 470 can receive the size of the actual code stream as an input to update the code rate control parameters, that is, realize QP update.

FIG. 13b is a schematic structural diagram of a JPEG XR encoder provided by another embodiment of the present application. The schematic diagram may also include a filter module 410 , a transform module 420 , a quantization module 430 , a prediction module 440 , an entropy encoding module 450 , a complexity calculation module 460 and a code rate control module 470 .

Different from Fig. 13a, the complexity calculation module 460 and the code rate control module 470 in Fig. 13b are located in the JPEG XR encoder, while the complexity calculation module 460 in Fig. 13a is located in the processor, and the code rate control module 470 Located in the JPEG XR encoder.

The two JPEG XR encoders shown in Figure 13a and Figure 13b can both implement the update of the QP of the tile in the embodiment of the present application. The difference is that the JPEG XR encoder shown in Figure 13a is currently encoding. The encoding of the current image to be encoded can only be started after the complexity calculation of the image to be encoded is completed. Compared with the JPEG XR encoder shown in Figure 13b, the time is slightly increased; the complexity calculation of the image to be encoded is completed before starting Encode the current image to be encoded, and then use it with a delay of one frame. Compared with the JPEG XR encoder shown in Figure 13b, the first frame has no input prior knowledge.

Optionally, in some embodiments, the code rate control module 1220 is further configured to: calculate the cumulative value of the target number of bytes of the i-th tile among the n1 tiles according to the image complexity information , i is a positive integer less than or equal to n; the QP of the i-th tile is updated according to the cumulative value of the target number of bytes of the i-th tile.

Optionally, in some embodiments, the code rate control module 1220 is further configured to: update the QP of the ith tile according to the absolute value of the first difference and the first threshold, the first difference is the difference between the cumulative value of the target number of bytes of the ith tile and the cumulative value of the actual coded number of bytes of the ith tile.

Optionally, in some embodiments, the code rate control module 1220 is further configured to: if the first difference is a positive value, calculate the difference between the QP of the i-1th tile and the first offset QP value as the QP of the ith tile; if the first difference is a negative value, the sum of the QP of the i-1th tile and the first offset QP is used as the ith QP of tiles; wherein, the first offset QP is obtained based on the absolute value of the first difference and the first threshold.

Optionally, in some embodiments, the code rate control module 1220 is further configured to: update the target frame according to the ratio of the absolute value of the second difference to the target byte number of the current frame and the size of the second threshold The QP of the n2 tiles; wherein, the second difference is the difference between the target number of bytes of the current frame and the actual number of encoded bytes of the current frame, and the target frame is the current The previous x frame and/or the subsequent y frame of the frame, x and y are positive integers greater than or equal to 1, and n2 is a positive integer greater than or equal to 2.

Optionally, in some embodiments, the code rate control module 1220 is further configured to: if the second difference is a positive value, combine the first parameter in the QP used to calculate the n1 tiles with The difference of the offset parameter is used as the updated first parameter; the QP of the n2 tiles is calculated according to the updated first parameter; or, if the second difference is negative, it will be used to calculate the The sum of the first parameter and the offset parameter in the QP of the n1 tiles is used as the updated first parameter; and the QP of the n2 tiles is calculated according to the updated first parameter.

Optionally, in some embodiments, if multi-channel encoding is performed on the current frame, the complexity calculation module 1210 is further configured to: separately obtain an image of a component of each tile in the n1 tiles Complexity information, the component includes the brightness component and/or at least one chrominance component of the current frame; the code rate control module 1220 is further configured to: update according to the image complexity information of at least one component in the component QP of the n1 tiles.

Optionally, in some embodiments, the complexity calculation module 1210 is further configured to: obtain multiple transformation parameters of each tile; select a parameter from the multiple transformation parameters as the The image complexity information of the tile.

Optionally, in some embodiments, the code rate control module 1220 is further configured to: if the updated QP of the n1 tiles is smaller than a third threshold, use the third threshold as the updated The QP of the n1 tiles; or, if the updated QP of the n1 tiles is greater than a fourth threshold, the fourth threshold is used as the updated QP of the n1 tiles; wherein , the third threshold is the minimum value of the QP used for encoding, and the fourth threshold is the maximum value of the QP used for encoding.

Optionally, in some embodiments, the encoding device 1200 is applied in the JPEG XR encoding format.

FIG. 14 is a schematic structural diagram of an encoding device 1400 provided by another embodiment of the present application. The encoding device 1400 may include a complexity calculation module 1410 and a code rate control module 1420 .

The complexity calculation module 1410 is configured to obtain image complexity information of the current frame, where the image complexity information includes transformation coefficients obtained after image kernel transformation processing (PCT processing) is performed on the pixel values of the current frame.

A code rate control module 1420, configured to determine an initial quantization parameter (initial QP) of the current frame according to the image complexity information.

The code rate control module 1420 is also configured to: update the initial QP of the target frame according to the initial QP of the current frame, the target frame is the previous x frame and/or the next y frame of the current frame, and x and y are A positive integer greater than or equal to 1.

The complexity calculation module 460 and the code rate control module 470 in FIG. 13b above can be the complexity calculation module 1210 and the code rate control module 1220 in the embodiment of the present application, which can realize the update of the QP of the target frame in the embodiment of the present application .

