WO2023045168A1 - 一种图像处理方法、装置及相关组件 - Google Patents

一种图像处理方法、装置及相关组件 Download PDF

Info

Publication number
WO2023045168A1
WO2023045168A1 PCT/CN2021/143274 CN2021143274W WO2023045168A1 WO 2023045168 A1 WO2023045168 A1 WO 2023045168A1 CN 2021143274 W CN2021143274 W CN 2021143274W WO 2023045168 A1 WO2023045168 A1 WO 2023045168A1
Authority
WO
WIPO (PCT)
Prior art keywords
ddr
image block
image data
image
lines
Prior art date
Application number
PCT/CN2021/143274
Other languages
English (en)
French (fr)
Inventor
宋金凤
Original Assignee
苏州浪潮智能科技有限公司
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 苏州浪潮智能科技有限公司 filed Critical 苏州浪潮智能科技有限公司
Priority to US18/259,681 priority Critical patent/US11930198B2/en
Publication of WO2023045168A1 publication Critical patent/WO2023045168A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/42—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation
    • H04N19/423—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation characterised by memory arrangements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/41—Bandwidth or redundancy reduction
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
    • H04N19/124—Quantisation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
    • H04N19/13—Adaptive entropy coding, e.g. adaptive variable length coding [AVLC] or context adaptive binary arithmetic coding [CABAC]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
    • H04N19/136—Incoming video signal characteristics or properties
    • H04N19/14—Coding unit complexity, e.g. amount of activity or edge presence estimation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
    • H04N19/176—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/186—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being a colour or a chrominance component
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/60—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding
    • H04N19/625—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding using discrete cosine transform [DCT]

Definitions

  • the present application relates to the field of image processing, in particular to an image processing method, device and related components.
  • VGA Video Graphics Array, Video Graphics Array
  • DDR Double Data Rate, double-rate synchronous dynamic random access memory
  • the VGA stores the video source image data in the DDR in units of frames.
  • the JPEG compression module reads the video source image data from the DDR to process the entire frame of video source image data. RGB to YUV format conversion, and then perform image compression on the converted image data.
  • the reading and writing of the entire frame source image data increases the transmission delay, and the entire frame image needs to be divided into blocks in the JPEG compression module, which wastes system resources.
  • the purpose of the present application is to provide an image processing method, device, electronic equipment and computer-readable storage medium, which can reduce data transmission delay and reduce waste of system resources.
  • an image processing method including:
  • the preset number of lines is an integer multiple of the number of lines of an image block determined according to the compression condition, and the image block Any image block in the image block matrix determined based on the entire frame of image;
  • the process of judging whether the number of writing lines of the source image data in the DDR is a preset number of lines includes:
  • the image processing method before reading the source image data of the current image block from the DDR based on the first read address of the current image block, the number of rows and the number of columns, the image processing method further includes:
  • If the current image block is the image block in the last column of the image block matrix, determine whether the number of columns of the source image data in the DDR is an integer multiple of the number of columns of the image block;
  • the column edge processing includes performing a padding operation on the source image data of the last column in the DDR, so that the number of columns of the source image data is an integer multiple of the number of columns of the image block.
  • the image processing method before reading the source image data of the current image block from the DDR based on the first read address of the current image block, the number of rows and the number of columns, the image processing method further includes:
  • If the current image block is the image block in the last row of the image block matrix, determine whether the number of rows of the source image data in the DDR is an integer multiple of the number of rows of the image block;
  • the line edge processing includes performing a padding operation with the source image data of the last line in the DDR, so that the number of lines of the source image data in the DDR is equal to the number of lines of the image block Integer multiples of .
  • the compression processing includes RGB-to-YUV format conversion operations, DCT transformation operations, quantization operations, and entropy coding operations.
