WO2020113766A1 - 补偿表压缩方法 - Google Patents

补偿表压缩方法 Download PDF

Info

Publication number
WO2020113766A1
WO2020113766A1 PCT/CN2019/071033 CN2019071033W WO2020113766A1 WO 2020113766 A1 WO2020113766 A1 WO 2020113766A1 CN 2019071033 W CN2019071033 W CN 2019071033W WO 2020113766 A1 WO2020113766 A1 WO 2020113766A1
Authority
WO
WIPO (PCT)
Prior art keywords
compensation table
sub
block
quantization
compression method
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2019/071033
Other languages
English (en)
French (fr)
Inventor
程琳
邓宇帆
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen China Star Optoelectronics Semiconductor Display Technology Co Ltd
Original Assignee
Shenzhen China Star Optoelectronics Semiconductor Display Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shenzhen China Star Optoelectronics Semiconductor Display Technology Co Ltd filed Critical Shenzhen China Star Optoelectronics Semiconductor Display Technology Co Ltd
Publication of WO2020113766A1 publication Critical patent/WO2020113766A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/14—Digital output to display device ; Cooperation and interconnection of the display device with other functional units
    • G06F3/1407—General aspects irrespective of display type, e.g. determination of decimal point position, display with fixed or driving decimal point, suppression of non-significant zeros
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T9/00—Image coding
    • G06T9/007—Transform coding, e.g. discrete cosine transform