Optionally, in some embodiments, the code rate control module 1420 is further configured to: update according to the ratio of the absolute value of the second difference to the target number of bytes of the current frame and the size of the second threshold The initial QP in the target frame; wherein, the second difference is the difference between the target byte count of the current frame and the actual coded byte count of the current frame.

Optionally, in some embodiments, the code rate control module 1420 is further configured to: if the second difference is a positive value, combine the first parameter used to calculate the initial QP of the current frame with the offset The difference of the shift parameter is used as the updated first parameter; the initial QP of the target frame is calculated according to the updated first parameter; or, if the second difference is negative, it will be used to calculate the initial QP of the previous frame The sum of the first parameter in the QP and the offset parameter is used as the updated first parameter; and the initial QP of the target frame is calculated according to the updated first parameter.

Optionally, in some embodiments, if multi-channel encoding is performed on the current frame, the complexity calculation module 1410 is further configured to: respectively acquire image complexity information of components of the current frame, the components include The luma component and/or at least one chrominance component of the current frame; the code rate control module 1420 is further configured to: update the initial QP of the target frame according to the image complexity information of at least one component in the component.

Optionally, in some embodiments, the complexity calculation module 1410 is further configured to: acquire multiple transformation parameters of the current frame; select a parameter from the multiple transformation parameters as the image of the current frame complexity information.

Optionally, in some embodiments, the code rate control module 1420 is further configured to: if the updated initial QP of the target frame is smaller than a third threshold, use the third threshold as the updated The initial QP of the target frame; or, if the updated initial QP of the target frame is greater than a fourth threshold, the fourth threshold is used as the updated initial QP of the target frame; wherein, The third threshold is the minimum value of the QP used for encoding, and the fourth threshold is the maximum value of the QP used for encoding.

Optionally, in some embodiments, the encoding device 1400 is applied in the JPEG XR encoding format.

FIG. 15 provides an encoding device 1500 according to yet another embodiment of the present application, and the encoding device 1500 may include a processor 1510 .

The processor 1510 is configured to: acquire image complexity information of each of the n1 tiles in the current frame, where the image complexity information includes image kernel transformation processing of pixel values of each tile ( PCT processing), n1 is a positive integer greater than or equal to 2; update the quantization parameters (QP) of the n1 tiles according to the image complexity information.

Optionally, in some embodiments, the processor 1510 is further configured to: calculate the cumulative value of the target number of bytes of the i-th tile among the n1 tiles according to the image complexity information, i is a positive integer less than or equal to n; the QP of the i-th tile is updated according to the cumulative value of the target number of bytes of the i-th tile.

Optionally, in some embodiments, the processor 1510 is further configured to: update the QP of the ith tile according to an absolute value of a first difference and a first threshold, where the first difference is The difference between the cumulative value of the target number of bytes of the ith tile and the cumulative value of the actual coded number of bytes of the ith tile.

Optionally, in some embodiments, the processor 1510 is further configured to: if the first difference is a positive value, use the difference between the QP of the i-1th tile and the first offset QP as The QP of the ith tile; if the first difference is a negative value, the sum of the QP of the i-1th tile and the first offset QP is used as the ith tile QP of a slice; wherein, the first offset QP is obtained based on the absolute value of the first difference and the first threshold.

Optionally, in some embodiments, the processor 1510 is further configured to: update the number of bytes in the target frame according to the ratio of the absolute value of the second difference to the target byte number of the current frame and the size of the second threshold. QP of n2 tiles; wherein, the second difference is the difference between the target number of bytes of the current frame and the actual number of encoded bytes of the current frame, and the target frame is the number of bytes of the current frame The previous x frame and/or the next y frame, x and y are positive integers greater than or equal to 1, and n2 is a positive integer greater than or equal to 2.

Optionally, in some embodiments, the processor 1510 is further configured to: if the second difference is a positive value, the first parameter and the offset used to calculate the QP of the n1 tiles The difference between the parameters is used as the updated first parameter; the QP of the n2 tiles is calculated according to the updated first parameter; or, if the second difference is negative, it will be used to calculate the n1 tiles The sum of the first parameter in the QP of the tile and the offset parameter is used as the updated first parameter; and the QP of the n2 tiles is calculated according to the updated first parameter.

Optionally, in some embodiments, if multi-channel encoding is performed on the current frame, the processor 1510 is further configured to: respectively obtain the image complexity of the components of each tile in the n1 tiles Information, the component includes the luma component and/or at least one chrominance component of the current frame; the QP of the n1 tiles is updated according to the image complexity information of at least one component in the component.

Optionally, in some embodiments, the processor 1510 is further configured to: acquire multiple transformation parameters of each tile; select one parameter from the multiple transformation parameters as the parameter for each tile image complexity information.

Optionally, in some embodiments, the processor 1510 is further configured to: if the updated QP of the n1 tiles is smaller than a third threshold, use the third threshold as the updated The QP of n1 tiles; or, if the updated QP of the n1 tiles is greater than the fourth threshold, the fourth threshold is used as the updated QP of the n1 tiles; wherein, The third threshold is the minimum value of the QP used for encoding, and the fourth threshold is the maximum value of the QP used for encoding.

Optionally, in some embodiments, the encoding device 1500 is applied in the JPEG XR encoding format.

Optionally, the encoding device 1500 may further include a memory 1520 . Wherein, the processor 1510 can invoke and run a computer program from the memory 1520, so as to implement the method in the embodiment of the present application.