  • an image processing device including:
  • Judging module for judging whether the number of writing lines of the source image data in the DDR is a preset number of lines or stop writing the number of lines, if so, generating a first trigger signal; the preset number of lines is determined according to compression conditions Integer multiples of the number of rows of the image block, the image block being any image block in the image block matrix determined based on the entire frame of image;
  • a reading module configured to read the source image data of the current image block from the DDR based on the first read address of the current image block, the number of rows, and the number of columns when the first trigger signal is received;
  • a compression module configured to perform compression processing on the source image data of the current image block to obtain compressed data, and write the compressed data into the DDR.
  • an electronic device including:
  • a processor configured to implement the steps of the image processing method described in any one of the above when executing the computer program.
  • the present application also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it realizes any of the above-mentioned The steps of the image processing method.
  • the present application provides an image processing method. Whenever the source image data written in the DDR reaches the number of lines of the image block corresponding to the compression condition or reaches the number of lines to stop writing, the source image data in the DDR is read to form the current image. Blocks are compressed without waiting for the entire frame source image data to be written into the DDR before processing. Compared with the existing entire frame read and write scheme, the amount of data read and written each time in this application is less, thereby reducing the data transmission delay. At the same time, the application completes the division of the entire frame of images before compression, which reduces the waste of system resources.
  • the present application also provides an image processing device, an electronic device, and a computer-readable storage medium, which have the same beneficial effect as the above-mentioned image processing method.
  • Fig. 1 is a flow chart of the steps of an image processing method provided by the present application
  • FIG. 2 is a schematic structural diagram of an image processing system provided by the present application.
  • FIG. 3 is a schematic structural diagram of another image processing system provided by the present application.
  • the core of the present application is to provide an image processing method, device, electronic equipment and computer-readable storage medium, which can reduce data transmission delay and reduce waste of system resources.
  • Fig. 1 is the step flow chart of a kind of image processing method provided by the present application, and this image processing method comprises:
  • S101 Determine whether the number of writing lines of the source image data in the DDR is the preset number of lines or the number of stopping writing lines, if so, execute S102; the preset number of lines is an integer multiple of the number of lines of the image block determined according to the compression condition , the image block is any image block in the image block matrix determined based on the entire frame image;
  • the number of rows and columns of each image block can be determined based on the compression conditions, and then the number of rows and columns of the image block matrix can be determined based on the resolution of the entire frame image.
  • the size of each image block is 8 ⁇ 8, that is, the number of rows and columns of the image block is 8, and when the JPEG image data is in RGB888 format, the data of one image block is 8 ⁇ 8 ⁇ 4byte. It can be understood that, assuming that the resolution of the whole frame image is 640 ⁇ 480 and the size of the image block is 8 ⁇ 8, the size of the image block matrix is 80 ⁇ 60.
  • the number of rows and columns of the image block matrix will be adjusted accordingly. It can be understood that the number of rows and columns of the source image data stored in the DDR is greater than the number of rows and columns of the image block.
  • the source image data written into the DDR In order to improve the image data processing efficiency, it is necessary to monitor the source image data written into the DDR by the VGA. Whenever the source image data is filled with 8 lines, that is, the number of write lines of the source image data in the DDR is the line of the image block. When it is an integer multiple of the number, the source image data in the DDR can be read and processed, thereby avoiding the disadvantages caused by storing the entire frame of image and then processing it.
  • the process of judging whether the number of writing lines of the source image data in the DDR is a preset number of lines includes:
  • AXI Advanced eXtensible Interface, advanced extensible interface
  • S102 Read the source image data of the current image block from the DDR based on the read first address, the number of rows and the number of columns of the current image block;
  • the number of writing lines of the source image data in the DDR reaches the preset number of lines, it is first determined whether the current image block is the image block of the last row in the image block matrix, if not, according to the reading of the current image block.
  • the first address and the number of rows and columns of the image block read the source image data of the current image block from the DDR. In this step, the block processing of the whole frame image is completed, so as to improve the image data processing efficiency.
  • the read first address of the first image block in the first row and first column in the image block matrix is the first address of the entire frame image
  • the read first addresses of other image blocks are based on the read first address of the first image block , the resolution of the whole frame image and the block sequence number can be obtained.