Definitions

  • the invention relates to the field of image processing, in particular to a compensation table compression method.
  • Data compression is a relatively mature technology for reducing data size. It is applied to the data stored in the memory subsystem of the computer system to increase the storage capacity. Data compression is also used when data is transferred between different subsystems within a computer system, or generally when the transfer is performed between two points in a data communication system that includes a communication network.
  • Data compression requires two basic operations: 1. Compression (also called encoding). Compression is to take uncompressed data as input and pass the corresponding codeword (also called encoding, word code or code in the literature) Replace data values to convert uncompressed data into compressed data: 2. Decompression (also called decoding). Decompression uses compressed data as input and replaces the codeword with the corresponding data value. The compressed data is converted to uncompressed. Data compression can be lossless or lossy, depending on whether the actual data value after decompression is exactly the same as the original data value before compression (lossless), or whether the data value after decompression is different The original data value and the original value cannot be obtained (lossy type). Compression and decompression can be implemented in software, or hardware, or a combination of software and hardware to implement corresponding methods, devices, and systems.
  • the compensation table is usually used to store the compensation information of each pixel.
  • the driver board looks up the compensation table, adjusts the signal, and adjusts the signal in the dark area of the panel up, and the signal in the bright area down, showing a uniform display effect.
  • each pixel corresponds to a set of compensation information, and each set of compensation information contains one or more compensation data.
  • the physical meaning of the compensation data depends on the algorithm, which is usually the adjustment value of a specific gray scale, and some algorithms set it to the voltage value to be adjusted.
  • the lossy compression algorithm of the compensation table in the prior art will visually see the loss.
  • the size of the compensation table It is equal to the number of panel pixels multiplied by the size of each set of compensation information. If a 55-inch UHD (Ultra High Definition) RGB panel is compensated (each pixel includes three sub-pixels of red, green and blue), assuming that each sub-pixel compensation information is 24 bits, the compensation table data The quantity is 2160*3840*24bit*3 ⁇ 597Mb.
  • the compensation table of the prior art occupies a lot of system storage resources, which has high requirements on the hardware system, and the process of transmitting and burning data on the production line takes time.
  • the purpose of the present invention is to provide a compensation table compression method, which can obtain the optimal quantization step without iteration, reduce the occupation of memory space, save the hardware resources of the system, and can reduce the cost and reduce the cost of transmitting and burning data time.
  • the present invention provides a compensation table compression method, including the following steps:
  • Step S1 Divide the original compensation table into multiple sub-blocks of the same size in the air domain
  • Step S2 Multiply the original compensation table divided into multiple sub-blocks by the sobel operator to obtain a gradient vector, and calculate the variance of each block separately;
  • Step S4 Each sub-block is quantized separately according to its corresponding quantization step size to obtain a quantization compensation table, and the coding algorithm is used to compress the quantization compensation table.
  • Step S5 Decompress the compressed quantization compensation table, and then dequantize each sub-block according to the information in the quantization compensation table to reconstruct the original compensation table.
  • each sub-block is 8*8.
  • step S2 the edge of each sub-block is detected by the sobel operator.
  • step S3 the calculation formula of the quantization step size is obtained through a visual lossless subjective experiment, where a and b are parameters obtained by fitting the results of the visual lossless subjective experiment.
  • the invention also provides a compensation table compression method, including the following steps:
  • Step S1' DCT transform the original compensation table, convert the original compensation table from the spatial domain to the frequency domain, and divide it into multiple sub-blocks of the same size;
  • Step S4' Each sub-block is quantized separately according to its corresponding quantization step size to obtain a quantization compensation table, and the encoding algorithm is used to compress the quantization compensation table.
  • Step S5' Decompress the compressed quantization compensation table, and then inverse quantize and inverse DCT transform each sub-block to reconstruct the original compensation table according to the information in the quantization compensation table.
  • each sub-block is 8*8.
  • the size of the lower right corner area of the DCT coefficient of each sub-block is 4*4; the DC coefficient is the value of the upper left corner of each sub-block.
  • step S3' the calculation formula of the quantization step size is obtained through a visual lossless subjective experiment, where a and b are parameters obtained by fitting the results of the visual lossless subjective experiment.
  • the invention also provides a compensation table compression method, including the following steps:
  • Step S1' DCT transform the original compensation table, convert the original compensation table from the spatial domain to the frequency domain, and divide it into multiple sub-blocks of the same size;
  • Step S4' each sub-block is quantized separately according to its corresponding quantization step size to obtain a quantization compensation table, and the coding algorithm is used to compress the quantization compensation table;
