An image compression method, system, and apparatus for use in LED display control systems.

Abstract

This invention discloses an image compression method, system, and apparatus for LED display control systems, belonging to the field of image compression. The method includes the following steps: hardware decoding of the HDMI or DVI video interface input signal to parse the original RGB image data stream; caching a complete image frame data in DRAM memory; performing overall compression encoding on the complete image frame data; packaging the image data after overall compression encoding into Ethernet frame data according to two-dimensional information; parsing the compressed image data from the Ethernet frame data; performing partitioned decompression decoding on the parsed image data; and caching the partitioned decompression decoding image frame data in DRAM memory. This invention employs an intra-frame compression transmission technology based on discrete wavelet transform, using overall compression at the encoding end and partitioned decompression at the decoding end, enabling decompression of compressed images at a very low cost.

Description

Technical Field

[0001] This invention relates to the field of image compression technology, specifically to an image compression method, system, and apparatus for use in LED display control systems. Background Technology

[0002] Currently, LED displays are typically composed of multiple small display units spliced ​​together. Each display unit is called a unit cabinet. The LED display control system is used to control the display of each pixel on the LED display screen. An LED display control system usually has at least two levels of topology: the first level is the transmitting end (transmitting card or other device), and the second level is the receiving end (receiving card). The receiving card is usually integrated with the LED display unit cabinet for display control (similar to the TCON in an LCD). Figure 4 As shown. The transmitting card and the receiving card are usually connected by a gigabit network cable, and the unit cabinets or receiving cards are also usually cascaded with each other by gigabit network cables.

[0003] For large stages or events, LED displays can achieve resolutions of tens of kilobytes per second (TK), requiring enormous bandwidth and resulting in extensive and complex wiring. Furthermore, as the pixel pitch of LED displays decreases to within 1.0mm, high-resolution (8K and above) displays will become increasingly mainstream, further limiting the layout space available for LED display control systems within the same area.

[0004] Currently, some solutions in the LED display control industry use 5G Ethernet cables as the transmission link between the transmitting and receiving cards to replace the previous gigabit Ethernet cable transmission, aiming to increase the transmission bandwidth of a single network cable. However, since LED displays typically have a large number of receiving cards, which are cost-sensitive products, 5G or 10G Ethernet transmission solutions often increase the cost of the receiving cards several times over, thus limiting their application to some high-end solutions at present.

[0005] It is evident that existing compression and decompression technologies cannot achieve decompression of compressed images at a minimal cost. Therefore, those skilled in the art provide an image compression method, system, and apparatus for LED display control systems to address the problems mentioned in the background section. Summary of the Invention

[0006] The purpose of this invention is to provide an image compression method, system, and apparatus for LED display control systems, which can decompress compressed images at a very low cost, thereby solving the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] An image compression method for use in LED display control systems includes the following steps:

[0009] Hardware decoding is performed on the HDMI or DVI video interface input signal to extract the original RGB image data stream;

[0010] A complete image frame data is cached in DRAM memory;

[0011] Perform overall compression encoding on the complete image frame data;

[0012] The image data, which has undergone overall compression and encoding, is packaged into Ethernet frame data according to two-dimensional information;

[0013] Extract compressed image data from Ethernet frame data;

[0014] The parsed image data is then partitioned, decompressed, and decoded.

[0015] The image frame data after partition decompression and decoding is cached in DRAM memory.

[0016] As a further aspect of the present invention: the overall compression encoding process includes:

[0017] Bit width extension maps the original 8-bit or 10-bit color depth data to higher bit precision;

[0018] Remove the DC component from the input RGB data;

[0019] The input data is converted from the RGB color space to the YCbCr color space to separate the brightness and chromaticity information.

[0020] Perform discrete wavelet transform on the vertical direction of the image to decompose the high-frequency and low-frequency components of the image in the vertical direction.

[0021] After vertical wavelet decomposition, the high and low frequency data are then subjected to discrete wavelet transform in the horizontal direction to decompose the high frequency and low frequency components of the image in the horizontal direction.

