Method for encoding and rate control of cutaneous tactile signals

Abstract

This invention relates to a method for encoding and controlling the bit rate of tactile signals. It includes: dividing the original signal sequence into data blocks of a specific length of 64 samples; converting these data blocks into an 8×8 matrix using a zigzag scanning method; applying a two-dimensional discrete cosine transform (DCT) to transform the data from the time domain to the frequency domain, while employing a non-uniform quantization strategy for the DCT coefficients at different frequency positions; encoding the quantized data blocks using run-length encoding and variable-length integer encoding; accurately fitting the R-Q curve using a linear regression method; dynamically allocating the bit rate by analyzing frame complexity; and deriving the optimal quantization step size for each frame based on the target bit rate. This invention achieves low-latency transmission while ensuring high signal fidelity, optimizes bandwidth resource allocation, and effectively solves the core problems of large data volume and insufficient bit rate control in tactile signal encoding.

Description

Technical Field

[0001] This invention relates to the field of tactile signal encoding technology, specifically to a method for encoding and controlling the bit rate of tactile signals. Background Technology

[0002] With the advancement of multimedia technology, humans are no longer satisfied with visual and auditory interactions and are beginning to pursue more immersive experiences. Haptic feedback, as a new interactive medium, adds a new dimension to human-computer interaction. By simulating physical sensations such as pressure, vibration, temperature, and texture, haptic feedback allows users to gain a richer sensory experience during interaction. Compared to traditional visual and auditory media, haptic feedback can directly convey information such as the surface texture, motion trajectory, and physical contact of objects, enhancing the user's perception of the virtual world. For example, in virtual games, users can feel the weight of virtual weapons and the force of collisions, and even subtle vibrations in the environment, through haptic devices, increasing the game's immersion and realism.

[0003] In fields such as medicine, education, and industrial design, haptic feedback offers new possibilities for tasks such as remote operation, training simulation, and design verification. The tactile signals are primarily acquired through electromyography (EMG) sensors, piezoresistive tactile sensors, capacitive tactile sensors, inertial measurement units (IMUs), and piezoelectric vibration sensors. EMG sensors capture electrical signals generated by muscle contraction, reflecting muscle activity. Piezoresistive and capacitive tactile sensors detect mechanical signals such as contact pressure and texture. Inertial measurement units acquire position and acceleration information related to motion posture. Piezoelectric vibration sensors collect tactile feedback signals such as vibration frequency and intensity. However, to accurately and comprehensively reproduce the full picture and details of human tactile perception, multiple tactile interaction points and multi-degree-of-freedom acquisition links need to be deployed using a combination of these sensors. This leads to an exponential increase in the amount of tactile signal data, posing significant challenges to data transmission, processing, and storage. Therefore, effectively encoding tactile signals has become an important research topic. Summary of the Invention

[0004] In view of the above problems, this application provides a method for encoding and controlling the bit rate of tactile signals to solve the technical problems of delayed and poor rate-distortion performance of tactile signals.

[0005] To achieve the above objectives, this application provides a method for encoding and controlling the bit rate of tactile signals, comprising the following steps:

[0006] Step S1: Divide the original tactile signal sequence into data blocks of 64 samples each, and convert each data block into an 8×8 matrix by zigzag scanning. The 8×8 matrix is ​​used as the input data for the encoding process.

[0007] Step S2: Apply two-dimensional discrete cosine transform to the 8×8 matrix form of the tactile signal to transform it from the time domain to the frequency domain, and then apply a non-uniform quantization strategy to the DCT coefficients at different frequency positions.