Wherein, the memory 1520 may be an independent device independent of the processor 1510 , or may be integrated in the processor 1510 .

Optionally, the encoding device 1500 may further include a transceiver 1530 . Wherein, the transceiver 1530 may be an independent device independent of the processor 1510 , or may be integrated in the processor 1510 .

Optionally, the encoding device can be, for example, an encoder, a terminal (including but not limited to mobile phones, cameras, drones, etc.), and the encoding device can implement the corresponding process in the encoding method 600 of the embodiment of the present application, for the sake of brevity , which will not be repeated here.

FIG. 16 provides an encoding device 1600 according to yet another embodiment of the present application, and the encoding device 1600 may include a processor 1610 .

Processor 1610, configured to: acquire image complexity information of the current frame, where the image complexity information includes transformation coefficients obtained after image kernel transformation processing (PCT processing) is performed on the pixel values of the current frame; The complexity information determines the initial quantization parameter (initial QP) of the current frame; updates the initial QP of the target frame according to the initial QP of the current frame, and the target frame is the first x frame and/or the last y of the current frame frame, x and y are positive integers greater than or equal to 1.

Optionally, in some embodiments, the processor 1610 is further configured to: update the The initial QP in the target frame; wherein, the second difference is the difference between the target byte count of the current frame and the actual coded byte count of the current frame.

Optionally, in some embodiments, the processor 1610 is further configured to: if the second difference is a positive value, the first parameter and the offset parameter used to calculate the initial QP of the current frame The difference is used as the updated first parameter; the initial QP of the target frame is calculated according to the updated first parameter; or, if the second difference is negative, it will be used to calculate the initial QP of the previous frame The sum of the first parameter and the offset parameter is used as the updated first parameter; and the initial QP of the target frame is calculated according to the updated first parameter.

Optionally, in some embodiments, if multi-channel encoding is performed on the current frame, the processor 1610 is further configured to: respectively acquire image complexity information of components of the current frame, the components including the A luma component and/or at least one chrominance component of the current frame; updating the initial QP of the target frame according to image complexity information of at least one component in the components.

Optionally, in some embodiments, the processor 1610 is further configured to: acquire multiple transformation parameters of the current frame; select one parameter from the multiple transformation parameters as the image complexity of the current frame information.

Optionally, in some embodiments, the processor 1610 is further configured to: if the updated initial QP of the target frame is smaller than a third threshold, use the third threshold as the updated target frame The initial QP of the frame; or, if the updated initial QP of the target frame is greater than a fourth threshold, the fourth threshold is used as the updated initial QP of the target frame; wherein, the The third threshold is the minimum value of the QP used for encoding, and the fourth threshold is the maximum value of the QP used for encoding.

Optionally, in some embodiments, the encoding device 1600 is applied in a JPEG XR encoding format.

Optionally, the encoding device 1600 may further include a memory 1620 . Wherein, the processor 1610 can invoke and run a computer program from the memory 1620, so as to implement the method in the embodiment of the present application.

Wherein, the memory 1620 may be an independent device independent of the processor 1610 , or may be integrated in the processor 1610 .

Optionally, the encoding device 1600 may further include a transceiver 1630 . Wherein, the transceiver 1630 may be an independent device independent of the processor 1610 , or may be integrated in the processor 1610 .

Optionally, the encoding device can be, for example, an encoder, a terminal (including but not limited to mobile phones, cameras, drones, etc.), and the encoding device can implement the corresponding process in the encoding method 1100 of the embodiment of the present application, for the sake of brevity , which will not be repeated here.

FIG. 17 is a schematic structural diagram of a chip according to an embodiment of the present application. The chip 1700 shown in FIG. 17 includes a processor 1710, and the processor 1710 can call and run a computer program from a memory, so as to implement the method in the embodiment of the present application.

Optionally, as shown in FIG. 17 , the chip 1700 may further include a memory 1720 . Wherein, the processor 1710 can invoke and run a computer program from the memory 1720, so as to implement the method in the embodiment of the present application.

Wherein, the memory 1720 may be a separate device independent of the processor 1710 , or may be integrated in the processor 1710 .

Optionally, the chip 1700 may also include an input interface 1730 . Wherein, the processor 1710 can control the input interface 1730 to communicate with other devices or chips, specifically, can obtain information or data sent by other devices or chips.

Optionally, the chip 1700 may also include an output interface 1740 . Wherein, the processor 1710 can control the output interface 1740 to communicate with other devices or chips, specifically, can output information or data to other devices or chips.

It should be understood that the chip mentioned in the embodiment of the present application may also be called a system-on-chip, a system-on-chip, a system-on-a-chip, or a system-on-a-chip.

It should be understood that the processor in the embodiment of the present application may be an integrated circuit image processing system, which has a signal processing capability. In the implementation process, each step of the above-mentioned method embodiments may be completed by an integrated logic circuit of hardware in a processor or instructions in the form of software. The above-mentioned processor can be a general-purpose processor, a digital signal processor (Digital Signal Processor, DSP), an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), an off-the-shelf programmable gate array (Field Programmable Gate Array, FPGA) or other available Program logic devices, discrete gate or transistor logic devices, discrete hardware components. Various methods, steps, and logic block diagrams disclosed in the embodiments of the present application may be implemented or executed. A general-purpose processor may be a microprocessor, or the processor may be any conventional processor, or the like. The steps of the method disclosed in connection with the embodiments of the present application may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software module can be located in a mature storage medium in the field such as random access memory, flash memory, read-only memory, programmable read-only memory or electrically erasable programmable memory, register. The storage medium is located in the memory, and the processor reads the information in the memory, and completes the steps of the above method in combination with its hardware.