  • the preset number of rows refers to an integer multiple of the number of rows of an image block. Taking an 8 ⁇ 8 image block as an example, the preset number of rows is 8, 16, 24, 32..., and so on.
  • the first reading is to write 8 rows
  • the source image data read by each image block in the first row of the image block matrix is the source image data of 1-8 rows in DDR
  • the second reading is to write 16 rows
  • the source image data read by each image block of the second row in the image block matrix is the source image data of rows 9-16 in the DDR, and so on.
  • S103 Perform compression processing on the source image data of the current image block to obtain compressed data, and write the compressed data into the DDR.
  • the source image data of the current image block is compressed to obtain compressed data, and the compressed data is written back to the DDR. It can be understood that the source image data and the compressed data are partitioned and stored in the DDR.
  • the source image data is compressed, including RGB to YUV format conversion operation, DCT transformation operation, quantization operation and entropy coding operation.
  • the read source image data of the current image block is converted from RGB to YUV format, and then based on DCT transformation, the data is quantized according to the quantization table, and finally the compressed data is obtained through entropy coding.
  • DCT Discrete Cosine Transform, discrete cosine transform
  • the source image data can be input to the read FIFO (First In First Out, first-in-first-out queue), because the read The size of source image data is 8 ⁇ 8, so the depth of reading FIFO does not need to be very large, which saves part of FIFO resources.
  • the compressed data can also be output to the DDR by writing the FIFO.
  • the source image data written in the DDR reaches the number of lines of the image block corresponding to the compression condition or reaches the number of lines to stop writing
  • the source image data in the DDR is read to form the current image block for compression processing, without waiting for the whole frame source image data to be written into DDR before processing.
  • the amount of data read and written each time in this application is less, thereby reducing data transmission delay.
  • this application The block of the whole frame image is completed before compression, which reduces the waste of system resources.
  • the image processing method before reading the source image data of the current image block from the DDR based on the read head address, the number of rows and the number of columns of the current image block, the image processing method also includes:
  • If the current image block is the image block of the last column in the image block matrix, determine whether the column number of the source image data in the DDR is an integer multiple of the column number of the image block;
  • the column edge processing includes performing a padding operation with the source image data of the last column in the DDR, so that the number of columns of the source image data is an integer multiple of the number of columns of the image block.
  • the image processing method before reading the source image data of the current image block from the DDR based on the read head address, the number of rows and the number of columns of the current image block, the image processing method also includes:
  • If the current image block is the image block of the last row in the image block matrix, determine whether the number of rows of the source image data in the DDR is an integer multiple of the number of rows of the image block;
  • the row edge processing includes performing a padding operation with the source image data of the last row in the DDR, so that the number of rows of the source image data in the DDR is an integer multiple of the row number of the image block.
  • the number of rows and columns of the source image data in DDR may not be an integer multiple of the number of rows and columns of the image block. Therefore, in order to maintain the visual continuity of the entire frame of image, it is necessary to The source image data of the image blocks in the last column and the image blocks in the last row are subjected to edge processing.
  • the source image data in the DDR is filled with the preset number of rows, that is, the reading condition of a certain row of image blocks in the image block matrix corresponding to the preset number of rows is satisfied, considering that the columns of the source image data in the DDR The number may not be an integer multiple of the number of columns in the image block, so the size of the read source image data does not meet the size of the image block.
  • perform column edge processing on the source image data in DDR and pass the last column in DDR
  • the source image data of the source image data or the edge part is filled by padding with 0, so that the read source image data of the last image block of the row meets the compression conditions.
  • the source image data of the entire frame of image is written into the DDR
  • the source image data of each image block in the last row in the image block matrix should be read at this time, considering that the number of rows of the source image data in the DDR is not In the case of an integer multiple of the number of image block rows, in order to ensure the visual continuity of the entire frame image, when it is determined that the number of source image data in the DDR is not an integer multiple of the image block row number, the source image data in the DDR is processed. Edge processing, the edge part is filled by the source image data of the last row in the DDR or by filling 0, so that the source image data of the image block of the last row read meets the compression conditions.