[0008] Step S3: Compress the quantized tactile signal by sequentially applying run-length encoding and variable-length integer encoding;

[0009] Step S4: Divide each 8×8 matrix into frames according to a preset number. Based on the bit overrun or saving of the previous frame group, and combined with the total number of bits, the number of bits consumed by encoding, the total number of unencoded frames, and the number of frames per frame group, dynamically adjust the target number of bits for the current frame group. Use variance to characterize the frame complexity of the tactile signal. Distribute the target number of bits for the current frame group to each frame according to the ratio of the variance of each frame to the average variance of the frame group to obtain the target number of bits per frame. Based on the historical encoded bitrate and quantization parameter data, fit a quadratic RQ curve through linear regression. According to the bitrate corresponding to the target number of bits per frame, solve the RQ curve in reverse to obtain the matching optimal scaling factor. Map this scaling factor to the optimal quantization step size corresponding to each frame. Match the optimal scaling factor through a preset 32-item scaling factor index table and limit the variation of the scaling factor index to stabilize the encoding quality.

[0010] Furthermore, in step S4, the dynamic adjustment of the target bit count of the frame group satisfies:

[0011] ;

[0012] in, The target bits for the current frame group, and The total number of bits and the number of bits consumed in encoding. This represents the total number of uncoded frames. The number of frames in each frame group.

[0013] Furthermore, in step S4, the frame complexity is characterized by the variance of the tactile signal. The proportional allocation of the target bits per frame is as follows: the target bits of the current i-th frame are equal to the target bits of the current frame group multiplied by the ratio of the variance of the current i-th frame to the average variance of the current frame group, thereby realizing the reasonable allocation of bit resources based on frame complexity.

[0014] Furthermore, in step S4, the quadratic RQ curve satisfies:

[0015] ;

[0016] A and b are model parameters determined through linear regression analysis;

[0017] ;

[0018] ;

[0019] in, This represents the number of frames observed in the past. The average scaling factor for the actual encoding in the past. This represents the average number of bits in past actual codes.

[0020] Furthermore, in step S4, the estimated scaling factor value Q is first obtained by back-calculating based on the relationship f(R,Q) between the code R and the quantization parameter Q corresponding to the RQ model, combined with the code rate corresponding to the target number of bits per frame. est The optimal scaling factor is then determined using the following formula:

[0021] ;

[0022] Among them, Q list This is a list of 32 scaling factors, where Q is the optimal scaling factor for the match.

[0023] Furthermore, in step S4, the index change of the scaling factor satisfies the following constraints:

[0024] ;

[0025] in, This is the index of the average scaling factor in past actual coding, where T is a preset change threshold, and T is an integer from 1 to 3.

[0026] Furthermore, in step S1, the zigzag scanning sequence is as follows: starting from the top left corner of the 8×8 matrix, fill the matrix diagonally step by step. The first diagonal is the matrix position (0,0), the second diagonal is the matrix positions (0,1) and (1,0), the third diagonal is the matrix positions (2,0), (1,1) and (0,2), and so on until the entire matrix is ​​filled.

[0027] Furthermore, in step S2, the two-dimensional discrete cosine transform can remove the correlation between the tactile signals in the horizontal and vertical directions, concentrate the signal energy in the low-frequency region, and improve data sparsity to optimize compression performance.

[0028] Furthermore, in step S3, run-length encoding is used to compress the continuous zero values ​​in the high-frequency part of the quantized tactile signal, and variable-length integer encoding is used to compress the non-zero DCT coefficients. The DCT coefficients are rearranged by zigzag scanning to concentrate the continuous zero values ​​and improve compression efficiency.

[0029] Unlike existing technologies, the above-mentioned haptic signal encoding and bitrate control method includes: Step S1, the original signal sequence is divided into data blocks of 64 samples of a specific length and zigzagged; Step S2, the encoder applies two-dimensional discrete cosine transform to transform the data from the time domain to the frequency domain, and simultaneously adopts a non-uniform quantization strategy for the DCT coefficients at different frequency positions; Step S3, run-length encoding and variable-length integer encoding are applied to encode the quantized data blocks; Step S4, linear regression is used to accurately fit the RQ curve, and the bitrate is dynamically allocated by analyzing frame complexity. The optimal quantization step size for each frame is derived in reverse by combining the target bitrate. This invention concentrates signal energy through block-to-matrix conversion, DCT transformation, and non-uniform quantization, and efficiently compresses haptic signals by combining run-length encoding and variable-length encoding, significantly reducing the data volume; by fitting the RQ curve through linear regression, dynamically allocating the bitrate, and inversely deriving the optimal quantization step size, it achieves low-latency transmission while ensuring high signal fidelity, optimizes bandwidth resource allocation, and effectively solves the core problems of large data volume and insufficient bitrate control in haptic signal encoding.

