A Compression Method, Device and Storage Method for Correction Coefficient

Abstract

The present invention relates to the field of correction coefficient processing, and particularly to a compression method and a storage method for correction coefficients. The compression method is used to compress correction coefficients, and the compression method includes: obtaining a preset target number of bits Q for the main coefficient and a preset target number of bits R for the complementary coefficient; compressing the main coefficient to be compressed from a P-bit fixed-point number to a Q-bit fixed-point number, or compressing it into Q-bit encoded data composed of an A-bit first exponent part and a B-bit first fixed-point decimal part; compressing the complementary coefficient to be compressed from a P-bit fixed-point number into R-bit encoded data composed of a C-bit second exponent part and a D-bit second fixed-point decimal part. By using the compression method for correction coefficients provided by the present invention, the data storage number of bits of the correction coefficients can be effectively reduced, and further, when storing the compressed correction coefficients in the dynamic random access memory of the receiving card, the consumption of bandwidth and storage capacity can be reduced.

Description

Technical Field

[0001] The present invention relates to the field of correction coefficient processing, and in particular, to a method and apparatus for compressing correction coefficients and a storage method thereof. Background Art

[0002] Currently, during the process of using correction coefficients for calibration, after the host computer calculates the correction coefficients, the correction coefficients will be sent to the receiving card. The receiving card reads the correction coefficients stored in the dynamic random access memory such as SDRAM in the receiving card. Currently, the number of data bits used for the correction coefficients is relatively large, that is, the required data storage space is relatively large. For dynamic random access memories such as SDRAM, the bandwidth and storage capacity consumption are relatively large, and the load-carrying capacity of SDRAM and the like may also be affected to a certain extent.

[0003] For example, if the correction coefficient is 14 bits, and each pixel point corresponds to a total of 9 correction coefficients, at this time, storing the correction coefficients of one pixel point requires 14 * 9 = 126 bits of storage space. For a display screen with a load-carrying resolution of 512 * 512, a total of 2G of bandwidth is required. For a display screen with a load-carrying resolution of 1024 * 512, a total of 4G of bandwidth is required.

[0004] Therefore, those skilled in the art urgently need to find a new technical solution to solve the above problems. Summary of the Invention

[0005] In view of the above problems, the present invention provides a method for compressing correction coefficients for compressing the correction coefficients of an LED display screen. The correction coefficients include a main coefficient and a supplementary coefficient, and the main coefficient and the supplementary coefficient are represented in the form of fixed-point numbers with a data storage bit number of P bits. The method includes:

[0006] Obtaining a preset target bit number Q of the main coefficient and a preset target bit number R of the supplementary coefficient, where 0 < Q < P and 0 < R < P;

[0007] Obtaining the main coefficient to be compressed, and compressing the main coefficient to be compressed from a fixed-point number of P bits to a fixed-point number of Q bits, or compressing the main coefficient to be compressed from a fixed-point number of P bits to encoded data with a data storage bit number of Q bits composed of a first exponent part of A bits and a first fixed-point decimal part of B bits, where A + B = Q and A ≤ B;

[0008] Obtaining the supplementary coefficient to be compressed, and compressing the supplementary coefficient to be compressed from a fixed-point number of P bits to encoded data with a data storage bit number of R bits composed of a second exponent part of C bits and a second fixed-point decimal part of D bits, where C + D = R and C ≤ D.

[0009] Further, compressing the main coefficient to be compressed from a fixed-point number of P bits to a fixed-point number of Q bits includes:

[0010] According to the preset target number of bits Q of the main coefficient, delete the last P - Q bits of the P - bit main coefficient to be compressed, so as to compress the main coefficient to be compressed from a P - bit fixed - point number to a Q - bit fixed - point number.

[0011] Further, compressing the main coefficient to be compressed from a P - bit fixed - point number into encoded data with a data storage bit number of Q bits, which consists of a first exponent part of A bits and a first fixed - point decimal part of B bits, includes:

[0012] Obtain a preset main coefficient exponent encoding - exponent value mapping table, which includes multiple preset main coefficient exponent encodings for representing multiple preset main coefficient exponent values. Among them, one main coefficient exponent value corresponds to one main coefficient exponent encoding;

[0013] According to the position S where the first significant digit of the main coefficient to be compressed is located, determine that the target exponent value of the main coefficient of the main coefficient to be compressed is 1 - S;

[0014] Obtain the main coefficient exponent encoding corresponding to the target exponent value of the main coefficient from the main coefficient exponent encoding - exponent value mapping table as the first exponent part;

[0015] According to the data storage bit number A of the first exponent part and the target bit number Q of the main coefficient, determine the data storage bit number B of the first fixed - point decimal part, where B = Q - A;

[0016] According to the data storage bit number B of the first fixed - point decimal part and the target exponent value 1 - S of the main coefficient, determine the target decimal value of the main coefficient of the main coefficient to be compressed, and use the target decimal value of the main coefficient as the first fixed - point decimal part.

[0017] Further, the preset main coefficient exponent encoding - exponent value mapping table represents different main coefficient exponent values in a fixed - bit exponent encoding form or a dynamic - bit exponent encoding form.

[0018] Further, compressing the complementary coefficient to be compressed from a P - bit fixed - point number into encoded data with a data storage bit number of R bits, which consists of a second exponent part of C bits and a second fixed - point decimal part of D bits, includes:

[0019] Obtain a preset first complementary coefficient exponent encoding - exponent value mapping table, which includes multiple preset complementary coefficient exponent encodings for representing multiple preset complementary coefficient exponent values. Among them, one complementary coefficient exponent value corresponds to one complementary coefficient exponent encoding;

[0020] According to the position T where the first significant digit of the complementary coefficient to be compressed is located, determine that the target exponent value of the complementary coefficient of the complementary coefficient to be compressed is 1 - T;

[0021] Obtain the complementary coefficient exponent code corresponding to the complementary coefficient target exponent value from the first complementary coefficient exponent code-exponent value mapping table as the second exponent part;

[0022] Determine the data storage bit number D of the first fixed-point decimal part according to the data storage bit number C of the second exponent part and the complementary coefficient target bit number R, where D = R - C;

[0023] Determine the complementary coefficient target decimal value of the complementary coefficient to be compressed according to the data storage bit number D of the second fixed-point decimal part and the main coefficient target exponent value 1 - T, and use the complementary coefficient target decimal value as the second fixed-point decimal part.

[0024] Further, the complementary coefficient includes a blue complementary coefficient, a red complementary coefficient, and a green complementary coefficient.