  • the following uses an 8 ⁇ 8 image block as an example to illustrate the source image data reading solution based on the AXI protocol in this application.
  • Step 201 System initialization, by detecting AWADDR to judge whether the source image data written by VGA to DDR has been greater than 8 lines, if not, wait until the data written into DDR is enough for the image block to read.
  • Step 202 If the source image data written into the DDR is greater than 8 rows, determine the first read address of the image block in the first row and first column of the image block matrix.
  • the AXI master reads the source image data of the last image block in the first row of the source image data in the DDR.
  • Step 204 Judging whether AWADDR satisfies the reading condition of the source image data of the next row of the image block matrix, if not, wait;
  • Step 206 judge whether the current line is the last line of the image block matrix, if this line is not the last line of the image block matrix, then the AXI main control reads the source image data of the first image block of this line, and determines the next line of this line in turn Read the first address of an image block, judge whether the current image block is the last image block of this line, if not, the AXI master reads the source image of the next image block of this line from DDR in turn Data, if this image block is the last image block of this row, judge whether the number of columns of the source image data in DDR is a multiple of 8, if it is a multiple of 8, column edge processing is not required, if it is not a multiple of 8, Then perform column edge processing, and after the column edge processing is completed, the AXI main control reads the source image data of the last image block of this row. Enter step 204, and read the source image data of the next row of the image block matrix sequentially.
  • Step 207 If the next row is the last row of the image block matrix, then judge whether the row number of the source image data in the DDR is a multiple of 8, if not a multiple of 8, then perform row edge processing, after processing, the AXI master reads Get the source image data of the first image block in the last row, and then determine the read first address of the next image block. Judging whether the next image block is the last image block of the last row, if not the last image block of the last row, the AXI main control reads the source image data of the current image block.
  • Block data is used as a unit for source image data capture, and subsequent format conversion and JPEG image compression. It saves the capture time of source image data, greatly reduces the time delay of data transmission, and the capture method based on AXI protocol replaces the block unit before JPEG image compression, saving system resources.
  • FIG. 2 is a schematic structural diagram of an image processing device provided by the present application.
  • the image processing device includes:
  • Judging module 1 for judging whether the number of writing lines of the source image data in the DDR is a preset number of lines or stop writing the number of lines, if so, generating a first trigger signal;
  • the preset number of lines is an image determined according to compression conditions an integer multiple of the number of rows of the block, and the image block is any image block in the image block matrix determined based on the entire frame of image;
  • the reading module 2 is used to read the source image data of the current image block from the DDR based on the read head address, the number of rows and the number of columns of the current image block when the first trigger signal is received;
  • the compression module 3 is used for compressing the source image data of the current image block to obtain compressed data, and writing the compressed data into the DDR.
  • the source image data written in the DDR reaches the number of lines of the image block corresponding to the compression condition or reaches the number of lines to stop writing
  • the source image data in the DDR is read to form the current image block for compression processing, without waiting for the whole frame source image data to be written into DDR before processing.
  • the amount of data read and written each time in this application is less, thereby reducing data transmission delay.
  • this application The block of the whole frame image is completed before compression, which reduces the waste of system resources.
  • the process of judging whether the number of writing lines of the source image data in the DDR is a preset number of lines includes:
  • the image processing device further includes:
  • Edge processing module 4 is used for if current image block is the image block of last row in the image block matrix, judges whether the column number of source image data in DDR is the integral multiple of the column number of image block, if not, to source image in DDR Data is column edge processed.
  • the column edge processing includes performing a padding operation with the source image data of the last column in the DDR, so that the number of columns of the source image data is an integer multiple of the number of columns of the image block.
  • the edge processing module 4 is also used to determine whether the number of rows of the source image data in the DDR is an integer of the number of rows of the image block if the current image block is the last row of the image block in the image block matrix times, if not, perform line edge processing on the source image data in DDR.