[0030] The above description of the invention is merely an overview of the technical solution of this application. In order to enable those skilled in the art to better understand the technical solution of this application and to implement it based on the description and drawings, and to make the above-mentioned objectives and other objectives, features and advantages of this application easier to understand, the following description is provided in conjunction with the specific embodiments and drawings of this application. Attached Figure Description

[0031] The accompanying drawings are only used to illustrate the principles, implementation methods, applications, features, and effects of specific embodiments of the present invention and other related contents, and should not be considered as limitations on this application.

[0032] In the accompanying drawings of the instruction manual:

[0033] Figure 1 This is a flowchart of the muscle sensory tactile signal encoding and bit rate control method described in a specific implementation;

[0034] Figure 2 This is a schematic diagram of the processing flow of the muscle sensation tactile signal encoding and bit rate control method described in the specific implementation embodiment;

[0035] Figure 3 This is a schematic diagram of the zigzag scanning of the data block as described in the specific implementation method;

[0036] Figure 4 The flowchart of the bitrate control algorithm described in the specific implementation method is shown below; Detailed Implementation

[0037] To illustrate the possible application scenarios, technical principles, implementable specific solutions, and achievable objectives and effects of this application in detail, the following description, in conjunction with the listed specific embodiments and accompanying drawings, provides a detailed explanation. The embodiments described herein are merely illustrative of the technical solutions of this application and are therefore intended to limit the scope of protection of this application.

[0038] In this document, the term "embodiment" means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The term "embodiment" appearing in various places throughout the specification does not necessarily refer to the same embodiment, nor does it specifically limit its independence or connection with other embodiments. In principle, in this application, as long as there are no technical contradictions or conflicts, the technical features mentioned in each embodiment can be combined in any way to form corresponding implementable technical solutions.

[0039] Unless otherwise defined, the technical terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the use of related terms herein is merely for the purpose of describing particular embodiments and is not intended to limit this application.

[0040] In the description of this application, the term "and / or" is used to describe the logical relationship between objects, indicating that three relationships can exist. For example, A and / or B means: A exists, B exists, and A and B exist simultaneously. Additionally, the character " / " in this document generally indicates that the preceding and following objects have an "or" logical relationship.

[0041] In this application, terms such as “first” and “second” are used only to distinguish one entity or operation from another, and do not necessarily require or imply any actual quantity, hierarchy or order relationship between these entities or operations.

[0042] Without further limitations, the use of terms such as “comprising,” “including,” “having,” or other similar open-ended expressions in this application is intended to cover non-exclusive inclusion, which does not exclude the presence of additional elements in a process, method, or product that includes the stated elements, such that a process, method, or product that includes a list of elements may include not only those defined elements but also other elements not expressly listed, or elements inherent to such a process, method, or product.

[0043] As understood in the Examination Guidelines, in this application, expressions such as "greater than," "less than," and "exceeding" are understood to exclude the stated number; expressions such as "above," "below," and "within" are understood to include the stated number. Furthermore, in the description of the embodiments in this application, "multiple" means two or more (including two), and similar expressions related to "multiple" are also understood in this way, such as "multiple groups" and "multiple times," unless otherwise explicitly specified.

[0044] In the description of the embodiments of this application, the space-related expressions used, such as "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "vertical," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential," indicate the orientation or positional relationship based on the orientation or positional relationship shown in the specific embodiments or drawings. They are only for the purpose of describing the specific embodiments of this application or for the reader's understanding, and do not indicate or imply that the device or component referred to must have a specific position, a specific orientation, or be constructed or operated in a specific orientation. Therefore, they should not be construed as limitations on the embodiments of this application.