It can be understood that the memory in the embodiments of the present application may be a volatile memory or a nonvolatile memory, or may include both volatile and nonvolatile memories. Among them, the non-volatile memory can be read-only memory (Read-Only Memory, ROM), programmable read-only memory (Programmable ROM, PROM), erasable programmable read-only memory (Erasable PROM, EPROM), electronically programmable Erase Programmable Read-Only Memory (Electrically EPROM, EEPROM) or Flash. The volatile memory can be Random Access Memory (RAM), which acts as external cache memory. By way of illustration and not limitation, many forms of RAM are available, such as Static Random Access Memory (Static RAM, SRAM), Dynamic Random Access Memory (Dynamic RAM, DRAM), Synchronous Dynamic Random Access Memory (Synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (Double Data Rate SDRAM, DDR SDRAM), enhanced synchronous dynamic random access memory (Enhanced SDRAM, ESDRAM), synchronous connection dynamic random access memory (Synchlink DRAM, SLDRAM ) and Direct Memory Bus Random Access Memory (Direct Rambus RAM, DR RAM). It should be noted that the memory of the systems and methods described herein is intended to include, but not be limited to, these and any other suitable types of memory.

It should be understood that the above-mentioned memory is illustrative but not restrictive. For example, the memory in the embodiment of the present application may also be a static random access memory (static RAM, SRAM), a dynamic random access memory (dynamic RAM, DRAM), Synchronous dynamic random access memory (synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (double data rate SDRAM, DDR SDRAM), enhanced synchronous dynamic random access memory (enhanced SDRAM, ESDRAM), synchronous connection Dynamic random access memory (synch link DRAM, SLDRAM) and direct memory bus random access memory (Direct Rambus RAM, DR RAM), etc. That is, the memory in the embodiments of the present application is intended to include, but not be limited to, these and any other suitable types of memory.

The memory in the embodiments of the present application may provide instructions and data to the processor. A portion of the memory may also include non-volatile random access memory. For example, the memory may also store device type information. The processor may be used to execute instructions stored in the memory, and when the processor executes the instructions, the processor may execute various steps corresponding to the terminal device in the foregoing method embodiments.

In the implementation process, each step of the above method can be completed by an integrated logic circuit of the hardware in the processor or an instruction in the form of software. The steps of the methods disclosed in connection with the embodiments of the present application may be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules in the processor. The software module can be located in a mature storage medium in the field such as random access memory, flash memory, read-only memory, programmable read-only memory or electrically erasable programmable memory, register. The storage medium is located in the memory, and the processor executes the instructions in the memory to complete the steps of the above method in combination with its hardware. To avoid repetition, no detailed description is given here.

It should also be understood that in the embodiment of the present application, the pixels in the image may be located in different rows and/or columns, wherein the length of A may correspond to the number of pixels in the same row included in A, and the height of A may be Corresponding to the number of pixels in the same column included in A. In addition, the length and height of A may also be referred to as width and depth of A, respectively, which is not limited in this embodiment of the present application.

It should also be understood that, in this embodiment of the present application, "distribution of distance from the boundary of A" may refer to being at least one pixel apart from the boundary of A, and may also be referred to as "not adjacent to the boundary of A" or "not located at the boundary of A". Boundary", which is not limited in this embodiment of the present application, where A may be an image, a rectangular area, or a sub-image, and so on.

It should also be understood that the above description of the embodiments of the present application focuses on emphasizing the differences between the various embodiments, and the same or similar points that are not mentioned can be referred to each other, and for the sake of brevity, details are not repeated here.

The embodiment of the present application also provides a computer-readable storage medium for storing computer programs.

Optionally, the computer-readable storage medium can be applied to the coding device in the embodiment of the present application, and the computer program enables the computer to execute the corresponding processes implemented by the coding device in the methods of the embodiment of the present application. For the sake of brevity, here No longer.

The embodiment of the present application also provides a computer program product, including computer program instructions.

Optionally, the computer program product can be applied to the encoding device in the embodiment of the present application, and the computer program instructions cause the computer to execute the corresponding processes implemented by the encoding device in the methods of the embodiment of the present application. For the sake of brevity, the Let me repeat.

The embodiment of the present application also provides a computer program.

Optionally, the computer program can be applied to the encoding device in the embodiment of the present application. When the computer program is run on the computer, the computer executes the corresponding processes implemented by the encoding device in the methods of the embodiment of the present application. For the sake of brevity , which will not be repeated here.

It should be understood that in this embodiment of the present application, the term "and/or" is only an association relationship describing associated objects, indicating that there may be three relationships. For example, A and/or B may mean that A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character "/" in this article generally indicates that the contextual objects are an "or" relationship.

Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the relationship between hardware and software Interchangeability. In the above description, the composition and steps of each example have been generally described according to their functions. Whether these functions are executed by hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to implement the described functions for each specific application, but such implementation should not be regarded as exceeding the scope of the present application.