[0022] The coefficients obtained from wavelet decomposition of the two-dimensional image data are divided according to the two-dimensional region controlled by each receiving card.

[0023] Quantizing signal values ​​within a certain range into a representative value reduces data precision.

[0024] Encode the quantized low-precision data according to entropy information;

[0025] Entropy evaluation is performed on the coefficients after wavelet level decomposition;

[0026] Based on the entropy information of the image data, different quantization coefficients are applied to different regions.

[0027] As a further aspect of the present invention: the quantization adopts a time series generation method of vector quantization to model continuous image data in the time-frequency domain.

[0028] As a further aspect of the present invention: the partition decompression and decoding process includes:

[0029] The image data of the receiving card that needs to be decoded is divided into two-dimensional information in the parsed compressed image.

[0030] Entropy decoding is performed to obtain quantized data for each pixel.

[0031] The quantized data is used to reconstruct a new output based on the quantization step size;

[0032] Combine high-frequency and low-frequency data in the horizontal direction into new data;

[0033] Combine high-frequency and low-frequency data in the vertical direction into new data;

[0034] Convert data in the YCbCr color space to data in the RGB color space;

[0035] The original DC component in the image is superimposed onto the AC data portion;

[0036] Extracting the high-bit width data from the computation process to the actual image color depth is equivalent to converting floating-point data to integer data.

[0037] This application also discloses an image compression system for use in LED display control systems, comprising:

[0038] The video input module is used to perform hardware decoding on the HDMI or DVI video interface input signals and parse out the original RGB image data stream;

[0039] The first frame buffer module is used to buffer a complete image frame data in DRAM memory;

[0040] The overall compression module is used to perform overall compression encoding processing on complete image frame data;

[0041] The Ethernet packet generation module is used to package the image data, which has undergone overall compression and encoding, into Ethernet frame data according to two-dimensional information.

[0042] The Ethernet data parsing module is used to parse compressed image data from Ethernet frame data;

[0043] The partition decompression module is used to perform partition decompression and decoding processing on the parsed image data;

[0044] The second frame buffer module is used to cache the image frame data after partition decompression and decoding in DRAM memory.

[0045] As a further embodiment of the present invention: the overall compression module includes:

[0046] Bit width extension unit is used to extend the bit width of the original 8-bit or 10-bit color depth data to a higher bit precision;

[0047] The DC bias removal unit is used to remove the DC component from the input RGB data;

[0048] The RGB to YCbCr unit is used to convert input data from the RGB color space to the YCbCr color space representation, separating the luminance and chrominance information.

[0049] The wavelet vertical decomposition unit is used to perform discrete wavelet transform on the vertical direction of the image, decomposing the high-frequency and low-frequency components of the image in the vertical direction.

[0050] The wavelet horizontal decomposition unit is used to perform discrete wavelet transform in the horizontal direction on the high and low frequency data after wavelet vertical decomposition, thereby decomposing the high frequency and low frequency components of the image in the horizontal direction.

[0051] The partitioning unit is used to divide the coefficients obtained from the wavelet decomposition of the two-dimensional image data according to the two-dimensional region controlled by each receiving card.

[0052] A quantization unit is used to quantize a signal value within a certain range into a representative value, thereby reducing the precision of the data.

[0053] Entropy coding unit is used to encode quantized low-precision data according to entropy information.

[0054] Entropy analysis unit, used to evaluate the entropy of the coefficients after wavelet level decomposition;

[0055] The compression ratio control unit is used to determine the appropriate quantization coefficients to apply to different regions based on the entropy information of the image data.

[0056] As a further aspect of the present invention: the quantization unit adopts a time series generation method of vector quantization to model continuous image data in the time-frequency domain.

[0057] As a further embodiment of the present invention: the partition decompression module includes:

[0058] The partition selection unit is used to divide the image data of the receiving card into two-dimensional information in the parsed compressed image to be decoded.

[0059] The entropy decoding unit is used to perform entropy decoding to obtain quantized data for each pixel.

[0060] The dequantization unit is used to recover a new output from the quantized data based on the quantization step size;

[0061] The wavelet horizontal synthesis unit is used to synthesize high-frequency and low-frequency data in the horizontal direction into new data.