[0045] Unless otherwise expressly specified or limited, the terms "installation," "connection," "linking," "fixing," and "setting," as used in the description of the embodiments of this application, should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral setting; it can be a mechanical connection, an electrical connection, or a communication connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be the internal connection of two components or the interaction between two components. For those skilled in the art to which this application pertains, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.

[0046] Please see Figures 1 to 4 This embodiment provides a method for encoding and controlling the bit rate of haptic signals. This method can be widely applied to scenarios such as VR / AR immersive interaction, telemedicine, industrial remote control, and haptic simulation training. It specifically addresses the core problems of existing technologies, such as large data volume, high transmission latency, lack of bit rate control, and insufficient fidelity in haptic signals. This embodiment achieves low-latency transmission and high-fidelity signal restoration through efficient encoding compression and dynamic bit rate allocation, while optimizing bandwidth resource allocation, thus supporting the practical application of haptic interaction technologies in various fields.

[0047] like Figure 1 and Figure 2 As shown, the method for encoding and controlling the bit rate of tactile signals in this embodiment includes the following steps:

[0048] Step S1: Divide the original tactile signal sequence into data blocks of 64 samples each, and convert each data block into an 8×8 matrix by zigzag scanning. The 8×8 matrix is ​​used as the input data for the encoding process.

[0049] Step S2: Apply two-dimensional discrete cosine transform to the 8×8 matrix form of the tactile signal to transform it from the time domain to the frequency domain, and then apply a non-uniform quantization strategy to the DCT coefficients at different frequency positions.

[0050] Step S3: Compress the quantized tactile signal by sequentially applying run-length encoding and variable-length integer encoding;

[0051] Step S4: Divide each 8×8 matrix into frames according to a preset number. Based on the bit overrun or saving of the previous frame group, and combined with the total number of bits, the number of bits consumed by encoding, the total number of unencoded frames, and the number of frames per frame group, dynamically adjust the target number of bits for the current frame group. Use variance to characterize the frame complexity of the tactile signal. Distribute the target number of bits for the current frame group to each frame according to the ratio of the variance of each frame to the average variance of the frame group to obtain the target number of bits per frame. Based on the historical encoded bitrate and quantization parameter data, fit a quadratic RQ curve through linear regression. According to the bitrate corresponding to the target number of bits per frame, solve the RQ curve in reverse to obtain the matching optimal scaling factor. Map this scaling factor to the optimal quantization step size corresponding to each frame. Match the optimal scaling factor through a preset 32-item scaling factor index table and limit the variation of the scaling factor index to stabilize the encoding quality.

[0052] The tactile signals can be acquired by a data acquisition system consisting of electromyography (EMG) sensors, piezoresistive tactile sensors, capacitive tactile sensors, inertial measurement units (IMUs), and piezoelectric vibration sensors. The original tactile signal sequence is then scanned in a zigzag pattern.

[0053] like Figure 3 The diagram illustrates how a zigzag scan converts each data block into an 8×8 matrix. In step S1, the zigzag scan sequence is as follows: starting from the top left corner of the 8×8 matrix, the matrix is ​​filled diagonally. The first diagonal is the matrix position (0,0), the second diagonal is the matrix positions (0,1) and (1,0), the third diagonal is the matrix positions (2,0), (1,1), and (0,2), and so on until the entire matrix is ​​filled. The specific scanning sequence is as follows: starting from the top left corner, the matrix is ​​filled diagonally, the first diagonal is... The second diagonal is , The third diagonal is , , And so on until it is filled. matrix.

[0054] In step S2, after the tactile signal is converted into a two-dimensional matrix, the two-dimensional DCT can transform the data in both the horizontal and vertical directions, and remove the correlation in these two directions, making the transformed data more sparse, that is, most coefficients are zero or close to zero, thereby improving the compression performance.