[0025] Obtaining the preset complementary coefficient target bit number R includes: obtaining the preset blue complementary coefficient target bit number R1 and obtaining the preset red complementary coefficient and green complementary coefficient target bit number R2, where 0 < R2 < R1 < P;

[0026] Obtaining the complementary coefficient to be compressed and compressing the complementary coefficient to be compressed from a P-bit fixed-point number into an encoded data with a data storage bit number of R bits composed of a C-bit second exponent part and a D-bit second fixed-point decimal part includes:

[0027] Obtain the preset second complementary coefficient exponent code-exponent value mapping table;

[0028] Obtain the complementary coefficient to be compressed. If the complementary coefficient to be compressed is a blue complementary coefficient, according to the second complementary coefficient exponent code-exponent value mapping table and the blue complementary coefficient target bit number R1, compress the complementary coefficient to be compressed from a P-bit fixed-point number into an encoded data with a data storage bit number of R1 bits composed of a C1-bit second exponent part and a D1-bit second fixed-point decimal part, where C1 + D1 = R1 and C1 ≤ D1;

[0029] If the complementary coefficient to be compressed is a red complementary coefficient or a green complementary coefficient, according to the second complementary coefficient exponent code-exponent value mapping table and the red complementary coefficient and green complementary coefficient target bit number R2, compress the complementary coefficient to be compressed from a P-bit fixed-point number into an encoded data with a data storage bit number of R2 bits composed of a C2-bit second exponent part and a D2-bit second fixed-point decimal part, where C2 + D2 = R2, C2 ≤ D2, and D1 > D2.

[0030] The present invention also provides a correction coefficient compression device for compressing correction coefficients. The correction coefficients include a main coefficient and a supplementary coefficient, and the main coefficient and the supplementary coefficient are represented in the form of fixed-point numbers with a data storage bit number of P bits. The device includes an acquisition module, a main coefficient compression module, and a supplementary coefficient compression module, where:

[0031] The acquisition module is connected to the main coefficient compression module and the supplementary coefficient compression module, and is used to acquire a preset target bit number Q of the main coefficient and a preset target bit number R of the supplementary coefficient, where 0 < Q < P and 0 < R < P;

[0032] The main coefficient compression module is used to acquire the main coefficient to be compressed, compress the main coefficient to be compressed from a fixed-point number of P bits to a fixed-point number of Q bits, or compress the main coefficient to be compressed from a fixed-point number of P bits to encoded data with a data storage bit number of Q bits, which is composed of a first exponent part of A bits and a first fixed-point decimal part of B bits, where A + B = Q and A ≤ B;

[0033] The supplementary coefficient compression module is used to acquire the supplementary coefficient to be compressed, and compress the supplementary coefficient to be compressed from a fixed-point number of P bits to encoded data with a data storage bit number of R bits, which is composed of a second exponent part of C bits and a second fixed-point decimal part of D bits, where C + D = R and C ≤ D.

[0034] The present invention also provides a storage method for correction coefficients. The method includes:

[0035] Receiving the correction coefficients of each pixel point of the LED display screen;

[0036] Compressing the correction coefficients of each pixel point of the LED display screen according to the above correction coefficient compression method;

[0037] Storing the compressed correction coefficients of each pixel point of the LED display screen into the dynamic random access memory in the receiving card.

[0038] The present invention also provides another storage method for correction coefficients. The method includes:

[0039] Receiving the correction coefficients of each pixel point of the LED display screen;

[0040] Dividing the LED display screen into multiple block areas equally;

[0041] Taking the first pixel point in each block area as a reference pixel point;

[0042] Compressing the correction coefficients of the reference pixel points in each block area according to the above correction coefficient compression method to obtain compressed correction coefficients;

[0043] According to the preset differential coding strategy and the compressed correction coefficients of the reference pixel points in each block area, obtain the differential coding corresponding to the correction coefficients of each pixel point except the reference pixel points in each block area.

[0044] Store the compressed correction coefficients of the reference pixel points of each block area and the differential coding corresponding to the correction coefficients of each pixel point except the reference pixel points in each block area into the dynamic random access memory in the receiving card.

[0045] The present invention also provides another method for storing correction coefficients, and the method includes:

[0046] Receive the correction coefficients corresponding to multiple gray levels of the LED display screen respectively.

[0047] According to the above-mentioned compression method of correction coefficients, compress the correction coefficients corresponding to the first gray level to obtain compressed correction coefficients.

[0048] According to the preset differential coding strategy and the compressed correction coefficients corresponding to the first gray level, obtain the differential coding corresponding to the correction coefficients of each gray level except the first gray level.

[0049] Store the compressed correction coefficients of the first gray level and the differential coding corresponding to the correction coefficients of each gray level except the first gray level into the dynamic random access memory in the receiving card.

[0050] The compression method, device and storage method of correction coefficients provided by the present invention at least include the following beneficial effects: Compress the main coefficients and complementary coefficients according to the target number of bits of the main coefficients and the target number of bits of the complementary coefficients respectively, reduce the data storage bits of the main coefficients and the complementary coefficients, and then when storing the compressed correction coefficients (main coefficients and complementary coefficients) into the dynamic random storage memory such as SDRAM in the receiving card, it can effectively save bandwidth and the storage capacity of the dynamic random storage memory such as SDRAM. Description of the Drawings

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to these drawings.

[0052] Figure 1 It is a flowchart of the steps of the compression method of correction coefficients in an embodiment of the present invention.

[0053] Figure 2 It is the flowchart of the main coefficient compression steps in an embodiment of the present inventionFigure 1 ;

[0054] Figure 3 This is the flowchart of the complementary coefficient compression step in an embodiment of the present invention Figure 1 ;

[0055] Figure 4 This is the flowchart of the complementary coefficient compression step in an embodiment of the present invention Figure 2 ;

[0056] Figure 5 This is the schematic structural diagram of the compression device for the correction coefficient in an embodiment of the present invention;

[0057] Figure 6 This is the first storage method for the correction coefficient in an embodiment of the present invention;

[0058] Figure 7 This is the second storage method for the correction coefficient in an embodiment of the present invention;

[0059] Figure 8 This is the third storage method for the correction coefficient in an embodiment of the present invention;

[0060] 501 - acquisition module, 502 - main coefficient compression module, 503 - complementary coefficient compression module. Detailed implementation manners

[0061] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the protection scope of the present invention.

[0062] The compression method for the correction coefficient provided by the present invention is used to compress the correction coefficient of the LED display screen. The correction coefficient includes the main coefficient and the complementary coefficient. Taking a pixel point in the LED display screen as an example, the correction coefficient corresponding to a pixel point includes 3 main coefficients and 6 complementary coefficients, which can be expressed as The main coefficients are R R , G G , B B , and the complementary coefficients are R G , R B , G R , G B , B R , B G .

[0063] The value range of the main coefficient is [0, 1), and the value range of the complementary coefficient is [0, 0.5]. More specifically, the blue complementary coefficient R in the complementary coefficientB , G B The value range of is [0, 0.25]. The red compensation coefficient and green compensation coefficient G in the compensation coefficients R , B R , B G , R G The value range of is [0, 0.5].

[0064] The main coefficient and the compensation coefficient are represented in the form of fixed-point numbers with P bits of data storage. P is the common data storage bit number of the calibration coefficient in the prior art. Further, the value of P can be 14.

[0065] Since the value range of the main coefficient is [0, 1), and the value range of the compensation coefficient is [0, 0.5]. More specifically, the blue compensation coefficient R B , G B The value range of is [0, 0.25]. The red compensation coefficient and green compensation coefficient G R , B R , B G , R G The value range of is [0, 0.5]. That is, whether it is the main coefficient or the compensation coefficient, they are all pure decimals less than 1 or 0. Therefore, it can be understood that when the main coefficient and the compensation coefficient are represented by P-bit fixed-point numbers, they are represented in the form of P-bit fixed-point decimals, that is, the position of the decimal point is fixed before the highest bit of the data. Taking P = 14 as an example, if the main coefficient is represented by a 14-bit fixed-point number as 10000000000000 (usually represented in binary in the computer), the actual value represented is 0.10000000000000, which is 0.5 in decimal conversion.