[0062] Wavelet vertical synthesis unit is used to synthesize high-frequency and low-frequency data in the vertical direction into new data;

[0063] The YCbCr to RGB unit is used to convert data in the YCbCr color space to data in the RGB space.

[0064] A DC bias unit is added to superimpose the original DC component in the image onto the AC data portion;

[0065] The bit-width truncation unit is used to truncate the high-bit-width data during the calculation process to the actual image color depth, which is equivalent to converting floating-point data to integer data.

[0066] This application also discloses an image compression device for use in an LED display control system, comprising:

[0067] A memory for storing instructions; wherein the instructions are instructions that implement the steps of an image compression method applied to an LED display control system;

[0068] A processor for executing instructions in the memory.

[0069] This application also discloses a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of an image compression method applied to an LED display control system.

[0070] Compared with the prior art, the beneficial effects of the present invention are:

[0071] This application employs an intra-frame compression transmission technology based on discrete wavelet transform between the transmitting and receiving cards. The encoding end uses overall compression, while the decoding end uses partitioned decompression. Specifically, compression encoding is performed on the transmitting card, and partitioned decompression and decoding are performed on the receiving card. Due to the partitioned decompression method, compared to some existing compression and decompression technologies, the consumption of logic, storage resources, and bandwidth at the compression and decompression ends is significantly reduced. This allows for the decompression of compressed images with minimal cost increase compared to existing low-cost receiving card solutions. Furthermore, this application addresses the fact that no one has yet implemented compressed transmission in LED display control systems. Based on the one-to-many topology of LED display control systems, it proposes an overall compression and partitioned decompression approach, achieving efficiency gains with virtually no increase in receiving card costs. Attached Figure Description

[0072] Figure 1 This is a flowchart of an image compression method applied to an LED display control system according to this application;

[0073] Figure 2 This is a structural block diagram of an image compression system applied to an LED display control system according to this application;

[0074] Figure 3 This is a combined view of the sending card and the receiving card in an embodiment of this application;

[0075] Figure 4 This is a combined view of the transmitting card and receiving card in the prior art of this application;

[0076] Figure 5 This is a flowchart of vector quantization in an embodiment of this application. Detailed Implementation

[0077] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0078] As mentioned in the background section of this application, the inventors have discovered through research that there are two existing image compression methods for LED display control systems. The first method uses gigabit network cables for connection, but this method has certain drawbacks. When laying out large LED displays, the on-site wiring is usually very large and complex, while when laying out small LED displays, the space is tight and difficult to arrange. The second method uses 5G Ethernet cables for connection, but this method also has certain drawbacks. Since there are usually many receiver cards on LED displays, and receiver cards are cost-sensitive products, 5G or 10G Ethernet transmission solutions often increase the cost of receiver cards several times, causing the overall cost to rise sharply.

[0079] To address the aforementioned shortcomings, this application discloses an image compression method, system, and apparatus for LED display control systems. It employs an intra-frame compression transmission technology based on discrete wavelet transform between the transmitting and receiving cards. The encoding end uses overall compression, while the decoding end uses partitioned decompression. Specifically, compression encoding is performed on the transmitting card, and partitioned decompression and decoding are performed on the receiving card. Due to the partitioned decompression method, compared to some existing compression and decompression technologies, the consumption of logic, storage resources, and bandwidth at the decompression end is significantly reduced. This allows for the decompression of compressed images with minimal cost increase compared to existing low-cost receiving card solutions. It should be noted that, as... Figure 3 As shown, the encoding end, i.e., the encoder, is integrated on the transmitting card, and the decoding end, i.e., the decoder, is integrated on the receiving card.

[0080] The following will describe in detail, with reference to the accompanying drawings, how the solution of this application solves the above-mentioned technical problems.