[0055] In step S3, run-length encoding is used to compress the continuous zero values ​​in the high-frequency part of the quantized tactile signal, and variable-length integer encoding is used to compress the non-zero DCT coefficients. Furthermore, the DCT coefficients are rearranged through zigzag scanning to concentrate the continuous zero values ​​and improve compression efficiency.

[0056] In step S3, the AC coefficients after two-dimensional discrete cosine transform (DCT) and quantization often contain a large number of zeros. This is because DCT concentrates most of the signal's energy in the low-frequency range, while the high-frequency range typically contains less information. Therefore, the large number of zeros in the high-frequency coefficients can be utilized to improve compression efficiency. To maximize the utilization of the zeros in the AC coefficients, a zigzag scan is used to rearrange the data. The zigzag scan reads the DCT coefficients sequentially according to their frequency, allowing the large number of zeros in the high-frequency coefficients to be concentrated together as much as possible. Run-length encoding is used to compress these consecutive zeros, and variable-length integer encoding is used to further compress the non-zero coefficients.

[0057] In step S4, each 8×8 matrix is ​​defined as a frame, and the size of the frame group is set to N. The bitrate of each frame group is initialized. Ideally, the bits consumed by the previously encoded frame groups are the initially allocated bits, so the target bitrate of each frame group is the initially allocated bitrate. However, due to factors such as changes in data complexity and error accumulation, it is difficult to consistently and accurately achieve the preset target bitrate during actual encoding. Therefore, when the actual number of bits in the current frame group exceeds or decreases, this method dynamically adjusts the target number of bits in the current frame group to achieve stable control of the overall bitrate and reasonable allocation of resources. The dynamic adjustment of the target number of bits in the frame group satisfies:

[0058] ;

[0059] in, The target bits for the current frame group, and The total number of bits and the number of bits consumed in encoding. This represents the total number of uncoded frames. The number of frames in each frame group.

[0060] Since the content and complexity of each frame signal are different, generally speaking, the more complex the signal, the more bits are needed for encoding. Therefore, in this embodiment, the bits of each frame signal are allocated proportionally according to the complexity in the frame group.

[0061] In step S4, the frame complexity is calculated using the variance of the tactile signals. The proportional allocation of target bits per frame is represented as follows: the target bit count of the current i-th frame is equal to the target bit count of the current frame group multiplied by the ratio of the variance of the current i-th frame to the average variance of the current frame group, thus achieving a reasonable allocation of bit resources based on frame complexity.

[0062] Suppose there is a variance of For a Gaussian source, rate-distortion theory defines the relationship between distortion D and bit rate R. Gaussian sources possess ideal statistical properties, and their rate-distortion function is often considered a theoretical upper bound on the optimal bit rate. According to Shannon's rate-distortion function, the minimum bit rate of a Gaussian source with a given mean squared error distortion D is expressed by the following formula:

[0063] ;

[0064] This formula describes the variance of a Gaussian source under ideal conditions. The relationship between variance and distortion D. When the variance of a signal increases, it means that the amplitude and dynamic range of the signal increase. Encoding it with the same distortion level D requires more bits, i.e., a higher bit rate. In this embodiment, the variance of each frame's signal is normalized, and the target bits for the current frame are obtained according to the following formula, where... Indicates the current number Frame variance This represents the average variance of the current frame group. Indicates the current frame bit allocation ratio. Indicates the target bits of the current frame.

[0065] ;

[0066] ;

[0067] ;

[0068] The coderate control for this chapter will be performed using the quadratic model shown in the following formula:

[0069] ;

[0070] Using previously encoded data and linear regression analysis, the model parameters are determined using the following formula. and This method allows for a more accurate fit to the data. Among other things, This represents the number of frames observed in the past. , The average scaling factor and number of bits for past actual encoding.