[0066] In the LED display screen mentioned in the present invention, the calibration coefficients of each pixel point are generally calculated by the calibration system in the host computer and then sent from the host computer to the receiving card. During use, the receiving card reads the calibration coefficients stored in the dynamic random access memory such as SDRAM in the receiving card to be used for calibrating each pixel point in the LED display screen.

[0067] Further, the execution subject of the calibration coefficient compression method provided by the present invention can be the receiving card in the LED box or the host computer, and the present invention does not limit this.

[0068] In an embodiment of the present invention, as Figure 1 shown, the calibration coefficient compression method includes the following steps:

[0069] Step S101: Obtain the preset target number of bits Q of the main coefficient and the preset target number of bits R of the compensation coefficient.

[0070] Among them, the specific values of the target number of bits Q of the main coefficient and the target number of bits R of the complementary coefficient are set by technicians according to actual needs, and the present invention does not limit this. As long as 0 < Q < P and 0 < R < P are satisfied during the setting.

[0071] It should be understood that when the value of Q or R is set smaller, when dividing the exponent part and the fixed-point decimal part in steps S102 and S103 subsequently, the number of bits of the fixed-point decimal part divided is less. At this time, the accuracy loss after compression of the correction coefficient will be greater. Therefore, when technicians set the value of Q or R, they need to set it comprehensively considering the accuracy loss.

[0072] Step S102: Obtain the main coefficient to be compressed, compress the main coefficient to be compressed from a P-bit fixed-point number to a Q-bit fixed-point number, or compress the main coefficient to be compressed from a P-bit fixed-point number into encoded data with a data storage bit number of Q bits, which consists of a first exponent part of A bits and a first fixed-point decimal part of B bits.

[0073] Specifically, in this step, A + B = Q and A ≤ B.

[0074] Step S103: Obtain the complementary coefficient to be compressed, and compress the complementary coefficient to be compressed from a P-bit fixed-point number into encoded data with a data storage bit number of R bits, which consists of a second exponent part of C bits and a second fixed-point decimal part of D bits.

[0075] Specifically, in this step, C + D = R and C ≤ D.

[0076] Assume that the value of P is 14. Preferably, set the value of Q to 12 and the value of R to 10. Thus, while taking into account the accuracy (with a slight loss of accuracy), the data storage bits of the compression correction coefficient are compressed. The data storage bits of the correction coefficient corresponding to each pixel point (3 main coefficients and 6 supplementary coefficients) are compressed from the original 14 bits * 9 = 126 bits to 12 bits * 3 + 10 bits * 6 = 96 bits. In comparison, the compressed correction coefficient can save approximately 25% of the bandwidth and storage capacity. After processing by the above compression method, the data storage space of the correction coefficient corresponding to one pixel point becomes 96 bit. For a 64-bit SDRAM dynamic random access memory, 64 * 3 = 96 * 2. Therefore, storing one pixel point only occupies 1.5 positions in the SDRAM. For a 32-bit SDRAM, 3 * 32 = 96 bit. Therefore, storing one pixel point only occupies 3 positions in the SDRAM. For a display screen with a resolution of 512 * 512, 0.5 G of bandwidth can be saved. For a display screen with a resolution of 1024 * 512, 1 G of bandwidth can be saved. For example, 512 * 512 * 126 / 1024 / 1024 / 1024 / (1 / 60) = 1.84 G, 512 * 512 * 96 / 1024 / 1024 / 1024 / (1 / 60) = 1.4 G, with a difference of approximately 0.5 Gbps.

[0077] More specifically, when the value of P is 14 and the value of Q is set to 12, the main coefficient to be compressed, which is a 14-bit fixed-point number, can be compressed into an encoded data with a data storage bit of 12 bits, consisting of a 3-bit first exponent part and a 9-bit first fixed-point decimal part.

[0078] Or the main coefficient to be compressed, which is a 14-bit fixed-point number, can be compressed into an encoded data with a data storage bit of 12 bits, consisting of a 2-bit first exponent part and a 10-bit first fixed-point decimal part.

[0079] When the value of P is 14 and the value of R is set to 10, the supplementary coefficient to be compressed, which is a 14-bit fixed-point number, can be compressed into an encoded data with a data storage bit of 10 bits, consisting of a 2-bit second exponent part and an 8-bit second fixed-point decimal part.

[0080] Or the supplementary coefficient to be compressed, which is a 14-bit fixed-point number, can be compressed into an encoded data with a data storage bit of 10 bits, consisting of a 3-bit second exponent part and a 7-bit second fixed-point decimal part.

[0081] Preferably, the number of bits of the compressed data storage of the correction coefficient corresponding to each pixel point is an integer multiple of 32. That is, when setting the Q value and the R value, Q and R satisfy 3*Q + 6*R = 32n, where n is a positive integer. When compressing the correction coefficient, ensure that the number of bits of the compressed data storage of the correction coefficient corresponding to each pixel point is an integer multiple of 32, which is convenient for storage.

[0082] Further, in the first implementation manner of this embodiment, compressing the main coefficient to be compressed from a fixed-point number with P bits to a fixed-point number with Q bits includes:

[0083] Deleting the last P - Q bits of the main coefficient to be compressed with P bits according to the preset target number of bits Q of the main coefficient, so as to compress the main coefficient to be compressed from a fixed-point number with P bits to a fixed-point number with Q bits.

[0084] Specifically, for the sake of easy understanding, the following is an example:

[0085] If P = 14 and the target number of bits Q of the main coefficient is 12, and the main coefficient to be compressed with 14 bits is 10000000000001, when compressing the main coefficient with 14 bits, the last 2 bits (01) of the 14-bit main coefficient are deleted, that is, compressed into a 12-bit fixed-point number 100000000000.

[0086] Further, in the second implementation manner of this embodiment, as Figure 2 shown, compressing the main coefficient from a fixed-point number with P bits into encoded data with a data storage number of bits Q composed of a first exponential part with A bits and a first fixed-point decimal part with B bits includes:

[0087] Step S201: Obtain a preset main coefficient exponential encoding - exponential value mapping table.

[0088] Among them, the preset main coefficient exponential encoding - exponential value mapping table includes multiple preset main coefficient exponential encodings, which are used to represent multiple preset main coefficient exponential values, and one main coefficient exponential value corresponds to one main coefficient exponential encoding.

[0089] The main coefficient exponential encoding - exponential value mapping table is preset by those skilled in the art according to needs.

[0090] Specifically, the main coefficient exponential encoding - exponential value mapping table can represent different main coefficient exponential values in the form of exponential encoding with a fixed number of bits, or can represent different main coefficient exponential values in the form of exponential encoding with a dynamic number of bits. Those skilled in the art can select the corresponding exponential encoding form according to actual needs, and the present invention does not limit this.