[0081] Please see Figures 1-5 In this embodiment of the invention, an image compression method applied to an LED display control system includes the following steps:

[0082] Hardware decoding is performed on the HDMI or DVI video interface input signal to extract the original RGB image data stream;

[0083] A complete image frame data is cached in DRAM memory;

[0084] Perform overall compression encoding on the complete image frame data;

[0085] The image data, which has undergone overall compression and encoding, is packaged into Ethernet frame data according to two-dimensional information;

[0086] Extract compressed image data from Ethernet frame data;

[0087] The parsed image data is then partitioned, decompressed, and decoded.

[0088] The image frame data after partition decompression and decoding is cached in DRAM memory.

[0089] This application enables the decompression of compressed images at a very low cost.

[0090] In this embodiment, the overall compression encoding process includes: the input video source RGB data is typically 8-bit or 10-bit color depth; the bit width is extended to map the original 8-bit or 10-bit color depth data to a higher bit precision, which is equivalent to achieving higher precision in subsequent calculations and retaining more decimal places; the DC component in the input RGB data is removed; the input data is converted from the RGB color space to the YCbCr color space representation, separating the luminance and chrominance information to facilitate the processing of luminance and chrominance differences during subsequent compression; a discrete wavelet transform is performed on the vertical direction of the image to decompose the high-frequency and low-frequency components of the image in the vertical direction; and the high and low frequency data after the wavelet vertical decomposition are further processed... A discrete wavelet transform is performed in the horizontal direction to decompose the high-frequency and low-frequency components of the image in the horizontal direction. The coefficients obtained from the wavelet decomposition of the two-dimensional image data are divided according to the two-dimensional region controlled by each receiver card. Quantizing the signal value within a certain interval (quantization step size) into a representative value to reduce the data precision is a key step in achieving data compression. The quantized low-precision data is encoded according to entropy information. The basic principle is that the number of codewords used for data encoding is proportional to the entropy information. Entropy evaluation is performed on the coefficients after the horizontal wavelet decomposition. Based on the entropy information of the image data, different quantization coefficients are applied to different regions. The quantization coefficients directly determine the precision of the quantized data, thereby controlling the compression ratio.

[0091] In this embodiment, the input data can first be converted from the RGB color space to the YCbCr color space, and then the YCbCr data can be de-DC biased, that is, the DC component in the input YCbCr data can be removed.

[0092] In this embodiment: the quantization adopts the vector quantization time series generation method, which models continuous image data in the time-frequency domain and divides it into low frequency (LF) and high frequency (HF) to retain the important features of the time series while also achieving compression.

[0093] In this embodiment, the partitioned decompression and decoding process includes: dividing the image data of the receiving card into regions that need to be decoded according to two-dimensional information in the parsed compressed image; performing entropy decoding to obtain quantized data pixel by pixel; recovering the new output from the quantized data according to the quantization step size. The data recovered after quantization and dequantization will introduce a certain reconstruction error depending on the compression ratio, which is reflected in the image as image distortion; combining high-frequency and low-frequency data in the horizontal direction into new data; combining high-frequency and low-frequency data in the vertical direction into new data; converting the data in the YCbCr color space into data in the RGB space; superimposing the original DC component in the image onto the AC data part; and extracting the high-bit-width data in the calculation process to the actual image color depth, which is equivalent to converting floating-point data into integer data.

[0094] This application also discloses an image compression system for LED display control systems, comprising: a video input module for hardware decoding of HDMI or DVI video interface input signals to parse the original RGB image data stream; a first frame buffer module for buffering a complete image frame data in DRAM memory; an overall compression module for performing overall compression encoding on the complete image frame data; an Ethernet packet generation module for packaging the image data after overall compression encoding into Ethernet frame data according to two-dimensional information; an Ethernet data parsing module for parsing the compressed image data from the Ethernet frame data; a partition decompression module for performing partition decompression decoding on the parsed image data; and a second frame buffer module for buffering the partitioned decompression decoding image frame data in DRAM memory.