  • the row edge processing includes performing a padding operation with the source image data of the last row in the DDR, so that the number of rows of the source image data in the DDR is an integer multiple of the row number of the image block.
  • the compression processing includes an RGB-to-YUV format conversion operation, a DCT transformation operation, a quantization operation, and an entropy encoding operation.
  • the image processing device also includes:
  • the block address generation module 5 is used to generate the first address of each image block in the image block matrix in turn when the first trigger signal is received;
  • Read FIFO 6 which is used to transmit the source image data obtained by reading module 2;
  • an electronic device including:
  • the processor is configured to implement the steps of the image processing method described in any one of the above embodiments when executing the computer program.
  • An electronic device provided by the present application has the same beneficial effects as the above image processing method.
  • the present application also provides a computer-readable storage medium, on which a computer program is stored.
  • a computer program is stored.
  • the steps of the image processing method described in any one of the above embodiments are implemented. .
  • a computer-readable storage medium provided by the present application has the same beneficial effects as the above image processing method.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • Discrete Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Compression Or Coding Systems Of Tv Signals (AREA)
  • Compression Of Band Width Or Redundancy In Fax (AREA)

Abstract

本申请公开了一种图像处理方法，包括：判断DDR中的源图像数据的写入行数是否为预设行数或停止写入行数；预设行数为根据压缩条件确定的图像块的行数的整数倍，图像块为基于整帧图像确定的图像块矩阵中的任一图像块；若是，基于当前图像块的读取首地址、行数及列数从DDR中读取当前图像块的源图像数据；对当前图像块的源图像数据进行压缩处理得到压缩数据，并将压缩数据写入DDR。本申请能够降低数据传输延迟，减少系统资源的浪费。本申请还公开了一种图像处理装置、电子设备及计算机可读存储介质，具有以上有益效果。

Description

一种图像处理方法、装置及相关组件