  • Step S5' by decompressing the compressed quantization compensation table, and then dequantizing and inverse DCT transforming each sub-block to reconstruct the original compensation table according to the information in the quantization compensation table;
  • each sub-block is 8*8.
  • the compensation table compression method of the present invention makes the human eye not notice the quality loss of the original compensation table through visual lossless compression of the compensation table, and greatly increases the compression efficiency, compared with the existing compensation table compression algorithm
  • the invention can obtain the optimal quantization step without iteration, reduce the occupation of memory space, save the hardware resources of the system, and can reduce the cost and the time spent in transferring and burning data.
  • FIG. 1 is a flowchart of a first embodiment of a compensation table compression method of the present invention
  • FIG. 2 is a logic diagram of a first embodiment of a compensation table compression method of the present invention
  • FIG. 3 is a flowchart of a second embodiment of the compensation table compression method of the present invention.
  • FIG. 4 is a logic diagram of a second embodiment of the compensation table compression method of the present invention.
  • the present invention provides a first embodiment of a compensation table compression method, including the following steps:
  • Step S1 Divide the original compensation table into multiple sub-blocks of the same size in the air domain
  • Step S2 Multiply the original compensation table divided into multiple sub-blocks by the sobel operator to obtain a gradient vector, and calculate the variance of each block separately;
  • Step S4 Each sub-block is quantized separately according to its corresponding quantization step size to obtain a quantization compensation table, and the coding algorithm is used to compress the quantization compensation table.
  • step S5 Decompressing the compressed quantization compensation table, and then dequantizing each sub-block according to the information in the quantization compensation table, the original compensation table can be reconstructed.
  • each sub-block is 8*8, that is, each sub-block corresponds to 8 rows and 8 columns of pixels (a total of 64 pixels) of compensation information.
  • step S2 the edge of each sub-block is detected by the sobel operator to determine the location of each sub-block.
  • step S3 the calculation formula of the quantization step size can be obtained according to the visual lossless subjective experiment, so as to losslessly compress the quantization compensation table, where a and b are parameters obtained by fitting the results of the visual lossless subjective experiment.
  • the present invention uses visual lossless compression compensation table to make human eyes not notice the quality loss of the original compensation table, and greatly increases the compression efficiency.
  • the present invention does not require iteration It can obtain the best quantization step size, reduce the occupation of memory space, save the hardware resources of the system, and can reduce the cost and reduce the time spent in transferring and burning data.
  • the present invention provides a second embodiment of a compensation table compression method, including the following steps:
  • Step S1' Perform a DCT transform (discrete cosine transform) on the original compensation table, convert the original compensation table from the spatial domain to the frequency domain, and divide it into multiple sub-blocks of the same size;
  • DCT transform discrete cosine transform
  • Step S4' Each sub-block is quantized separately according to its corresponding quantization step size to obtain a quantization compensation table, and the encoding algorithm is used to compress the quantization compensation table.
  • step S5' decompressing the compressed quantization compensation table, dequantizing each sub-block according to the information in the quantization compensation table, and inversely DCT transforming each sub-block to reconstruct the original compensation table.
  • each sub-block is 8*8, that is, each sub-block corresponds to 8 rows and 8 columns of pixels (a total of 64 pixels) of compensation information.
  • the size of the lower right corner area of each sub-block DCT coefficient is 4*4, that is, the compensation information of the lower right corner area of each sub-block DCT coefficient corresponds to 4 rows and 4 columns of pixels (a total of 16 pixels); the DC coefficient is in each sub-block The value in the upper left corner.
  • the calculation formula of the quantization step size can be obtained according to the visual lossless subjective experiment, and thus the lossless compression quantization compensation table, where a and b are parameters obtained by fitting the results of the visual lossless subjective experiment.
  • the present invention uses visual lossless compression compensation table to make human eyes not notice the quality loss of the original compensation table, and greatly increases the compression efficiency.
  • the present invention does not require iteration It can obtain the best quantization step size, reduce the occupation of memory space, save the hardware resources of the system, and can reduce the cost and reduce the time spent in transferring and burning data.
  • the compensation table compression method of the present invention compresses the compensation table visually, so that human eyes cannot perceive the quality loss of the original compensation table, and greatly increases the compression efficiency.
  • the invention can obtain the optimal quantization step size without iteration, reduce the occupation of the memory space, save the hardware resources of the system, and can reduce the cost and the time spent in transferring and burning data.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Discrete Mathematics (AREA)
  • Multimedia (AREA)
  • Human Computer Interaction (AREA)
  • General Engineering & Computer Science (AREA)
  • Compression Or Coding Systems Of Tv Signals (AREA)
  • Compression, Expansion, Code Conversion, And Decoders (AREA)
  • Compression Of Band Width Or Redundancy In Fax (AREA)