[0095] It should be noted that both the first and second frame buffer modules include frame buffer circuits. This circuit design refers to the design of a circuit capable of reading and writing video data from a memory buffer. A frame buffer circuit typically includes the following components: a FIFO (First-In, First-Out) memory for buffering continuous data streams, preventing data loss during loading and storage operations, and allowing the system to perform DMA (Direct Memory Access) operations to improve data transfer speed; control logic for controlling the FIFO's read and write operations based on read / write clocks and enable signals; and a video interface for converting the data in the FIFO into a format suitable for the display, such as VGA or HDMI. Furthermore, some frame buffer circuits can implement double buffering or page-flipping techniques, using two frame buffers to alternately display the current frame and the next frame's data to improve image smoothness and stability.

[0096] In this embodiment, the overall compression module includes: a bit-width expansion unit for expanding the original 8-bit or 10-bit color depth data mapping to higher bit precision; a DC bias removal unit for removing the DC component from the input RGB data; an RGB-to-YCbCr unit for converting the input data from the RGB color space to the YCbCr color space representation, separating the luminance and chrominance information; a wavelet vertical decomposition unit for performing discrete wavelet transform on the vertical direction of the image, decomposing the high-frequency and low-frequency components of the image in the vertical direction; and a wavelet horizontal decomposition unit for decomposing the high and low frequencies after the wavelet vertical decomposition. The image data is processed by performing a discrete wavelet transform in the horizontal direction to decompose the high-frequency and low-frequency components in the horizontal direction. A partitioning unit divides the coefficients from the wavelet decomposition of the two-dimensional image data according to the two-dimensional region controlled by each receiver card. A quantization unit quantizes the signal values ​​within a certain interval into a representative value, reducing the data precision. An entropy coding unit encodes the quantized low-precision data according to entropy information. An entropy analysis unit evaluates the entropy of the coefficients after horizontal wavelet decomposition. A compression ratio control unit determines the application of different quantization coefficients to different regions based on the entropy information of the image data.

[0097] It's important to note that bit-width expansion units (BWL) extend the bit width of an image using an image bit-width expansion algorithm. This algorithm is a type of image processing algorithm used to change the resolution or compression ratio of an image. One commonly used BWL algorithm is interpolation, which estimates unknown pixel values ​​based on known pixel values, thereby increasing or decreasing the number of bits in the image. There are various types of interpolation algorithms, such as nearest neighbor interpolation, linear interpolation, and bicubic interpolation. Another commonly used BWL algorithm is transform, which converts the image from the spatial domain to the frequency domain for processing, enabling image enhancement, filtering, and compression. Transform algorithms also come in various types, such as Fourier transform, Walsh transform, and discrete cosine transform.

[0098] Different image bit-width extension algorithms have different advantages, disadvantages, and application scenarios:

[0099] The advantages of interpolation algorithms are their simplicity and ease of implementation, allowing for rapid image scaling. The disadvantages include potential image distortion, blurring, or jagged edges, especially with large interpolation ratios. Applications of interpolation algorithms include image scaling, rotation, and distortion correction.

[0100] The advantage of transform algorithms is that they can perform complex image processing in the frequency domain, such as enhancement, filtering, and compression, which can improve image quality or save storage space. The disadvantages are that they require significant computational resources and time, and may introduce noise or distortion. Applications of transform algorithms include image compression, encryption, watermarking, and feature extraction.

[0101] To avoid the drawbacks of interpolation and transformation algorithms while retaining their advantages, this application employs a bilinear interpolation algorithm. This is a linear interpolation algorithm, but instead of using only two known pixel values ​​to estimate unknown pixel values, it uses four known pixel values ​​for interpolation. This retains the advantages of simple implementation and fast processing of interpolation algorithms, while avoiding the jagged and blurring effects of nearest-neighbor interpolation algorithms, thus improving image quality. The bilinear interpolation algorithm can also be combined with transformation algorithms to process the image in the frequency domain and then transform it back to the spatial domain, thereby achieving image enhancement. The specific steps are as follows: 1. Perform wavelet transform on the original image to obtain its vertical and horizontal component frequency domain representations; 2. Perform bilinear interpolation on the frequency domain representation; 3. Perform inverse Fourier transform on the transformed frequency domain image to obtain the interpolated image.