[0071] Therefore, in step S4, the quadratic RQ curve satisfies:

[0072] ;

[0073] and These are the model parameters determined through linear regression analysis;

[0074] ;

[0075] ;

[0076] in, This represents the number of frames observed in the past. The average scaling factor for the actual encoding in the past. This represents the average number of bits in past actual codes.

[0077] In this embodiment, a scaling factor index table was designed and constructed. Each time encoding occurs, only the index value of the scaling factor in this table needs to be written into the bitstream to accurately restore the required scaling factor. The table is designed to contain 32 candidate options, thus requiring only 5 bits to complete index encoding. To achieve precise control of the target bitrate, the encoding process requires deriving a suitable scaling factor based on a pre-established RQ model and the target bitrate. After deriving the optimal scaling factor corresponding to the target bitrate through the RQ model, the closest entry in the scaling factor index table is searched and used to replace the floating-point output in the model. The specific calculation process is shown in the following formula:

[0078] ;

[0079] ;

[0080] Where f(R,Q) represents the relationship between the code R and the quantization parameter Q corresponding to the RQ model. Represents the estimated scaling factor value, Indicates the determination of the scaling factor The index corresponding to the value Indicates the scaling factor List.

[0081] Therefore, in step S4, the estimated scaling factor value Q is first obtained by back-calculating based on the relationship f(R,Q) between the code R and the quantization parameter Q corresponding to the RQ model, combined with the code rate corresponding to the target number of bits per frame. est The optimal scaling factor is then determined using the following formula:

[0082] ;

[0083] Among them, Q list This is a list of 32 scaling factors, where Q is the optimal scaling factor for the match.

[0084] To avoid excessive fluctuations in encoding quality, the scaling factor The value cannot change drastically and should satisfy the following formula constraint, where, The average scaling factor in past actual coding The index.

[0085] .

[0086] This formula restricts the range of variation of the scaling factor index, requiring the index of the current optimal scaling factor.

[0087] idx Must be in the "historical average scaling factor index" The purpose of keeping the scaling factor within the ±1 range is to avoid drastic fluctuations and ensure stable coding quality.

[0088] For the interval of the above inequality, and idx Since the absolute value of is no more than 1, we can perform an equivalent transformation on the above inequality to obtain the following: In step S4, the index change of the scaling factor satisfies the constraint condition:

[0089] ;

[0090] in, The average scaling factor in past actual coding The index, i.e., the one mentioned above. T is a preset change threshold, where T is an integer from 1 to 3.

[0091] In the haptic signal encoding and rate control method of this embodiment, the signal energy is concentrated by block-to-matrix conversion, DCT transformation and non-uniform quantization, and combined with run-length encoding and variable-length encoding to efficiently compress the haptic signal and significantly reduce the amount of data. By fitting the RQ curve with linear regression, dynamically allocating the rate of data and deriving the optimal quantization step size in reverse, low-latency transmission is achieved while ensuring high signal fidelity, optimizing bandwidth resource allocation, and effectively solving the core problems of large data volume and insufficient rate control in haptic signal encoding.

[0092] Finally, it should be noted that although the above embodiments have been described in the text and drawings of this application, this should not limit the scope of patent protection of this application. Any technical solutions that are based on the essential concept of this application and utilize the content described in the text and drawings of this application, resulting in equivalent structural or procedural substitutions or modifications, as well as the direct or indirect application of the technical solutions of the above embodiments to other related technical fields, are all included within the scope of patent protection of this application.