[0091] Among them, representing different main coefficient exponent values in the form of fixed - digit exponential coding means that for different main coefficient exponent values, the same fixed - digit main coefficient exponent coding is used for representation. For example, for the main coefficient exponent value - 1, it is represented by 010 (main coefficient exponent coding) with 3 bits, and for the main coefficient exponent value - 2, it is also represented by 100 with 3 bits.

[0092] As shown in the main coefficient exponent coding - exponent value mapping tables in Table 1 and Table 2 below, different main coefficient exponent values are represented in the form of fixed - digit exponential coding.

[0093] Table 1:

[0094]

[0095] Table 2:

[0096]

[0097] It should be understood that the above two mapping tables are only examples and not limitations of the present invention. Those skilled in the art can set different main coefficient exponent values according to actual needs and correspond different main coefficient exponent value codings.

[0098] When setting the main coefficient exponent coding - exponent value mapping table, as shown in Table 1 and Table 2, both the main coefficient exponent values in the mapping table and the binary main coefficient exponent codings corresponding to these main coefficient exponent values are set by those skilled in the art according to the actual situation, as long as it is ensured that each main coefficient exponent value corresponds to a unique main coefficient exponent coding.

[0099] Furthermore, the maximum value of the main coefficient exponent value can be determined in advance, and using the offset - center coding method, the main coefficient exponent values less than or equal to the maximum value are encoded in sequence (both Table 1 and Table 2 are in the offset - center coding method and only encode the main coefficient exponent values less than the maximum value of 1). Specifically, since the main coefficients are all less than 1, the maximum value of the main coefficient exponent value can be set to 1.

[0100] Even further, when representing different main coefficient exponent values in the form of fixed - digit coding, there is a relationship between the maximum number of main coefficient exponent values that the mapping table can represent and the number of bits of the main coefficient exponent coding. For example, when the number of bits of the main coefficient exponent coding is 2 bits (as shown in Table 2), then this mapping table can represent at most 2 2 main coefficient exponent values (because there are only 4 different main coefficient exponent codings for 2 - bit binary numbers). Therefore, when those skilled in the art preset the mapping table, they need to determine the number of bits of the main coefficient exponent coding according to the number of main coefficient exponent values that the mapping table needs to represent.

[0101] More specifically, the number of main coefficient exponent values, the maximum value of the main coefficient exponent values, and the actual values of each main coefficient exponent value are all set by the technical personnel according to actual needs.

[0102] Using an exponential coding form with dynamic bit numbers to represent different main coefficient exponent values means that for different main coefficient exponent values, a main coefficient exponent coding with a non-fixed number of bits is used for representation. For example, for the main coefficient exponent value -2, it is represented by 1 bit of 1 (main coefficient exponent coding), and for the main coefficient exponent value -3, it is represented by 2 bits of 01 (main coefficient exponent coding).

[0103] As shown in the following Table 3, Table 4, and Table 5, which are the main coefficient exponent coding - exponent value mapping tables, they are using the exponential coding form with dynamic bit numbers to represent different main coefficient exponent values.

[0104] Table 3:

[0105]

[0106] Table 4:

[0107]

[0108] Table 5:

[0109]

[0110]

[0111] It should be understood that the above mapping table of the dynamic bit number coding form is only an example and not a limitation to the present invention. The technical personnel can set different main coefficient exponent values according to actual needs and correspond different main coefficient exponent value codings.

[0112] When setting the main coefficient exponent coding - exponent value mapping table, as shown in Table 3, Table 4, and Table 5, both the main coefficient exponent values in the mapping table and the binary main coefficient exponent codings corresponding to the main coefficient exponent values are set by the technical personnel according to the actual situation, as long as it is ensured that each main coefficient exponent value only corresponds to a uniquely determined main coefficient exponent coding.

[0113] Furthermore, the maximum value of the main coefficient exponent values can also be determined in advance, and an offset - center coding method is used to sequentially code the main coefficient exponent values less than or equal to the maximum value (for example, Table 3, Table 4, and Table 5 are all in the offset - center coding method, and only code the main coefficient exponent values less than the maximum value by 1). Specifically, since the main coefficients are all less than 1, the maximum value of the main coefficient exponent values can be set to 1.

[0114] Furthermore, when using a dynamic-digit coding form to represent different main coefficient exponent values, it is also necessary for the technician to first determine the number of main coefficient exponent values to be represented by the mapping table. When setting, taking Table 3 and Table 4 as examples, in Table 3, 4 main coefficient exponent values are represented by main coefficient exponent encodings 1, 01, 001, and 000. When the number of main coefficient exponent values to be represented is 6, 0 or 1 can be added after the main coefficient exponent encoding 1, and 0 or 1 can be added after the main coefficient exponent encoding 01, so as to add 2 main coefficient exponent encodings to represent 6 main coefficient exponent values. It should be understood that the above Table 3, Table 4, and Table 5 are only examples and not limitations to the present invention.

[0115] Specifically, assuming that the number of main coefficient exponent values to be represented is 4, and the technician pre-selects four main coefficient exponent values of -2, -1, 0, and 1, then the number of main coefficient exponent encodings required is 4, and the obtained main coefficient exponent encoding-exponent value mapping table is as shown in Table 3 above.

[0116] In dynamic coding, when setting the mapping relationship, multiple sets of corresponding mapping values can be set for each main coefficient exponent encoding (such as the first set of mapping values 1 and the second set of mapping values 2 in Table 3). In actual use, one set can be selected according to actual needs for compressing the correction coefficient (it should be understood that in actual use, the same set of mapping values used during compression needs to be used to decompress the compressed correction coefficient to ensure that the obtained correction coefficient value after decompression will not be disordered).

[0117] Step S202: According to the digit S where the first significant digit of the main coefficient to be compressed is located, determine that the main coefficient target exponent value of the main coefficient to be compressed is 1 - S.

[0118] Specifically, the first significant digit refers to: the first non-0 digit from the left of a number.

[0119] Step S203: Obtain the main coefficient exponent encoding corresponding to the main coefficient target exponent value from the main coefficient exponent encoding-exponent value mapping table as the first exponent part.

[0120] Step S204: According to the data storage digit A of the first exponent part and the main coefficient target digit Q, determine the data storage digit B of the first fixed-point decimal part, where B = Q - A.

[0121] Step S205: According to the data storage digit B of the first fixed-point decimal part and the main coefficient target exponent value 1 - S, determine the main coefficient target decimal value of the main coefficient to be compressed, and use the main coefficient target decimal value as the first fixed-point decimal part.

[0122] Specifically, in this step, it is necessary to use the formula: the main coefficient to be compressed = 2 主系数目标指数值*f, substitute the main coefficient to be compressed and the main coefficient target exponent value 1 - S into the above formula to calculate f.

[0123] Use H to represent the number of bits for storing the data of f, and determine whether H is equal to B. If so, directly use f as the main coefficient target decimal value of the main coefficient to be compressed;

[0124] Otherwise, determine whether H is greater than B. If so, delete the last H - B bits of f, and use f after deleting the last H - B bits as the main coefficient target decimal value of the main coefficient to be compressed. Otherwise, fill 0 after f to convert the number of bits for storing the data of f from H bits to B bits, and use f after filling 0 as the main coefficient target decimal value of the main coefficient to be compressed.