[0102] In this embodiment: the quantization unit employs a vector quantization time series generation method, modeling continuous image data in the time-frequency domain, dividing it into low-frequency (LF) and high-frequency (HF) sequences. This preserves important features of the time series while achieving compression. The specific process of vector quantization is as follows: Figure 5 As shown, the codebook stores K discrete codes or discrete latent vectors, and uses neural network learning to project the input continuous sampling rate space onto a discrete sampling rate space. The codebook is responsible for transforming the continuous sampling rate space into a discrete sampling rate space through an iterative optimization process. In this process, each continuous vector is compared with each discrete vector in the codebook by Euclidean distance, and replaced with the closest discrete vector.

[0103] In this embodiment, the partition decompression module includes: a partition selection unit, used to divide the image data of the receiving card to be decoded in the parsed compressed image according to two-dimensional information; an entropy decoding unit, used to perform entropy decoding to obtain quantized data pixel by pixel; an inverse quantization unit, used to recover a new output from the quantized data according to the quantization step size; a wavelet horizontal synthesis unit, used to synthesize high-frequency and low-frequency data in the horizontal direction into new data; a wavelet vertical synthesis unit, used to synthesize high-frequency and low-frequency data in the vertical direction into new data; a YCbCr to RGB conversion unit, used to convert data in the YCbCr color space into data in the RGB space; a DC bias addition unit, used to superimpose the original DC component in the image onto the AC data part; and a bit width truncation unit, used to truncate the high bit width data in the calculation process to the actual image color depth, which is equivalent to converting floating-point data to integer data.

[0104] This application also discloses an image compression device for use in an LED display control system, comprising: a memory for storing instructions; wherein the instructions are instructions that implement the steps of an image compression method for use in an LED display control system; and a processor for executing the instructions in the memory.

[0105] This application also discloses a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of an image compression method applied to an LED display control system.

[0106] This invention employs an intra-frame compression transmission technology based on discrete wavelet transform between the transmitting and receiving cards. The encoding end uses overall compression, while the decoding end uses partitioned decompression. Specifically, compression encoding is performed on the transmitting card, and partitioned decompression decoding is performed on the receiving card. Due to the partitioned decompression method, compared to some existing compression and decompression technologies, the consumption of logic, storage resources, and bandwidth at the compression and decompression ends is significantly reduced. This allows for the decompression of compressed images with minimal cost increase compared to existing low-cost receiving card solutions. Furthermore, this application addresses the fact that compressed transmission has never been implemented in LED display control systems. Based on the one-to-many topology of LED display control systems, it proposes an overall compression and partitioned decompression approach, achieving efficiency gains with virtually no increase in receiving card costs.

[0107] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

[0108] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. An image compression method applied to an LED display control system, characterized in that, Includes the following steps: Hardware decoding is performed on the HDMI or DVI video interface input signal to extract the original RGB image data stream; A complete image frame data is cached in DRAM memory; Perform overall compression encoding on the complete image frame data; The image data, which has undergone overall compression and encoding, is packaged into Ethernet frame data according to two-dimensional information; Extract compressed image data from Ethernet frame data; The parsed image data is then partitioned, decompressed, and decoded. The image frame data after partition decompression and decoding is cached in DRAM memory; The overall compression encoding process includes: Bit width extension maps the original 8-bit or 10-bit color depth data to higher bit precision; Remove the DC component from the input RGB data; The input data is converted from the RGB color space to the YCbCr color space to separate the brightness and chromaticity information. Perform discrete wavelet transform on the vertical direction of the image to decompose the high-frequency and low-frequency components of the image in the vertical direction. After vertical wavelet decomposition, the high and low frequency data are then subjected to discrete wavelet transform in the horizontal direction to decompose the high frequency and low frequency components of the image in the horizontal direction. The coefficients obtained from wavelet decomposition of the two-dimensional image data are divided according to the two-dimensional region controlled by each receiving card. Quantizing signal values ​​within a certain range into a representative value reduces data precision. Encode the quantized low-precision data according to entropy information; Entropy evaluation is performed on the coefficients after wavelet level decomposition; Based on the entropy information of the image data, different quantization coefficients are applied to different regions.

2. The image compression method for an LED display control system according to claim 1, characterized in that, The quantization employs a vector quantization time series generation method to model continuous image data in the time-frequency domain.