Abstract

本发明提供一种补偿表压缩方法。该补偿表压缩方法通过视觉无损压缩补偿表，使人眼察觉不出原始补偿表的质量损失，且极大增加压缩效率，和现有的补偿表压缩算法相比，本发明无需迭代即可获得最佳量化步长，减少对内存空间的占用，节省系统的硬件资源，且可以降低成本及减少传输和烧录数据耗费的时间。

Description

补偿表压缩方法 技术领域
本发明涉及图像处理领域，尤其涉及一种补偿表压缩方法。
背景技术
数据压缩是用于减小数据大小的一种较为成熟的技术。其应用于保存在计算机系统的存储器子系统中的数据以，增加存储能力。当数据在计算机系统内的不同子系统之间传输时，或者通常当在包括通信网络的数据通信系统中的两个点之间进行所述传输时，数据压缩也被使用。
数据压缩需要两个基本的操作:1、压缩(也称为编码)，压缩是将未压缩的数据作为输入，并通过用相应的码字(在文献中也称为编码、字码或代码)替换数据值来将未压缩的数据转换为经压缩的数据:2、解压缩(也称为解码)，解压缩是将经压缩的数据作为输入并通过用相应的数据值替换码字来将该经压缩的数据转换为未压缩的。数据压缩可以是无损式的或者有损式的，这取决于是否解压缩后的实际数据值与压缩前的原始数据值完全相同(无损式)，或者取决于是否解压缩后的数据值不同于原始数据值且原始值无法取得(有损式)。可以用软件、或硬件、或软件和硬件的组合来实施压缩和解压缩，以实现相应的方法、设备和系统。
为消除显示器的Mura，通常采用补偿表存储各像素的补偿信息。显示影像时，驱动板(TCON)查找补偿表，调整信号，将面板过暗区域的信号调高，过亮区域的信号调低，呈现均匀的显示效果。在补偿表中，每个像素对应于一组补偿信息，每组补偿信息包含一个或多个补偿数据。补偿数据的物理意义视算法而定，通常为特定灰阶的调整值，也有算法将其设定为待调整的电压值。现有技术的补偿表有损压缩算法，在视觉上会看出损失，使用二分法，需要大量迭代，不断迭代求最佳量化步长，对硬件要求高且较耗时；而补偿表的大小等于面板像素数目乘以每组补偿信息的大小，如果补偿55寸UHD(超高清)RGB面板(每个像素包括红绿蓝三个子像素)，假设每个子像素补偿信息为24bit，则补偿表数据量大小为2160*3840*24bit*3≈597Mb，现有技术的补偿表占用大量系统存储资源，对硬件系统要求较高，并在产线上传输、烧录数据的过程耗费时间。
发明内容
本发明的目的在于提供一种补偿表压缩方法，无需迭代即可获得最佳量化步长，减少对内存空间的占用，节省系统的硬件资源，且可以降低成本及减少传输和烧录数据耗费的时间。
为实现上述目的，本发明提供了一种补偿表压缩方法，包括如下步骤：
步骤S1、将原始补偿表在空域中分成多个相同大小的子块；
步骤S2、对分成多个子块后的原始补偿表乘以sobel算子得到梯度矢量，分别计算每个字块的方差；
步骤S3、根据量化步长计算公式：Q step＝aσ 2+b得到每个子块对应的量化步长，其中，Q step为量化步长，σ 2为方差，a和b均为常数；
步骤S4、每个子块根据其对应的量化步长分别量化，得到量化补偿表，采用编码算法对量化补偿表进行压缩。
步骤S5、通过解压压缩后的量化补偿表，再根据量化补偿表中的信息反量化每个子块重建原始补偿表。
每个子块的大小为8*8。
所述步骤S2中通过sobel算子检测每个子块的边缘。
所述步骤S3中通过视觉无损主观实验得到量化步长计算公式，其中，a和b是通过拟合视觉无损主观实验结果得到的参数。
本发明还提供一种补偿表压缩方法，包括如下步骤：
步骤S1’、对原始补偿表进行DCT变换，将原始补偿表从空域转换到频域，且分成多个相同大小的子块；
步骤S2’、根据能量函数计算公式：E＝sum(D)+K计算每个子块DCT系数右下角区域的能量函数，其中，sum(D)为子块DCT系数右下角区域中每个点的绝对值之和，K为DC系数，E为每个子块DCT系数右下角区域的能量函数；
步骤S3’、根据量化步长计算公式：Q step＝aE+b得到每个子块对应的量化步长，其中，Qstep为量化步长，a和b均为常数；
步骤S4’、每个子块根据其对应的量化步长分别量化，得到量化补偿表，采用编码算法对量化补偿表进行压缩。
步骤S5’、通过解压压缩后的量化补偿表，再根据量化补偿表中的信息 反量化及反DCT变换每个子块重建原始补偿表。
每个子块的大小为8*8。
每个子块DCT系数右下角区域的大小为4*4；DC系数为每个子块中最左上角的值。
所述步骤S3’中通过视觉无损主观实验得到量化步长计算公式，其中，a和b是通过拟合视觉无损主观实验结果得到的参数。
本发明还提供一种补偿表压缩方法，包括如下步骤：
步骤S1’、对原始补偿表进行DCT变换，将原始补偿表从空域转换到频域，且分成多个相同大小的子块；
步骤S2’、根据能量函数计算公式：E＝sum(D)+K计算每个子块DCT系数右下角区域的能量函数，其中，sum(D)为子块DCT系数右下角区域中每个点的绝对值之和，K为DC系数，E为每个子块DCT系数右下角区域的能量函数；