[0125] To facilitate the understanding of the above steps S201 - S205, the following uses a specific example for illustration:

[0126] Take "the main coefficient to be compressed is 00110001000001, the preset main coefficient target number of bits Q = 12, and the preset main coefficient exponent coding - exponent value mapping table is Table 1" as an example.

[0127] Then the first significant digit of the main coefficient to be compressed is the first non - 0 digit 1 (located at the 3rd position) from the left, that is, the position where the first significant digit is located is 3, S = 3. Therefore, determine that the main coefficient target exponent value K of the main coefficient to be compressed is 1 - 3 = - 2.

[0128] When the main coefficient target exponent value is - 2, the corresponding main coefficient exponent coding in Table 1 is 100, that is, after compressing the main coefficient to be compressed, the first exponent part is 011.

[0129] If the first exponent part is 011, then the number of bits A for storing the data of the first exponent part is 3 bits, and the number of bits B for storing the data of the first fixed - point decimal part is B = Q - A = 12 - 3 = 9 bits;

[0130] Since when the main coefficient target exponent value K is - 2, according to the formula: main coefficient to be compressed = 2 k *f, calculate f = 110001000001. Since f is 12 bits at this time, which is greater than 9 bits, therefore delete the last 12 - 9 = 3 bits of f = 110001000001, then obtain the main coefficient target decimal value 110001000, and use the main coefficient target decimal value 110001000 as the first fixed - point decimal part.

[0131] After the above steps, the main coefficient to be compressed of 14 - bit 00110001000001 is compressed into 12 - bit 011110001000, where the first 3 - bit 011 is the first exponent part, and the last 9 - bit 110001000 is the first fixed - point decimal part.

[0132] In the above example, if the preset main coefficient exponent encoding - exponent value mapping table is Table 2, when the target exponent value of the main coefficient is -2, the main coefficient exponent encoding corresponding to Table 2 is 11. That is, after compressing the main coefficient to be compressed, the first exponent part is 11.

[0133] If the first exponent part is 11, then the number of data storage bits A of the first exponent part is 2 bits, and the number of data storage bits B of the first fixed - point decimal part is B = Q - A = 12 - 2 = 10 bits;

[0134] Since when the target exponent value K of the main coefficient is -2, according to the formula: the main coefficient to be compressed = 2 k *f, we get f = 110001000001. Since f at this time is 12 bits, which is greater than 10 bits, the last 12 - 10 = 2 bits of f = 110001000001 are deleted, then the target small value of the main coefficient 1100010000 is obtained, and the target small value of the main coefficient 1100010000 is used as the first fixed - point decimal part.

[0135] After the above steps, the main coefficient to be compressed, which is 14 - bit 00110001000001, is compressed into 12 - bit 111100010000, where the first 2 - bit 11 is the first exponent part, and the last 10 - bit 1100010000 is the first fixed - point decimal part.

[0136] It should be understood that when subsequent correction coefficients are needed to correct the pixel points of the LED display screen, the compressed main coefficient needs to be decoded according to the main coefficient exponent encoding - exponent value mapping table to obtain the specific main coefficient value for subsequent correction.

[0137] In the third implementation manner of this embodiment, compressing the supplementary coefficient to be compressed, which is a P - bit fixed - point number, into encoded data with a data storage bit number of R bits composed of a C - bit second exponent part and a D - bit second fixed - point decimal part includes the following steps:

[0138] Step S301: Obtain the preset first supplementary coefficient exponent encoding - exponent value mapping table.

[0139] The first supplementary coefficient exponent encoding - exponent value mapping table includes multiple preset supplementary coefficient exponent encodings for representing multiple preset supplementary coefficient exponent values. Among them, one supplementary coefficient exponent value corresponds to one supplementary coefficient exponent encoding;

[0140] Similar to the main coefficient exponent encoding-exponent value mapping table, the complementary coefficient exponent encoding-exponent value mapping table is also preset by technicians according to needs. Similarly, the complementary coefficient exponent encoding-exponent value mapping table can also use a fixed-bit exponent encoding form to represent different complementary coefficient exponent values, or use a dynamic-bit exponent encoding form to represent different complementary coefficient exponent values. Technicians can choose the corresponding exponent encoding form according to actual needs, and the present invention does not limit this. Since the fixed-bit and dynamic-bit exponent encoding forms have been specifically described above, they will not be elaborated here.

[0141] As shown in Table 6 and Table 7 below, the complementary coefficient exponent encoding-exponent value mapping table uses a fixed-bit exponent encoding form to represent different complementary coefficient exponent values (since the value range of the complementary coefficient is [0, 0.5], the maximum value of the complementary coefficient exponent value can be set to -2. Similarly, using the method of offset center, each complementary coefficient exponent value less than or equal to the maximum value is encoded). It should be understood that, like the main coefficient exponent value, the number, actual value, and maximum value of the complementary coefficient exponent values in the complementary coefficient exponent encoding-exponent value mapping table are set by technicians according to actual needs.

[0142] Table 6:

[0143]

[0144] Table 7:

[0145]

[0146] It should be understood that the above two mapping tables are only examples and not limitations to the present invention.

[0147] Step S302: Determine that the complementary coefficient target exponent value of the complementary coefficient to be compressed is 1 - T according to the digit position T where the first significant digit of the complementary coefficient to be compressed is located;

[0148] For the specific meaning of the first significant digit, refer to Step S202 and will not be elaborated here.

[0149] Step S303: Obtain the complementary coefficient exponent encoding corresponding to the complementary coefficient target exponent value from the first complementary coefficient exponent encoding-exponent value mapping table as the second exponent part;

[0150] Step S304: Determine the data storage bit number D of the first fixed-point decimal part according to the data storage bit number C of the second exponent part and the complementary coefficient target bit number R, where D = R - C.

[0151] Step S305: Determine the target fractional value of the supplementary coefficient to be compressed according to the number of data storage bits D of the fractional part of the second fixed point and the main coefficient target exponent value 1 - T, and use the target fractional value of the supplementary coefficient as the fractional part of the second fixed point.

[0152] Since the specific implementation methods of steps S301 - S305 are the same as those of steps S201 - S205, except that there are differences in the specific values set for R and Q, and there are differences between the preset first supplementary coefficient exponent encoding - exponent value mapping table and the main coefficient exponent encoding - exponent value mapping table, the steps S301 - S305 will not be introduced in detail. The specific implementation method can be directly referred to the description in steps S201 - S205, and will not be elaborated here.

[0153] The present invention compresses the correction coefficient through the above implementation method. Under the condition of losing weak precision, the data storage bits of the correction coefficient can be reduced. Then, the dynamic random access memory in the receiving card stores the compressed correction coefficient, which can save bandwidth and the storage capacity of the SDRAM, and improve the loading capacity of random access memories such as the SDRAM.

[0154] In another embodiment of the present invention, the supplementary coefficient includes 2 blue supplementary coefficients, 2 red supplementary coefficients, and 2 green supplementary coefficients.