Those skilled in the art can clearly understand that for the convenience and brevity of the description, the specific working process of the above-described system, device and unit can refer to the corresponding process in the foregoing method embodiment, and will not be repeated here.

In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods may be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or May be integrated into another system, or some features may be ignored, or not implemented. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices or units, and may also be electrical, mechanical or other forms of connection.

The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment of the present application.

In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

If the integrated unit is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of software products, and the computer software products are stored in a storage medium In, several instructions are included to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: various media capable of storing program codes such as U disk, mobile hard disk, ROM, RAM, magnetic disk or optical disk.

The above is only a specific embodiment of the application, but the scope of protection of the application is not limited thereto. Any person familiar with the technical field can easily think of various equivalents within the scope of the technology disclosed in the application. Modifications or replacements, these modifications or replacements shall be covered within the scope of protection of this application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

A coding method, characterized in that, comprising:

Obtain the image complexity information of each of the n1 tiles in the current frame, the image complexity information includes the transformation obtained after the pixel value of each tile is processed by image kernel transformation (PCT processing) Coefficient, n1 is a positive integer greater than or equal to 2;

Updating the quantization parameters (QP) of the n1 tiles according to the image complexity information.
The encoding method according to claim 1, wherein the updating the QPs of the n1 tiles according to the image complexity information includes:

calculating the cumulative value of the target number of bytes of the i-th tile among the n1 tiles according to the image complexity information, where i is a positive integer less than or equal to n1;

Updating the QP of the i-th tile according to the cumulative value of the target number of bytes of the i-th tile.
The encoding method according to claim 2, wherein updating the QP of the i-th tile according to the cumulative value of the target number of bytes of the i-th tile includes:

Update the QP of the i-th tile according to the absolute value of the first difference and the first threshold, the first difference is the cumulative value of the target number of bytes of the i-th tile and the i-th tile The difference between the cumulative value of the actual number of encoded bytes for tiles.
The encoding method according to claim 3, wherein updating the QP of the ith tile according to the absolute value of the first difference and the first threshold includes:

If the first difference is a positive value, use the difference between the QP of the i-1th tile and the first offset QP as the QP of the i-th tile;

If the first difference is a negative value, using the sum of the QP of the i-1th tile and the first offset QP as the QP of the i-th tile;

Wherein, the first offset QP is obtained based on the absolute value of the first difference and the first threshold.
The encoding method according to claim 3 or 4, wherein the target number of bytes of the i-th tile is related to the first information;

The first information is at least one of the following information:

The target number of bytes of the current frame, the transformation coefficient of the ith tile or the transformation coefficient of the current frame.
The encoding method according to claim 5, wherein the target byte count of the current frame is related to the second information;

The second information is at least one of the following information:

The width of the current frame, the height of the current frame, the bit depth of the current frame, the encoding format of the current frame, or the image compression ratio of the current frame.
The encoding method according to any one of claims 1 to 6, wherein the method further comprises:

Updating the QP of n2 tiles in the target frame according to the ratio of the absolute value of the second difference to the target byte number of the current frame and the size of the second threshold;

Wherein, the second difference is the difference between the target number of bytes of the current frame and the actual number of encoded bytes of the current frame, and the target frame is the first x frames and/or the last x frames of the current frame y frame, x and y are positive integers greater than or equal to 1, and n2 is a positive integer greater than or equal to 2.
The encoding method according to claim 7, wherein the n2 tiles in the target frame are updated according to the ratio of the absolute value of the second difference to the target byte number of the current frame and the size of the second threshold Slices of QP, including:

If the second difference is a positive value, the difference between the first parameter and the offset parameter used to calculate the QP of the n1 tiles is used as the updated first parameter;

Calculate the QP of the n2 tiles according to the updated first parameter;

or,

If the second difference is a negative value, the sum of the first parameter and the offset parameter used to calculate the QP of the n1 tiles is used as the updated first parameter;

Calculate the QP of the n2 tiles according to the updated first parameter.
The encoding method according to any one of claims 1 to 8, wherein if multi-channel encoding is performed on the current frame, the acquisition of the image of each of the n1 tiles in the current frame is complicated Degree information, including:

Respectively acquire image complexity information of components of each of the n1 tiles, where the components include a luminance component and/or at least one chrominance component of the current frame;

Updating the QP of n1 tiles according to the image complexity information includes:

Updating the QPs of the n1 tiles according to the image complexity information of at least one of the components.
The encoding method according to any one of claims 1 to 9, wherein the acquiring the image complexity information of each of the n1 tiles in the current frame comprises:

Obtain multiple transformation parameters of each tile;

Selecting one parameter from the plurality of transformation parameters as the image complexity information of each tile.
The encoding method according to any one of claims 1 to 10, wherein the acquiring the image complexity information of each of the n1 tiles in the current frame comprises:

After the blocks in the tile are mapped, preset transformation is performed to generate preset transformation parameters, and the image complexity information of the tile is generated according to the preset transformation parameters.
The encoding method according to claim 11, wherein the preset transformation comprises PCT transformation.
The encoding method according to any one of claims 1 to 12, wherein the method further comprises:

If the updated QP of the n1 tiles is less than a third threshold, use the third threshold as the updated QP of the n1 tiles; or,

If the updated QP of the n1 tiles is greater than a fourth threshold, use the fourth threshold as the updated QP of the n1 tiles;