步骤S3’、根据量化步长计算公式：Q step＝aE+b得到每个子块对应的量化步长，其中，Q step为量化步长，a和b均为常数；
步骤S4’、每个子块根据其对应的量化步长分别量化，得到量化补偿表，采用编码算法对量化补偿表进行压缩；
步骤S5’、通过解压压缩后的量化补偿表，再根据量化补偿表中的信息反量化及反DCT变换每个子块重建原始补偿表；
每个子块的大小为8*8。
本发明的有益效果：本发明的补偿表压缩方法通过视觉无损压缩补偿表，使人眼察觉不出原补偿表的质量损失，且极大增加压缩效率，和现有的补偿表压缩算法相比，本发明无需迭代即可获得最佳量化步长，减少对内存空间的占用，节省系统的硬件资源，且可以降低成本及减少传输和烧录数据耗费的时间。
附图说明
为了能更进一步了解本发明的特征以及技术内容，请参阅以下有关本发明的详细说明与附图，然而附图仅提供参考与说明用，并非用来对本发明加以限制。
附图中，
图1为本发明的补偿表压缩方法第一实施例的流程图；
图2为本发明的补偿表压缩方法第一实施例的逻辑图；
图3为本发明的补偿表压缩方法第二实施例的流程图；
图4为本发明的补偿表压缩方法第二实施例的逻辑图。
具体实施方式
为更进一步阐述本发明所采取的技术手段及其效果，以下结合本发明的优选实施例及其附图进行详细描述。
请参阅图1及图2，本发明提供一种补偿表压缩方法的第一实施例，包括如下步骤：
步骤S1、将原始补偿表在空域中分成多个相同大小的子块，
步骤S2、对分成多个子块后的原始补偿表乘以sobel算子得到梯度矢量，分别计算每个字块的方差；
步骤S3、根据量化步长计算公式：Q step＝aσ 2+b得到每个子块对应的量化步长，其中，Q step为量化步长，σ 2为方差，a和b均为常数；
步骤S4、每个子块根据其对应的量化步长分别量化，得到量化补偿表，采用编码算法对量化补偿表进行压缩。
具体的，还包括步骤S5、通过解压压缩后的量化补偿表，再根据量化补偿表中的信息反量化每个子块，即可重建原始补偿表。
具体的，每个子块的大小为8*8，即每个子块对应8行8列像素(总共64个像素)的补偿信息。
具体的，所述步骤S2中通过sobel算子检测每个子块的边缘，确定每个子块所在的位置。
具体的，所述步骤S3中，量化步长计算公式可以根据视觉无损主观实验得到，从而无损压缩量化补偿表，其中，a和b是通过拟合视觉无损主观实验结果得到的参数。
需要说明的是，本发明通过视觉无损压缩补偿表，使人眼察觉不出原补偿表的质量损失，且极大增加压缩效率，和现有的补偿表压缩算法相比，本发明无需迭代即可获得最佳量化步长，减少对内存空间的占用，节省系统的硬件资源，且可以降低成本及减少传输和烧录数据耗费的时间。
请参阅图3及图4，本发明提供一种补偿表压缩方法的第二实施例，包括如下步骤：
步骤S1’、对原始补偿表进行DCT变换(离散余弦变换)，将原始补 偿表从空域转换到频域，且分成多个相同大小的子块；
步骤S2’、根据能量函数计算公式：E＝sum(D)+K计算每个子块DCT系数右下角区域的能量函数，其中，sum(D)为子块DCT系数右下角区域中每个点的绝对值之和，K为DC系数，E为每个子块DCT系数右下角区域的能量函数；
步骤S3’、根据量化步长计算公式：Q step＝aE+b得到每个子块对应的量化步长，其中，Q step为量化步长，a和b均为常数；
步骤S4’、每个子块根据其对应的量化步长分别量化，得到量化补偿表，采用编码算法对量化补偿表进行压缩。
具体的，还包括步骤S5’、通过解压压缩后的量化补偿表，再根据量化补偿表中的信息反量化每个子块，再反DCT变换每个子块即可重建原始补偿表。
具体的，每个子块的大小为8*8，即每个子块对应8行8列像素(总共64个像素)的补偿信息。
进一步的，每个子块DCT系数右下角区域的大小为4*4，即每个子块DCT系数右下角区域对应4行4列像素(总共16个像素)的补偿信息；DC系数为每个子块中最左上角的值。
具体的，所述步骤S3’中，量化步长计算公式可以根据视觉无损主观实验得到，从而无损压缩量化补偿表，其中，a和b是通过拟合视觉无损主观实验结果得到的参数。
需要说明的是，本发明通过视觉无损压缩补偿表，使人眼察觉不出原补偿表的质量损失，且极大增加压缩效率，和现有的补偿表压缩算法相比，本发明无需迭代即可获得最佳量化步长，减少对内存空间的占用，节省系统的硬件资源，且可以降低成本及减少传输和烧录数据耗费的时间。
综上所述，本发明的补偿表压缩方法通过视觉无损压缩补偿表，使人眼察觉不出原补偿表的质量损失，且极大增加压缩效率，和现有的补偿表压缩算法相比，本发明无需迭代即可获得最佳量化步长，减少对内存空间的占用，节省系统的硬件资源，且可以降低成本及减少传输和烧录数据耗费的时间。
以上所述，对于本领域的普通技术人员来说，可以根据本发明的技术方案和技术构思作出其他各种相应的改变和变形，而所有这些改变和变形都应属于本发明权利要求的保护范围。