[0155] Obtaining the preset target number of bits R of the supplementary coefficient in step S101 includes obtaining the preset target number of bits R1 of the blue supplementary coefficient and obtaining the preset target number of bits R2 of the red and green supplementary coefficients, where 0 < R2 < R1 < P.

[0156] As Figure 4 shown, the step of compressing the supplementary coefficient from a P - bit fixed - point number to an encoded data with a data storage bit number of R bits composed of a C - bit second exponent part and a D - bit second fractional part in step S103 includes the following steps:

[0157] Step S401: Obtain the preset second supplementary coefficient exponent encoding - exponent value mapping table.

[0158] Specifically, the second supplementary coefficient exponent encoding - exponent value mapping table includes multiple preset supplementary coefficient exponent encodings for representing multiple preset supplementary coefficient exponent values, where one supplementary coefficient exponent value corresponds to one supplementary coefficient exponent encoding.

[0159] Similarly, the second supplementary coefficient exponent encoding - exponent value mapping table is also preset by those skilled in the art according to actual needs. Specifically, it can refer to the first supplementary coefficient exponent encoding - exponent value mapping table, and the present invention does not limit this.

[0160] Step S402: Obtain the coefficient to be compressed and complemented, and determine whether the coefficient to be compressed and complemented is a blue complement coefficient. If so, execute Step S403; otherwise, execute Step S404.

[0161] Step S403: According to the second complement coefficient exponent encoding-exponent value mapping table and the target number of bits R1 of the blue complement coefficient, compress the coefficient to be compressed and complemented from a P-bit fixed-point number into an encoded data with a storage bit number of R1 bits, which consists of a second exponent part of C1 bits and a second fixed-point decimal part of D1 bits.

[0162] Where C1 + D1 = R1, and C1 ≤ D1;

[0163] Specifically, since the specific implementation method of Step S403 is the same as that of Steps S201 - S205, except for the difference in the specific values set for R1 and Q, and the difference between the preset second complement coefficient exponent encoding-exponent value mapping table and the main coefficient exponent encoding-exponent value mapping table, Step S403 will not be introduced in detail. The specific implementation method can be directly referred to the description in Steps S201 - S205 and will not be elaborated here.

[0164] Step S404: According to the second complement coefficient exponent encoding-exponent value mapping table and the target number of bits R2 of the red complement coefficient and the green complement coefficient, compress the coefficient to be compressed and complemented from a P-bit fixed-point number into an encoded data with a storage bit number of R2 bits, which consists of a second exponent part of C2 bits and a second fixed-point decimal part of D2 bits.

[0165] Where C2 + D2 = R2, and C2 ≤ D2, D1 > D2.

[0166] Specifically, since the specific implementation method of Step S404 is the same as that of Steps S201 - S205, except for the difference in the specific values set for R2 and Q, and the difference between the preset second complement coefficient exponent encoding-exponent value mapping table and the main coefficient exponent encoding-exponent value mapping table, Step S404 will not be introduced in detail. The specific implementation method can be directly referred to the description in Steps S201 - S205 and will not be elaborated here.

[0167] Specifically, in this embodiment, taking R1 = 11 and R2 = 10 as examples, and taking the second complementary coefficient exponential coding - exponential value mapping table as an example of representing different complementary coefficient exponential values in a fixed - bit coding form, assuming that the number of bits of the complementary coefficient exponential coding is 3 bits, then the second exponential part of the blue complementary coefficient after compression is 3 bits, and the second fixed - point decimal part is 8 bits; the second exponential part of the red complementary coefficient is 3 bits, and the second fixed - point decimal part is 7 bits. Further, based on the above example, Q can also be set to 11. When compressing the main coefficient, both the main coefficient and the blue complementary coefficient are compressed into 11 bits, and the red and green complementary coefficients are compressed into 10 bits. Of course, Q can also be set to other values, and the present invention does not limit this.

[0168] Using the method in steps S401 - S404 for compression can improve the accuracy of the compressed blue complementary coefficient (compared with other complementary coefficients, the fixed - point decimal part has a higher number of bits, so the accuracy is greater). Since there are more problems with blue light points in actual use, a higher - accuracy blue complementary coefficient is required to correct them.

[0169] In an embodiment of the present invention, as Figure 5 shown, the present invention also provides a compression device for correction coefficients, which is used to compress correction coefficients. Among them, the correction coefficients include a main coefficient and complementary coefficients, and the main coefficient and complementary coefficients are represented in the form of fixed - point numbers with a data storage bit number of P bits. The device includes an acquisition module 501, a main - coefficient compression module 502, and a complementary - coefficient compression module 503, where:

[0170] The acquisition module 501 is connected to the main - coefficient compression module 502 and the complementary - coefficient compression module 503, and is used to acquire a preset main - coefficient target bit number Q and a preset complementary - coefficient target bit number R, where 0 < Q < P and 0 < R < P;

[0171] The main - coefficient compression module 502 is used to acquire the main coefficient to be compressed, and compress the main coefficient to be compressed from a fixed - point number of P bits into a fixed - point number of Q bits, or compress the main coefficient to be compressed from a fixed - point number of P bits into coded data with a data storage bit number of Q bits, which consists of a first exponential part of A bits and a first fixed - point decimal part of B bits, where A + B = Q and A ≤ B;

[0172] The complementary - coefficient compression module 503 is used to acquire the complementary coefficient to be compressed, and compress the complementary coefficient to be compressed from a fixed - point number of P bits into coded data with a data storage bit number of R bits, which consists of a second exponential part of C bits and a second fixed - point decimal part of D bits, where C + D = R and C ≤ D.

[0173] In the calibration coefficient compression device provided in this embodiment, the number of data storage bits of the calibration coefficient can be reduced under the condition of weak precision loss. Then, the dynamic random access memory in the receiving card stores the compressed calibration coefficient, which can save bandwidth and the storage capacity of the SDRAM, and improve the loading capacity of the random access memory such as the SDRAM.

[0174] In an embodiment of the present invention, a first method for storing calibration coefficients is provided. The execution subject of this method is the receiving card or the host computer of the LED cabinet, as Figure 6 shown. The method includes the following steps:

[0175] Step S601: Receive the calibration coefficients of each pixel of the LED display screen;

[0176] Step S602: Compress the calibration coefficients of each pixel of the LED display screen according to the above calibration coefficient compression method;

[0177] Step S603: Store the compressed calibration coefficients of each pixel of the LED display screen in the dynamic random access memory in the receiving card.

[0178] In the calibration coefficient storage method provided in this embodiment, after obtaining the calibration coefficients of each pixel of the LED display screen, first use the calibration coefficient compression method provided by the present invention to compress the calibration coefficients of each pixel, reduce the data storage space required for the calibration coefficients of each pixel, and store the compressed calibration coefficients in the dynamic random access memory in the receiving card, which can save bandwidth and the storage capacity of the SDRAM, and improve the loading capacity of the dynamic random access memory in the receiving card.