Wherein, the third threshold is the minimum value of the QP used for encoding, and the fourth threshold is the maximum value of the QP used for encoding.
The encoding method according to any one of claims 1 to 13, wherein the encoding method is applied in the Joint Photographic Experts Group Extended Range encoding format (JPEG XR encoding format).
A coding method, characterized in that, comprising:

Obtaining image complexity information of the current frame, the image complexity information including transformation coefficients obtained after image kernel transformation processing (PCT processing) is performed on the pixel values of the current frame;

Determine the initial quantization parameter (initial QP) of the current frame according to the image complexity information;

Update the initial QP of the target frame according to the initial QP of the current frame, the target frame is the previous x frames and/or the subsequent y frames of the current frame, and x and y are positive integers greater than or equal to 1.
The encoding method according to claim 15, wherein the updating the initial QP of the target frame according to the initial QP of the current frame comprises:

updating the initial QP in the target frame according to the ratio of the absolute value of the second difference to the target byte number of the current frame and the size of a second threshold;

Wherein, the second difference is the difference between the target number of bytes of the current frame and the actual number of encoded bytes of the current frame.
The encoding method according to claim 16, characterized in that, updating the target frame in the target frame according to the ratio of the absolute value of the second difference to the target byte number of the current frame and the size of the second threshold The initial QP, including:

If the second difference is a positive value, the difference between the first parameter and the offset parameter used to calculate the initial QP of the current frame is used as the updated first parameter;

calculating the initial QP of the target frame according to the updated first parameter;

or,

If the second difference is a negative value, the sum of the first parameter and the offset parameter used to calculate the initial QP of the previous frame is used as the updated first parameter;

Calculate the initial QP of the target frame according to the updated first parameter.
The encoding method according to any one of claims 15 to 17, wherein if multi-channel encoding is performed on the current frame, the acquiring the image complexity information of the current frame includes:

respectively acquiring image complexity information of components of the current frame, where the components include a luma component and/or at least one chrominance component of the current frame;

Updating the initial QP of the target frame according to the initial QP of the current frame includes:

Updating the initial QP of the target frame according to the image complexity information of at least one of the components.
The encoding method according to any one of claims 15 to 18, wherein said acquiring the image complexity information of the current frame comprises:

Obtain multiple transformation parameters of the current frame;

Selecting one parameter from the plurality of transformation parameters as the image complexity information of the current frame.
The encoding method according to any one of claims 15 to 19, wherein the method further comprises:

If the updated initial QP of the target frame is less than a third threshold, using the third threshold as the updated initial QP of the target frame; or,

If the updated initial QP of the target frame is greater than a fourth threshold, use the fourth threshold as the updated initial QP of the target frame;

Wherein, the third threshold is the minimum value of the QP used for encoding, and the fourth threshold is the maximum value of the QP used for encoding.
The encoding method according to any one of claims 15 to 20, wherein the encoding method is applied in the Joint Photographic Experts Group Extended Range encoding format (JPEG XR encoding format).
An encoding device, characterized in that it comprises:

The complexity calculation module is used to obtain the image complexity information of each tile of the n1 tiles in the current frame, and the image complexity information includes the image kernel transformation process on the pixel value of each tile ( PCT processing), the transformation coefficient obtained after n1 is a positive integer greater than or equal to 2;

A code rate control module, configured to update the quantization parameters (QP) of the n1 tiles according to the image complexity information.
The encoding device according to claim 22, wherein the code rate control module is further used for:

calculating the cumulative value of the target number of bytes of the i-th tile among the n1 tiles according to the image complexity information, where i is a positive integer less than or equal to n;

Updating the QP of the i-th tile according to the cumulative value of the target number of bytes of the i-th tile.
The encoding device according to claim 23, wherein the code rate control module is further used for:

Update the QP of the i-th tile according to the absolute value of the first difference and the first threshold, the first difference is the cumulative value of the target number of bytes of the i-th tile and the i-th tile The difference between the cumulative value of the actual number of encoded bytes for tiles.
The encoding device according to claim 24, wherein the code rate control module is further used for:

If the first difference is a positive value, use the difference between the QP of the i-1th tile and the first offset QP as the QP of the i-th tile;

If the first difference is a negative value, using the sum of the QP of the i-1th tile and the first offset QP as the QP of the i-th tile;

Wherein, the first offset QP is obtained based on the absolute value of the first difference and the first threshold.
The encoding device according to claim 24 or 25, wherein the target number of bytes of the ith tile is related to the first information;

The first information is at least one of the following information:

The target number of bytes of the current frame, the transformation coefficient of the ith tile or the transformation coefficient of the current frame.
The encoding device according to claim 26, wherein the target number of bytes of the current frame is related to the second information;

The second information is at least one of the following information:

The width of the current frame, the height of the current frame, the bit depth of the current frame, the encoding format of the current frame, or the image compression ratio of the current frame.
According to the encoding device according to any one of claims 22 to 27, the code rate control module is further used for:

Updating the QP of n2 tiles in the target frame according to the ratio of the absolute value of the second difference to the target byte number of the current frame and the size of the second threshold;

Wherein, the second difference is the difference between the target number of bytes of the current frame and the actual number of encoded bytes of the current frame, and the target frame is the first x frames and/or the last x frames of the current frame y frame, x and y are positive integers greater than or equal to 1, and n2 is a positive integer greater than or equal to 2.
The encoding device according to claim 28, wherein the code rate control module is further used for:

If the second difference is a positive value, the difference between the first parameter and the offset parameter used to calculate the QP of the n1 tiles is used as the updated first parameter;

Calculate the QP of the n2 tiles according to the updated first parameter;

or,

If the second difference is a negative value, the sum of the first parameter and the offset parameter used to calculate the QP of the n1 tiles is used as the updated first parameter;

Calculate the QP of the n2 tiles according to the updated first parameter.
The encoding device according to any one of claims 22 to 29, wherein if multi-channel encoding is performed on the current frame, the complexity calculation module is further used for:

Respectively acquire image complexity information of components of each of the n1 tiles, where the components include a luminance component and/or at least one chrominance component of the current frame;

The code rate control module is further used for:

Updating the QPs of the n1 tiles according to the image complexity information of at least one of the components.
The encoding device according to any one of claims 22 to 30, wherein the complexity calculation module is further used for:

Obtain multiple transformation parameters of each tile;

Selecting one parameter from the plurality of transformation parameters as the image complexity information of each tile.
The encoding device according to any one of claims 22 to 31, wherein the code rate control module is further used for:

If the updated QP of the n1 tiles is less than a third threshold, use the third threshold as the updated QP of the n1 tiles; or,

If the updated QP of the n1 tiles is greater than a fourth threshold, use the fourth threshold as the updated QP of the n1 tiles;

Wherein, the third threshold is the minimum value of the QP used for encoding, and the fourth threshold is the maximum value of the QP used for encoding.
The encoding device according to any one of claims 22 to 32, wherein the encoding device is applied in the Joint Photographic Experts Group Extended Range encoding format (JPEG XR encoding format).
An encoding device, characterized in that it comprises:

The complexity calculation module is used to obtain the image complexity information of the current frame, and the image complexity information includes transformation coefficients obtained after image kernel transformation processing (PCT processing) is performed on the pixel values of the current frame;

A code rate control module, configured to determine an initial quantization parameter (initial QP) of the current frame according to the image complexity information;

The code rate control module is also used to: update the initial QP of the target frame according to the initial QP of the current frame, the target frame is the previous x frame and/or the next y frame of the current frame, and x and y are greater than or a positive integer equal to 1.
The encoding device according to claim 34, wherein the code rate control module is further used for:

updating the initial QP in the target frame according to the ratio of the absolute value of the second difference to the target byte number of the current frame and the size of a second threshold;

Wherein, the second difference is the difference between the target number of bytes of the current frame and the actual number of encoded bytes of the current frame.
The encoding device according to claim 35, wherein the code rate control module is further used for:

If the second difference is a positive value, the difference between the first parameter and the offset parameter used to calculate the initial QP of the current frame is used as the updated first parameter;

calculating the initial QP of the target frame according to the updated first parameter;

or,

If the second difference is a negative value, the sum of the first parameter and the offset parameter used to calculate the initial QP of the previous frame is used as the updated first parameter;

Calculate the initial QP of the target frame according to the updated first parameter.
The encoding device according to any one of claims 34 to 36, wherein, if multi-channel encoding is performed on the current frame, the complexity calculation module is further used for:

Respectively acquire image complexity information of components of the current frame, where the components include a luminance component and/or at least one chrominance component of the current frame;

The code rate control module is further used for:

Updating the initial QP of the target frame according to the image complexity information of at least one of the components.
The encoding device according to any one of claims 34 to 37, wherein the complexity calculation module is further used for:

Obtain a plurality of transformation parameters of the current frame;

Selecting one parameter from the plurality of transformation parameters as the image complexity information of the current frame.
The encoding device according to any one of claims 34 to 38, wherein the code rate control module is further used for:

If the updated initial QP of the target frame is less than a third threshold, using the third threshold as the updated initial QP of the target frame; or,

If the updated initial QP of the target frame is greater than a fourth threshold, use the fourth threshold as the updated initial QP of the target frame;

Wherein, the third threshold is the minimum value of the QP used for encoding, and the fourth threshold is the maximum value of the QP used for encoding.
The encoding device according to any one of claims 34 to 39, wherein the encoding device is applied in the Joint Photographic Experts Group Extended Range encoding format (JPEG XR encoding format).
An encoding device, characterized in that it comprises:

The processor is configured to: acquire image complexity information of each of the n1 tiles in the current frame, the image complexity information including image kernel transformation processing (PCT) for the pixel value of each tile The transformation coefficient obtained after processing), n1 is a positive integer greater than or equal to 2;

Updating the quantization parameters (QP) of the n1 tiles according to the image complexity information.
The encoding device according to claim 41, wherein the processor is further used for:

calculating the cumulative value of the target number of bytes of the i-th tile among the n1 tiles according to the image complexity information, where i is a positive integer less than or equal to n;

Updating the QP of the i-th tile according to the cumulative value of the target number of bytes of the i-th tile.
The encoding device according to claim 42, wherein the processor is further used for:

Update the QP of the i-th tile according to the absolute value of the first difference and the first threshold, the first difference is the cumulative value of the target number of bytes of the i-th tile and the i-th tile The difference between the cumulative value of the actual number of encoded bytes for tiles.
The encoding device according to claim 43, wherein the processor is further used for:

If the first difference is a positive value, use the difference between the QP of the i-1th tile and the first offset QP as the QP of the i-th tile;