Claims (13)

  1. 一种补偿表压缩方法，包括如下步骤：
    步骤S1、将原始补偿表在空域中分成多个相同大小的子块；
    步骤S2、对分成多个子块后的原始补偿表乘以sobel算子得到梯度矢量，分别计算每个字块的方差；
    步骤S3、根据量化步长计算公式：Q step＝aσ 2+b得到每个子块对应的量化步长，其中，Q step为量化步长，σ 2为方差，a和b均为常数；
    步骤S4、每个子块根据其对应的量化步长分别量化，得到量化补偿表，采用编码算法对量化补偿表进行压缩。
  2. 如权利要求1所述的补偿表压缩方法，还包括步骤S5、通过解压压缩后的量化补偿表，再根据量化补偿表中的信息反量化每个子块重建原始补偿表。
  3. 如权利要求1所述的补偿表压缩方法，其中，每个子块的大小为8*8。
  4. 如权利要求1所述的补偿表压缩方法，其中，所述步骤S2中通过sobel算子检测每个子块的边缘。
  5. 如权利要求1所述的补偿表压缩方法，其中，所述步骤S3中通过视觉无损主观实验得到量化步长计算公式，其中，a和b是通过拟合视觉无损主观实验结果得到的参数。
  6. 一种补偿表压缩方法，包括如下步骤：
    步骤S1’、对原始补偿表进行DCT变换，将原始补偿表从空域转换到频域，且分成多个相同大小的子块；
    步骤S2’、根据能量函数计算公式：E＝sum(D)+K计算每个子块DCT系数右下角区域的能量函数，其中，sum(D)为子块DCT系数右下角区域中每个点的绝对值之和，K为DC系数，E为每个子块DCT系数右下角区域的能量函数；
    步骤S3’、根据量化步长计算公式：Q step＝aE+b得到每个子块对应的量化步长，其中，Q step为量化步长，a和b均为常数；
    步骤S4’、每个子块根据其对应的量化步长分别量化，得到量化补偿表， 采用编码算法对量化补偿表进行压缩。
  7. 如权利要求6所述的补偿表压缩方法，还包括步骤S5’、通过解压压缩后的量化补偿表，再根据量化补偿表中的信息反量化及反DCT变换每个子块重建原始补偿表。
  8. 如权利要求6所述的补偿表压缩方法，其中，每个子块的大小为8*8。
  9. 如权利要求8所述的补偿表压缩方法，其中，每个子块DCT系数右下角区域的大小为4*4；DC系数为每个子块中最左上角的值。
  10. 如权利要求6所述的补偿表压缩方法，其中，所述步骤S3’中通过视觉无损主观实验得到量化步长计算公式，其中，a和b是通过拟合视觉无损主观实验结果得到的参数。
  11. 一种补偿表压缩方法，包括如下步骤：
    步骤S1’、对原始补偿表进行DCT变换，将原始补偿表从空域转换到频域，且分成多个相同大小的子块；
    步骤S2’、根据能量函数计算公式：E＝sum(D)+K计算每个子块DCT系数右下角区域的能量函数，其中，sum(D)为子块DCT系数右下角区域中每个点的绝对值之和，K为DC系数，E为每个子块DCT系数右下角区域的能量函数；
    步骤S3’、根据量化步长计算公式：Q step＝aE+b得到每个子块对应的量化步长，其中，Q step为量化步长，a和b均为常数；
    步骤S4’、每个子块根据其对应的量化步长分别量化，得到量化补偿表，采用编码算法对量化补偿表进行压缩；
    步骤S5’、通过解压压缩后的量化补偿表，再根据量化补偿表中的信息反量化及反DCT变换每个子块重建原始补偿表；
    其中，每个子块的大小为8*8。
  12. 如权利要求11所述的补偿表压缩方法，其中，每个子块DCT系数右下角区域的大小为4*4；DC系数为每个子块中最左上角的值。
  13. 如权利要求11所述的补偿表压缩方法，其中，所述步骤S3’中通过视觉无损主观实验得到量化步长计算公式，其中，a和b是通过拟合视觉无损主观实验结果得到的参数。
PCT/CN2019/071033 2018-12-04 2019-01-09 补偿表压缩方法 Ceased WO2020113766A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811475457.2A CN109324778B (zh) 2018-12-04 2018-12-04 补偿表压缩方法
CN201811475457.2 2018-12-04