本申请要求在2021年9月22日提交中国专利局、申请号为202111103935.9、发明名称为“一种图像处理方法、装置及相关组件”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及图像处理领域，特别涉及一种图像处理方法、装置及相关组件。
背景技术
在居民生活、工业生产、航空航天领域，数字图像处理技术越来越重要。由于图像的数据量大，因此，需要采用图像压缩编码技术以满足高质量下使图像以较小的比特率传输的目的。本地视频图像经过VGA(Video Graphics Array，视频图形阵列)处理以后存入VGA独有的内存空间DDR(Double Data Rate，双倍速率同步动态随机存储器)中。为了实现视频图像的远程传输，JPEG压缩模块从DDR中读取视频源图像数据进行图像压缩，将压缩后的压缩图像数据存入DDR中，以太网控制器从DDR中读取压缩图像数据，通过以太网传输到远程终端。
现有技术中，VGA以帧为单位将视频源图像数据存入DDR中，整帧视频源图像数据存储完成后，JPEG压缩模块从DDR中读取视频源图像数据，进行整帧视频源图像数据的RGB转YUV格式转换，然后对格式转换后的图像数据进行图像压缩。在传输大分辨率的视频图像时，整帧源图像数据的读写增大了传输延迟，且需要在JPEG压缩模块内实现整帧图像的分块，浪费系统资源。
因此，如何提供一种解决上述技术问题的方案是本领域技术人员目前需要解决的问题。
发明内容
本申请的目的是提供一种图像处理方法、装置、电子设备及计算机可读存储介质，能够降低数据传输延迟，减少系统资源的浪费。
为解决上述技术问题，本申请提供了一种图像处理方法，包括：
判断DDR中的源图像数据的写入行数是否为预设行数或停止写入行数；所述预设行数为根据压缩条件确定的图像块的行数的整数倍，所述图像块为基于整帧图像确定的图像块矩阵中的任一图像块；
若是，基于当前图像块的读取首地址、所述行数及列数从所述DDR中读取所述当前图像块的源图像数据；
对所述当前图像块的源图像数据进行压缩处理得到压缩数据，并将所述压缩数据写入所述DDR。
可选的，所述判断DDR中的源图像数据的写入行数是否为预设行数的过程包括：
检测DDR中的当前写地址是否为预设地址；
若是，判定所述DDR中的源图像数据的写入行数为预设行数；
若否，判定所述DDR中的源图像数据的写入行数不为所述预设行数。
可选的，所述基于当前图像块的读取首地址、所述行数及列数从所述DDR中读取所述当前图像块的源图像数据之前，该图像处理方法还包括：
若所述当前图像块为所述图像块矩阵中最后一列的所述图像块，判断所述DDR中所述源图像数据的列数是否为所述图像块的列数的整数倍；
若否，对所述DDR中所述源图像数据进行列边缘处理。
可选的，所述列边缘处理包括以所述DDR中最后一列的所述源图像数据进行补齐操作，以使所述源图像数据的列数为所述图像块的列数的整数倍。
可选的，所述基于当前图像块的读取首地址、所述行数及列数从所述DDR中读取所述当前图像块的源图像数据之前，该图像处理方法还包括：
若所述当前图像块为所述图像块矩阵中最后一行的所述图像块，判断所述DDR中所述源图像数据的行数是否为所述图像块的行数的整数倍；
若否，对所述DDR中所述源图像数据进行行边缘处理。
可选的，所述行边缘处理包括以所述DDR中最后一行的所述源图像数据进行补齐操作，以使所述DDR中所述源图像数据的行数为所述图像块的行数的整数倍。
可选的，所述压缩处理包括RGB转YUV格式转换操作、DCT变换操作、量化操作及熵编码操作。
为解决上述技术问题，本申请还提供了一种图像处理装置，包括：
判断模块，用于判断DDR中的源图像数据的写入行数是否为预设行数或停止写入行数，若是，生成第一触发信号；所述预设行数为根据压缩条件确定的图像块的行数的整数倍，所述图像块为基于整帧图像确定的图像块矩阵中的任一图像块；
读取模块，用于当接收到所述第一触发信号，基于当前图像块的读取首地址、所述行数及列数从所述DDR中读取所述当前图像块的源图像数据；
压缩模块，用于对所述当前图像块的源图像数据进行压缩处理得到压缩数据，并将所述压缩数据写入所述DDR。
为解决上述技术问题，本申请还提供了一种电子设备，包括：
存储器，用于存储计算机程序；
处理器，用于执行所述计算机程序时实现如上文任意一项所述的图像处理方法的步骤。
为解决上述技术问题，本申请还提供了一种计算机可读存储介质，所述计算机可读存储介质上存储有计算机程序，所述计算机程序被处理器执行时实现如上文任意一项所述的图像处理方法的步骤。
本申请提供了一种图像处理方法，每当DDR中写入的源图像数据达到压缩条件对应的图像块的行数或者达到停止写入行数时，读取DDR中的源图像数据构成当前图像块进行压缩处理，而无需等待整帧源图像数据均写入DDR后再处理，相对于现有的整帧读写方案，本申请每次读写的数据量要少，从而降低数据传输延迟，同时本申请在压缩之前完成整帧图像的分块，减少了系统资源的浪费。本申请还提供了一种图像处理装置、电子设备及计算机可读存储介质，具有和上述图像处理方法相同的有益效果。