Claims

1. A method for encoding and controlling the bit rate of tactile signals, characterized in that, Includes the following steps: Step S1: Divide the original tactile signal sequence into data blocks of 64 samples each, and convert each data block into an 8×8 matrix by zigzag scanning. The 8×8 matrix is ​​used as the input data for the encoding process. Step S2: Apply two-dimensional discrete cosine transform to the 8×8 matrix form of the tactile signal to transform it from the time domain to the frequency domain, and then apply a non-uniform quantization strategy to the DCT coefficients at different frequency positions. Step S3: Compress the quantized tactile signal by sequentially applying run-length encoding and variable-length integer encoding; Step S4: Divide each 8×8 matrix into frames according to a preset number. Based on the bit overrun or saving situation of the previous frame group, combined with the total number of bits, the number of bits consumed by encoding, the total number of unencoded frames and the number of frames in each frame group, dynamically adjust the target number of bits in the current frame group. Variance is used to characterize the frame complexity of tactile signals. The target number of bits in the current frame group is allocated to each frame according to the ratio of the variance of each frame to the average variance of the frame group to obtain the target number of bits per frame. Based on the bitrate and quantization parameter data of historical encoding, a quadratic RQ curve is fitted by linear regression. According to the bitrate corresponding to the target number of bits per frame, the RQ curve is solved in reverse to obtain the matching optimal scaling factor. This scaling factor is mapped to the optimal quantization step size corresponding to each frame. The optimal scaling factor is matched by a preset 32-item scaling factor index table, and the variation of the scaling factor index is limited to stabilize the encoding quality.

2. The method for encoding and controlling the bit rate of tactile signals according to claim 1, characterized in that, In step S4, the dynamic adjustment of the target bit count of the frame group satisfies: ； in, The target bits for the current frame group, and The total number of bits and the number of bits consumed in encoding. This represents the total number of uncoded frames. The number of frames in each frame group.

3. The method for encoding and controlling the bit rate of tactile signals according to claim 1, characterized in that, In step S4, frame complexity is characterized by the variance of the tactile signal. The proportional allocation of the target bits per frame is as follows: the target bits of the current i-th frame are equal to the target bits of the current frame group multiplied by the ratio of the variance of the current i-th frame to the average variance of the current frame group, thereby achieving reasonable allocation of bit resources based on frame complexity.

4. The method for encoding and controlling the bit rate of tactile signals according to claim 1, characterized in that, In step S4, the quadratic RQ curve satisfies: ； A and b are model parameters determined through linear regression analysis; ； ； in, This represents the number of frames observed in the past. The average scaling factor for the actual encoding in the past. This represents the average number of bits in past actual codes.

5. The method for encoding and controlling the bit rate of tactile signals according to claim 1, characterized in that, In step S4, the estimated scaling factor value Q is first obtained by back-calculating based on the relationship f(R,Q) between the code R and the quantization parameter Q corresponding to the RQ model, combined with the code rate corresponding to the target number of bits per frame. est The optimal scaling factor is then determined using the following formula: ； Among them, Q list This is a list of 32 scaling factors, where Q is the optimal scaling factor for the match.

6. The method for encoding and controlling the bit rate of tactile signals according to claim 5, characterized in that, In step S4, the index change of the scaling factor satisfies the following constraints: ； in, This is the index of the average scaling factor in past actual coding, where T is a preset change threshold, and T is an integer from 1 to 3.

7. The method for encoding and controlling the bit rate of tactile signals according to claim 1, characterized in that, In step S1, the zigzag scanning sequence is as follows: starting from the top left corner of the 8×8 matrix, fill the matrix diagonally step by step. The first diagonal is the matrix position (0,0), the second diagonal is the matrix positions (0,1) and (1,0), the third diagonal is the matrix positions (2,0), (1,1) and (0,2), and so on until the entire matrix is ​​filled.

8. The method for encoding and controlling the bit rate of tactile signals according to claim 1, characterized in that, In step S2, the two-dimensional discrete cosine transform can remove the correlation between the tactile signals in the horizontal and vertical directions, concentrate the signal energy in the low-frequency region, and improve data sparsity to optimize compression performance.

9. The method for encoding and controlling the bit rate of tactile signals according to claim 1, characterized in that, In step S3, run-length encoding is used to compress the continuous zero values ​​in the high-frequency part of the quantized tactile signal, and variable-length integer encoding is used to compress the non-zero DCT coefficients. Furthermore, the DCT coefficients are rearranged through zigzag scanning to concentrate the continuous zero values ​​and improve compression efficiency.

Images