Publications (1)

Publication Number Publication Date
WO2020113766A1 true WO2020113766A1 (zh) 2020-06-11

Family

ID=65256221

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/071033 Ceased WO2020113766A1 (zh) 2018-12-04 2019-01-09 补偿表压缩方法

Country Status (2)

Country Link
CN (1) CN109324778B (zh)
WO (1) WO2020113766A1 (zh)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110166783B (zh) 2019-06-03 2021-03-23 Tcl华星光电技术有限公司 补偿表压缩方法、显示器制造设备和具有存储功能的装置
CN110796995B (zh) * 2019-11-28 2021-09-03 Tcl华星光电技术有限公司 显示面板的补偿数据的处理方法及显示装置
CN111225216A (zh) * 2020-01-10 2020-06-02 Tcl华星光电技术有限公司 显示器补偿表压缩方法、装置、系统及显示器
CN111223438B (zh) * 2020-03-11 2022-11-04 Tcl华星光电技术有限公司 像素补偿表的压缩方法及装置
CN116229870B (zh) * 2023-05-10 2023-08-15 苏州华兴源创科技股份有限公司 补偿数据的压缩、解压缩方法及显示面板补偿方法

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2005032117A2 (en) * 2003-09-24 2005-04-07 Texas Instruments Incorporated 8x8 transform and quantization
CN108416725A (zh) * 2018-03-13 2018-08-17 中山大学 一种基于消除块效应的非对称jpeg隐写方法