3. The image compression method for an LED display control system according to claim 1, characterized in that, The partition decompression and decoding process includes: The image data of the receiving card that needs to be decoded is divided into two-dimensional information in the parsed compressed image. Entropy decoding is performed to obtain quantized data for each pixel. The quantized data is used to reconstruct a new output based on the quantization step size; Combine high-frequency and low-frequency data in the horizontal direction into new data; Combine high-frequency and low-frequency data in the vertical direction into new data; Convert data in the YCbCr color space to data in the RGB color space; The original DC component in the image is superimposed onto the AC data portion; Extracting the high-bit width data from the computation process to the actual image color depth is equivalent to converting floating-point data to integer data.

4. An image compression system applied to an LED display control system, characterized in that, include: The video input module is used to perform hardware decoding on the HDMI or DVI video interface input signals and parse out the original RGB image data stream; The first frame buffer module is used to buffer a complete image frame data in DRAM memory; The overall compression module is used to perform overall compression encoding processing on complete image frame data; The Ethernet packet generation module is used to package the image data, which has undergone overall compression and encoding, into Ethernet frame data according to two-dimensional information. The Ethernet data parsing module is used to parse compressed image data from Ethernet frame data; The partition decompression module is used to perform partition decompression and decoding processing on the parsed image data; The second frame buffer module is used to buffer the image frame data after partition decompression and decoding in DRAM memory; The overall compression module includes: Bit width extension unit is used to extend the bit width of the original 8-bit or 10-bit color depth data to a higher bit precision; The DC bias removal unit is used to remove the DC component from the input RGB data; The RGB to YCbCr unit is used to convert input data from the RGB color space to the YCbCr color space representation, separating the luminance and chrominance information. The wavelet vertical decomposition unit is used to perform discrete wavelet transform on the vertical direction of the image, decomposing the high-frequency and low-frequency components of the image in the vertical direction. The wavelet horizontal decomposition unit is used to perform discrete wavelet transform in the horizontal direction on the high and low frequency data after wavelet vertical decomposition, thereby decomposing the high frequency and low frequency components of the image in the horizontal direction. The partitioning unit is used to divide the coefficients obtained from the wavelet decomposition of the two-dimensional image data according to the two-dimensional region controlled by each receiving card. A quantization unit is used to quantize a signal value within a certain range into a representative value, thereby reducing the precision of the data. Entropy coding unit is used to encode quantized low-precision data according to entropy information. Entropy analysis unit, used to evaluate the entropy of the coefficients after wavelet level decomposition; The compression ratio control unit is used to determine the appropriate quantization coefficients to apply to different regions based on the entropy information of the image data.

5. The image compression system for an LED display control system according to claim 4, characterized in that, The quantization unit employs a vector quantization time series generation method to model continuous image data in the time-frequency domain.

6. The image compression system for an LED display control system according to claim 4, characterized in that, The partition decompression module includes: The partition selection unit is used to divide the image data of the receiving card into two-dimensional information in the parsed compressed image to be decoded. The entropy decoding unit is used to perform entropy decoding to obtain quantized data for each pixel. The dequantization unit is used to recover a new output from the quantized data based on the quantization step size; The wavelet horizontal synthesis unit is used to synthesize high-frequency and low-frequency data in the horizontal direction into new data. Wavelet vertical synthesis unit is used to synthesize high-frequency and low-frequency data in the vertical direction into new data; The YCbCr to RGB unit is used to convert data in the YCbCr color space to data in the RGB space. A DC bias unit is added to superimpose the original DC component in the image onto the AC data portion; The bit-width truncation unit is used to truncate the high-bit-width data during the calculation process to the actual image color depth, which is equivalent to converting floating-point data to integer data.

7. An image compression device for use in an LED display control system, characterized in that, include: A memory for storing instructions; wherein the instructions are instructions capable of implementing the steps of the image compression method applied to an LED display control system as described in any one of claims 1-3; A processor for executing instructions in the memory.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the image compression method for an LED display control system as described in any one of claims 1-3.

Images