[0179] In another embodiment of the present invention, a second method for storing calibration coefficients is provided. Similarly, the execution subject of this method is also the receiving card or the host computer of the LED cabinet, as Figure 7 shown. The method includes:

[0180] Step S701: Receive the calibration coefficients of each pixel of the LED display screen;

[0181] Step S702: Divide the LED display screen into multiple block areas equally;

[0182] Specifically, the LED display screen can be divided into multiple block areas equally, and each block area includes M*N pixels. The values of M and N are preset by technicians according to actual needs, and the present invention does not limit this.

[0183] Step S703: Use the first pixel in each block area as the reference pixel.

[0184] Specifically, the pixel at the first row and the first column of each block area is regarded as the first pixel.

[0185] Step S704: Compress the correction coefficients of the reference pixels in each block area according to the above compression method of the correction coefficients to obtain the compressed correction coefficients.

[0186] Step S705: Obtain the differential encoding corresponding to the correction coefficients of each pixel other than the reference pixel in each block area according to the preset differential encoding strategy and the compressed correction coefficients of the reference pixels in each block area.

[0187] Specifically, taking one of the block areas as an example, calculate the difference between the correction coefficients of each pixel other than the reference pixel in this block area and the compressed correction coefficients of the reference pixel, and compress the difference according to the above compression method of the correction coefficients. The compressed difference is the differential encoding corresponding to the correction coefficients of each pixel in this block area.

[0188] Step S706: Store the compressed correction coefficients of the reference pixels in each block area and the differential encoding corresponding to the correction coefficients of each pixel other than the reference pixel in each block area into the dynamic random access memory in the receiving card.

[0189] In the storage method of the correction coefficients provided in this embodiment, after receiving the correction coefficients, the LED display screen is equally divided into multiple block areas, the correction coefficients of the first pixels in each block area are compressed and saved, and the remaining pixels save the differential encoding of the correction coefficients. During actual use, when obtaining the correction coefficients from the receiving card, only the correction coefficients corresponding to the first pixels in each block area need to be obtained, and the correction coefficients corresponding to the remaining pixels in this block area can be obtained through the differential encoding.

[0190] In another embodiment of the present invention, a third storage method of the correction coefficients is provided, as Figure 8 shown, the method includes:

[0191] Step S801: Receive the correction coefficients corresponding to multiple gray levels of the LED display screen.

[0192] Specifically, in the hierarchical correction method, the gray levels (generally 0 - 255) of the LED display screen can be pre-divided into multiple gray levels, and the corresponding correction coefficients (also including 3 main coefficients and 6 supplementary coefficients) are calculated for each gray level. When later correcting the pixels of the LED display screen, according to which gray level the gray value of each pixel belongs to, determine the correction coefficient corresponding to this pixel.

[0193] Step S802: Compress the correction coefficient corresponding to the first gray stage according to the above-mentioned compression method of the correction coefficient to obtain the compressed correction coefficient;

[0194] Specifically, sort in ascending or descending order according to the gray level values corresponding to the gray stages, and take the first gray stage as the first gray stage.

[0195] Step S803: Obtain the differential coding corresponding to the correction coefficients of each gray stage except the first gray stage according to the preset differential coding strategy and the compressed correction coefficient corresponding to the first gray stage.

[0196] Specifically, calculate the differences between the correction coefficients of each gray stage except the first gray stage in multiple gray stages and the compressed correction coefficient of the first gray stage, and compress the differences according to the above-mentioned compression method of the correction coefficient. The compressed differences are the differential coding corresponding to the correction coefficients of each gray stage except the first gray stage.

[0197] Furthermore, in this step, when compressing the differences according to the above-mentioned compression method of the correction coefficient, the differences can be compressed into a form with a smaller target number of bits. For example, for a 14-bit correction coefficient, set the target number of bits of the main coefficient / blue complementary coefficient / red-green complementary coefficient of the first gray stage to 12 bits / 10 bits / 10 bits respectively, and set the target number of bits of the differences between the main coefficients / blue complementary coefficients / red-green complementary coefficients of the remaining gray stages and the first gray stage to 8 bits / 6 bits / 6 bits respectively.

[0198] Step S804: Store the compressed correction coefficient of the first gray stage and the differential coding corresponding to the correction coefficients of each gray stage except the first gray stage in the dynamic random access memory in the receiving card.

[0199] In the storage method of the correction coefficient provided in this embodiment, after receiving the correction coefficients corresponding to multiple gray stages of the LED display screen, compress and store the correction coefficient corresponding to the first gray stage, and the remaining gray stages store the differential coding corresponding to the correction coefficients. In the actual use process, when obtaining the correction coefficient from the receiving card, only need to obtain the correction coefficient corresponding to the first gray stage, and then the correction coefficients corresponding to the remaining gray stages can be obtained through the differential coding.

[0200] The terms and expressions used in the specification of the present invention are for illustrative purposes only and do not constitute a limitation. Those skilled in the art should understand that various changes can be made to the details of the above embodiments without departing from the basic principles of the disclosed embodiments. Therefore, the scope of the present invention is only determined by the claims, and in the claims, unless otherwise specified, all terms should be understood in the broadest reasonable sense.

Claims

1. A method for compressing correction coefficients, characterized in that, For compressing the correction coefficients of an LED display screen, wherein the correction coefficients include a main coefficient and a supplementary coefficient, and the main coefficient and the supplementary coefficient are represented in the form of fixed-point numbers with a data storage bit number of P bits. The method includes: Obtaining a preset target bit number Q for the main coefficient and a preset target bit number R for the supplementary coefficient, where 0 < Q < P and 0 < R < P; Obtaining the main coefficient to be compressed, and compressing the main coefficient to be compressed from a fixed-point number of P bits to a fixed-point number of Q bits, or compressing the main coefficient to be compressed from a fixed-point number of P bits to encoded data with a data storage bit number of Q bits, which is composed of a first exponential part of A bits and a first fixed-point decimal part of B bits, where A + B = Q and A ≤ B; Obtaining the supplementary coefficient to be compressed, and compressing the supplementary coefficient to be compressed from a fixed-point number of P bits to encoded data with a data storage bit number of R bits, which is composed of a second exponential part of C bits and a second fixed-point decimal part of D bits, where C + D = R and C ≤ D.

2. The method for compressing correction coefficients according to claim 1, characterized in that, The compressing the main coefficient to be compressed from a fixed-point number of P bits to a fixed-point number of Q bits includes: Deleting the last P - Q bits of the main coefficient to be compressed of P bits according to the preset target bit number Q for the main coefficient, so as to compress the main coefficient to be compressed from a fixed-point number of P bits to a fixed-point number of Q bits.

3. The method for compressing correction coefficients according to claim 1, characterized in that, The compressing the main coefficient to be compressed from a fixed-point number of P bits to encoded data with a data storage bit number of Q bits, which is composed of a first exponential part of A bits and a first fixed-point decimal part of B bits, includes: Obtaining a preset main coefficient exponential encoding - exponential value mapping table, where the preset main coefficient exponential encoding - exponential value mapping table includes a plurality of preset main coefficient exponential encodings for representing a plurality of preset main coefficient exponential values, and one main coefficient exponential value corresponds to one main coefficient exponential encoding; determining that the target exponential value of the main coefficient of the main coefficient to be compressed is 1 - S according to the position S of the first significant digit of the main coefficient to be compressed; Obtaining, from the main coefficient exponential encoding - exponential value mapping table, the main coefficient exponential encoding corresponding to the target exponential value of the main coefficient as the first exponential part; Determining the data storage bit number B of the first fixed-point decimal part according to the data storage bit number A of the first exponential part and the target bit number Q of the main coefficient, where B = Q - A; Determining the target decimal value of the main coefficient of the main coefficient to be compressed according to the data storage bit number B of the first fixed-point decimal part and the target exponential value 1 - S of the main coefficient, and using the target decimal value of the main coefficient as the first fixed-point decimal part.