Family Cites Families (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR0129573B1 (ko) * 1994-04-30 1998-04-10 배순훈 디지탈 화상 복호화를 위한 디씨(dc) 계수의 오차보상 방법
US6963668B2 (en) * 1998-11-13 2005-11-08 Lightsurf Technologies, Inc. Method and system for fast image correction
KR100539923B1 (ko) * 2003-02-10 2005-12-28 삼성전자주식회사 화상통화시 화자의 영상을 구분하여 차등적 부호화할 수있는 비디오 엔코더 및 이를 이용한 비디오신호의 압축방법
KR20080065820A (ko) * 2007-01-10 2008-07-15 삼성전자주식회사 디지털 멀티미디어 중계 신호처리장치 및 그 방법
CN100496127C (zh) * 2007-06-05 2009-06-03 南京大学 Mpeg-2到h.264码的快速转换方法
CN102256126A (zh) * 2011-07-14 2011-11-23 北京工业大学 混合图像的编码方法
CN103313056B (zh) * 2013-06-03 2016-08-24 南京邮电大学 一种基于图像融合和边缘Hash的子块修复方法
CN104202599B (zh) * 2014-09-23 2017-07-28 浙江工商大学 一种应用于hevc帧间编码的全零块检测方法
CN105700847B (zh) * 2016-03-25 2019-01-22 深圳市华星光电技术有限公司 Oled显示面板补偿数据的存储方法
CN106341576B (zh) * 2016-08-25 2020-07-03 深圳市华星光电技术有限公司 图像处理方法
CN106339196B (zh) * 2016-08-31 2019-03-15 深圳市华星光电技术有限公司 DeMura表的数据压缩、解压缩方法及Mura补偿方法
CN107068050A (zh) * 2017-06-09 2017-08-18 深圳市华星光电技术有限公司 Oled显示面板的补偿表存储方法
CN107294538B (zh) * 2017-06-09 2020-09-01 深圳市华星光电半导体显示技术有限公司 Oled显示装置的补偿表压缩方法及解压方法
CN107799065B (zh) * 2017-11-02 2019-11-26 深圳市华星光电半导体显示技术有限公司 Oled显示面板的灰阶补偿表的压缩方法
CN108008553A (zh) * 2017-11-22 2018-05-08 深圳市华星光电技术有限公司 一种Mura检测校正方法和系统
CN108172168B (zh) * 2017-12-22 2019-11-15 深圳市华星光电半导体显示技术有限公司 一种补偿表压缩方法

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2005032117A2 (en) * 2003-09-24 2005-04-07 Texas Instruments Incorporated 8x8 transform and quantization
CN108416725A (zh) * 2018-03-13 2018-08-17 中山大学 一种基于消除块效应的非对称jpeg隐写方法

Also Published As

Publication number Publication date
CN109324778B (zh) 2020-03-27
CN109324778A (zh) 2019-02-12

Similar Documents

Publication Publication Date Title
WO2020113766A1 (zh) 补偿表压缩方法
KR102118234B1 (ko) 프레임 버퍼 압축을 위한 혼합 모드
US9460527B2 (en) Pattern mode for frame buffer compression
CN111819848B (zh) 用于图像压缩和解压缩的方法和装置以及存储介质
US20140098854A1 (en) Lossless intra-prediction video coding
US11494946B2 (en) Data compression device and compression method configured to gradually adjust a quantization step size to obtain an optimal target quantization step size
US20200169742A1 (en) Single-channel inverse mapping for image/video processing
US20190320173A1 (en) Spatially adaptive quantization-aware deblocking filter
JP3105335B2 (ja) 画像の直交変換符号化による圧縮・伸張方法
CN103957426A (zh) 一种rgb565真彩色图像有损压缩及解压方法
US9153017B1 (en) System and method for optimized chroma subsampling
US10560703B2 (en) FLC-based image compression method and device
US12526418B2 (en) Image compression device and image compression method
US20250030883A1 (en) Video decoder, video encoder, and bitstream
Joshua et al. Comparison of DCT and DWT image compression
US12505505B2 (en) Super-resolution video processing method and system for effective video compression
GB2620228A (en) Image compression with inter-channel encoding of codewords
CN115695793A (zh) 一种jpeg图像压缩系统
CN107027027A (zh) 图像的编码、解码方法及装置、以及图像的编解码系统
US11854235B1 (en) Lossless integer compression scheme
CN117579847A (zh) 一种基于相似块差值编码的屏幕图像压缩方法
CN118715773A (zh) 图像编解码方法、装置、设备、系统、及存储介质
CN114598881A (zh) 一种基于图像分布的实时帧缓存压缩系统
Arjun et al. Verilog implementation of adaptive compression of RGB images at low bit rates
EP2605515B1 (en) Method and apparatus for discrete cosine transform/inverse discrete cosine transform

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19893418

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19893418

Country of ref document: EP

Kind code of ref document: A1