If the first difference is a negative value, using the sum of the QP of the i-1th tile and the first offset QP as the QP of the i-th tile;

Wherein, the first offset QP is obtained based on the absolute value of the first difference and the first threshold.
The encoding device according to claim 43 or 44, wherein the target number of bytes of the ith tile is related to the first information;

The first information is at least one of the following information:

The target number of bytes of the current frame, the transformation coefficient of the ith tile or the transformation coefficient of the current frame.
The encoding device according to claim 45, wherein the target number of bytes of the current frame is related to the second information;

The second information is at least one of the following information:

The width of the current frame, the height of the current frame, the bit depth of the current frame, the encoding format of the current frame or the image compression ratio of the current frame.
The encoding device according to any one of claims 41 to 46, wherein the processor is further configured to:

Updating the QP of n2 tiles in the target frame according to the ratio of the absolute value of the second difference to the target byte number of the current frame and the size of the second threshold;

Wherein, the second difference is the difference between the target number of bytes of the current frame and the actual number of encoded bytes of the current frame, and the target frame is the first x frames and/or the last x frames of the current frame y frame, x and y are positive integers greater than or equal to 1, and n2 is a positive integer greater than or equal to 2.
The encoding device according to claim 47, wherein the processor is further used for:

If the second difference is a positive value, the difference between the first parameter and the offset parameter used to calculate the QP of the n1 tiles is used as the updated first parameter;

Calculate the QP of the n2 tiles according to the updated first parameter;

or,

If the second difference is a negative value, the sum of the first parameter and the offset parameter used to calculate the QP of the n1 tiles is used as the updated first parameter;

Calculate the QP of the n2 tiles according to the updated first parameter.
The encoding device according to any one of claims 41 to 48, wherein if multi-channel encoding is performed on the current frame, the processor is further configured to:

Respectively acquire image complexity information of components of each of the n1 tiles, where the components include a luminance component and/or at least one chrominance component of the current frame;

Updating the QPs of the n1 tiles according to the image complexity information of at least one of the components.
The encoding device according to any one of claims 41 to 49, wherein the processor is further configured to:

Obtain multiple transformation parameters of each tile;

Selecting one parameter from the plurality of transformation parameters as the image complexity information of each tile.
The encoding device according to any one of claims 41 to 50, wherein the processor is further configured to:

If the updated QP of the n1 tiles is less than a third threshold, use the third threshold as the updated QP of the n1 tiles; or,

If the updated QP of the n1 tiles is greater than a fourth threshold, use the fourth threshold as the updated QP of the n1 tiles;

Wherein, the third threshold is the minimum value of the QP used for encoding, and the fourth threshold is the maximum value of the QP used for encoding.
The encoding device according to any one of claims 41 to 51, wherein the encoding device is applied in the Joint Photographic Experts Group Extended Range encoding format (JPEG XR encoding format).
An encoding device, characterized in that it comprises:

Processor for:

Obtaining image complexity information of the current frame, the image complexity information including transformation coefficients obtained after image kernel transformation processing (PCT processing) is performed on the pixel values of the current frame;

Determine the initial quantization parameter (initial QP) of the current frame according to the image complexity information;

Update the initial QP of the target frame according to the initial QP of the current frame, the target frame is the previous x frames and/or the subsequent y frames of the current frame, and x and y are positive integers greater than or equal to 1.
The encoding device according to claim 53, wherein the processor is further used for:

updating the initial QP in the target frame according to the ratio of the absolute value of the second difference to the target byte number of the current frame and the size of a second threshold;

Wherein, the second difference is the difference between the target number of bytes of the current frame and the actual number of encoded bytes of the current frame.
The encoding device according to claim 54, wherein the processor is further used for:

If the second difference is a positive value, the difference between the first parameter and the offset parameter used to calculate the initial QP of the current frame is used as the updated first parameter;

calculating the initial QP of the target frame according to the updated first parameter;

or,

If the second difference is a negative value, the sum of the first parameter and the offset parameter used to calculate the initial QP of the previous frame is used as the updated first parameter;

Calculate the initial QP of the target frame according to the updated first parameter.
The encoding device according to any one of claims 53 to 55, wherein if multi-channel encoding is performed on the current frame, the processor is further used for:

respectively acquiring image complexity information of components of the current frame, where the components include a luma component and/or at least one chrominance component of the current frame;

Updating the initial QP of the target frame according to the image complexity information of at least one of the components.
The encoding device according to any one of claims 53 to 56, wherein the processor is further configured to:

Obtain a plurality of transformation parameters of the current frame;

Selecting one parameter from the plurality of transformation parameters as the image complexity information of the current frame.
The encoding device according to any one of claims 53 to 57, wherein the processor is further configured to:

If the updated initial QP of the target frame is less than a third threshold, using the third threshold as the updated initial QP of the target frame; or,

If the updated initial QP of the target frame is greater than a fourth threshold, use the fourth threshold as the updated initial QP of the target frame;

Wherein, the third threshold is the minimum value of the QP used for encoding, and the fourth threshold is the maximum value of the QP used for encoding.
The encoding device according to any one of claims 53 to 58, wherein the encoding device is applied in the Joint Photographic Experts Group Extended Range encoding format (JPEG XR encoding format).
A computer-readable storage medium, characterized by comprising program instructions, and when the program instructions are executed by a computer, the computer executes the encoding method according to any one of claims 1 to 19.