4. The method for compressing correction coefficients according to claim 3, characterized in that, The preset main coefficient exponential encoding - exponential value mapping table represents different main coefficient exponential values in the form of exponential encodings with a fixed number of bits or in the form of exponential encodings with a dynamic number of bits.

5. The method for compressing correction coefficients according to claim 1, characterized in that, The compressing the supplementary coefficient to be compressed from a fixed-point number of P bits to encoded data with a data storage bit number of R bits, which is composed of a second exponential part of C bits and a second fixed-point decimal part of D bits, includes: Obtain a preset first complementary coefficient exponent encoding - exponent value mapping table, where the first complementary coefficient exponent encoding - exponent value mapping table includes a plurality of preset complementary coefficient exponent encodings for representing a plurality of preset complementary coefficient exponent values. Among them, one complementary coefficient exponent value corresponds to one complementary coefficient exponent encoding; According to the digit position T where the first significant digit of the to - be - compressed complementary coefficient is located, determine that the target exponent value of the complementary coefficient of the to - be - compressed complementary coefficient is 1 - T; Obtain, from the first complementary coefficient exponent encoding - exponent value mapping table, the complementary coefficient exponent encoding corresponding to the target exponent value of the complementary coefficient as the second exponent part; According to the data storage bit number C of the second exponent part and the target bit number R of the complementary coefficient, determine the data storage bit number D of the first fixed - point decimal part, where D = R - C; According to the data storage bit number D of the second fixed - point decimal part and the target exponent value 1 - T of the main coefficient, determine the target decimal value of the complementary coefficient of the to - be - compressed complementary coefficient, and use the target decimal value of the complementary coefficient as the second fixed - point decimal part.

6. The method for compressing correction coefficients according to claim 1, characterized in that, The complementary coefficient includes a blue complementary coefficient, a red complementary coefficient, and a green complementary coefficient. The obtaining of the preset target bit number R of the complementary coefficient includes: obtaining the preset target bit number R1 of the blue complementary coefficient and obtaining the preset target bit numbers R2 of the red complementary coefficient and the green complementary coefficient, where 0 < R2 < R1 < P; The obtaining of the to - be - compressed complementary coefficient and the compression of the to - be - compressed complementary coefficient from a P - bit fixed - point number into an encoded data with a data storage bit number of R bits composed of a C - bit second exponent part and a D - bit second fixed - point decimal part includes: Obtain a preset second complementary coefficient exponent encoding - exponent value mapping table, where the second complementary coefficient exponent encoding - exponent value mapping table represents different complementary coefficient exponent values in a fixed - bit encoding form; Obtain the to - be - compressed complementary coefficient. If the to - be - compressed complementary coefficient is a blue complementary coefficient, according to the second complementary coefficient exponent encoding - exponent value mapping table and the target bit number R1 of the blue complementary coefficient, compress the to - be - compressed complementary coefficient from a P - bit fixed - point number into an encoded data with a data storage bit number of R1 bits composed of a C1 - bit second exponent part and a D1 - bit second fixed - point decimal part, where C1 + D1 = R1 and C1 ≤ D1; If the to - be - compressed complementary coefficient is a red complementary coefficient or a green complementary coefficient, according to the second complementary coefficient exponent encoding - exponent value mapping table and the target bit numbers R2 of the red complementary coefficient and the green complementary coefficient, compress the to - be - compressed complementary coefficient from a P - bit fixed - point number into an encoded data with a data storage bit number of R2 bits composed of a C2 - bit second exponent part and a D2 - bit second fixed - point decimal part, where C2 + D2 = R2 and C2 ≤ D2, and D1 > D2.

7. A device for compressing correction coefficients, characterized in that, For compressing a correction coefficient, where the correction coefficient includes a main coefficient and a complementary coefficient, the main coefficient and the complementary coefficient are represented in the form of a P - bit fixed - point number. The device includes an acquisition module, a main coefficient compression module, and a complementary coefficient compression module, where: The obtaining module, connected to the main coefficient compression module and the complementary coefficient compression module, is configured to obtain a preset target number of bits Q for the main coefficient and a preset target number of bits R for the complementary coefficient, where 0 < Q < P and 0 < R < P; The main coefficient compression module is configured to obtain the main coefficient to be compressed, compress the main coefficient to be compressed from a fixed-point number with P bits into a fixed-point number with Q bits, or compress the main coefficient to be compressed from a fixed-point number with P bits into encoded data with a data storage bit number of Q bits, which is composed of a first exponential part with A bits and a first fixed-point decimal part with B bits, where A + B = Q and A ≤ B; The complementary coefficient compression module is configured to obtain the complementary coefficient to be compressed, compress the complementary coefficient to be compressed from a fixed-point number with P bits into encoded data with a data storage bit number of R bits, which is composed of a second exponential part with C bits and a second fixed-point decimal part with D bits, where C + D = R and C ≤ D.

8. A method for storing correction coefficients, characterized in that, The method includes: Receiving the correction coefficients of each pixel point of the LED display screen; Compressing the correction coefficients of each pixel point of the LED display screen according to the correction coefficient compression method according to any one of claims 1-6; Storing the compressed correction coefficients of each pixel point of the LED display screen into the dynamic random access memory in the receiving card.

9. A storage method for correction coefficients, characterized in that, The method includes: Receiving the correction coefficients of each pixel point of the LED display screen; Dividing the LED display screen into multiple block areas equally; Taking the first pixel point in each block area as the reference pixel point; Compressing the correction coefficients of the reference pixel points in each block area according to the correction coefficient compression method according to any one of claims 1-6 to obtain compressed correction coefficients; Obtaining the differential encoding corresponding to the correction coefficients of each pixel point except the reference pixel point in each block area according to a preset differential encoding strategy and the compressed correction coefficients of the reference pixel points in each block area; Storing the compressed correction coefficients of the reference pixel points in each block area and the differential encoding corresponding to the correction coefficients of each pixel point except the reference pixel point in each block area into the dynamic random access memory in the receiving card.

10. A storage method for correction coefficients, characterized in that, The method includes: Receiving the correction coefficients corresponding to multiple gray levels of the LED display screen; Compressing the correction coefficients corresponding to the first gray level according to the correction coefficient compression method according to any one of claims 1-6 to obtain compressed correction coefficients; Obtaining the differential encoding corresponding to the correction coefficients of each gray level except the first gray level according to a preset differential encoding strategy and the compressed correction coefficients corresponding to the first gray level; Storing the compressed correction coefficients corresponding to the first gray level and the differential encoding corresponding to the correction coefficients of each gray level except the first gray level into the dynamic random access memory in the receiving card.

Images