附图说明
为了更清楚地说明本申请实施例，下面将对实施例中所需要使用的附图做简单的介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例， 对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为本申请所提供的一种图像处理方法的步骤流程图；
图2为本申请所提供的一种图像处理系统的结构示意图；
图3为本申请所提供的另一种图像处理系统的结构示意图。
具体实施方式
本申请的核心是提供一种图像处理方法、装置、电子设备及计算机可读存储介质，能够降低数据传输延迟，减少系统资源的浪费。
为使本申请实施例的目的、技术方案和优点更加清楚，下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
请参照图1，图1为本申请所提供的一种图像处理方法的步骤流程图，该图像处理方法包括：
S101：判断DDR中的源图像数据的写入行数是否为预设行数或停止写入行数，若是，执行S102；预设行数为根据压缩条件确定的图像块的行数的整数倍，图像块为基于整帧图像确定的图像块矩阵中的任一图像块；
具体的，基于压缩条件可以确定每一个图像块的行数和列数，再基于整帧图像的分辨率就可以确定图像块矩阵的行数和列数，举例说明，在基于DCT的JPEG压缩过程中，每一图像块的尺寸都是8×8，即图像块的行数和列数均为8，JPEG图像数据是RGB888格式时，一个图像块的数据是8×8×4byte。可以理解的是，假设整帧图像的分辨率是640×480，图像块的尺寸为8×8，则图像块矩阵的尺寸为80×60。不同的压缩条件对图像块的尺寸有不同的要求，相应的，图像块矩阵的行列数也会相应调整。可以理解的是，DDR中存储的源图像数据的行数和列数大于图像块的行数和列数。
为提高图像数据处理效率，需要对VGA写入DDR中的源图像数据进行监控，每当源图像数据写满8行，即每当DDR中的源图像数据的写入行数为 图像块的行数的整数倍时，即可以读取并处理DDR中的源图像数据，从而避免整帧图像存储后再进行处理所带来的弊端。当然，考虑到一帧图像的源图像数据在DDR中写完时，其行数可能并不是图像块的行数的整数倍，因此，当判定整帧图像的所有源图像数据均写入DDR后，即判定满足停止写入行数。作为一种可选的实施例，判断DDR中的源图像数据的写入行数是否为预设行数的过程包括：
检测DDR中的当前写地址是否为预设地址；
若是，判定DDR中的源图像数据的写入行数为预设行数；
若否，判定DDR中的源图像数据的写入行数不为预设行数。
具体的，通过AXI(Advanced eXtensible Interface，高级可扩展接口)主控判断DDR中AWADDR写入地址是否为预设地址来判断DDR中的源图像数据的写入行数是否为预设行数。
S102：基于当前图像块的读取首地址、行数及列数从DDR中读取当前图像块的源图像数据；
具体的，当DDR中的源图像数据的写入行数达到预设行数后，首先确定当前图像块是否为图像块矩阵中最后一行的图像块，如果不是，则根据当前图像块的读取首地址以及图像块的行数、列数从DDR中读取当前图像块的源图像数据。本步骤中完成了对整帧图像的分块处理，以便提高图像数据处理效率。可以理解的是，图像块矩阵中首行首列第一个图像块的读取首地址为整帧图像的首地址，其他图像块的读取首地址基于第一个图像块的读取首地址、整帧图像的分辨率以及块序列号即可得到。
可以理解的是，预设行数是指图像块的行数的整数倍，以8×8的图像块为例，那么预设行数为8、16、24、32……，以此类推，第一次读取是写满8行，图像块矩阵中第一行的各图像块读取的源图像数据即DDR中1-8行的源图像数据，第二次读取是写满16行，图像块矩阵中第二行的各图像块读取的源图像数据即DDR中9-16行的源图像数据，以此类推。
S103：对当前图像块的源图像数据进行压缩处理得到压缩数据，并将压缩数据写入DDR。
具体的，当获取到当前图像块的源图像数据后，对当前图像块的源图像数据进行压缩处理，得到压缩数据，并将压缩数据写回DDR。可以理解的是，源图像数据和压缩数据在DDR中分区存储。
具体的，在基于DCT(Discrete Cosine Transform，离散余弦变换)的JPEG压缩过程，对源图像数据进行压缩，包括RGB转YUV格式转换操作、DCT变换操作、量化操作及熵编码操作。具体的，将读取到的当前图像块的源图像数据进行RGB转YUV格式转换，然后基于DCT变换，将数据依据量化表进行量化，最后通过熵编码得到压缩数据。可以理解的是，本申请在对源图像数据进行JPEG压缩之前完成了整帧图像的分块，可以节约JPEG压缩资源。
进一步的，为保证数据处理的完整性，当读取到当前图像块的源图像数据后，可将源图像数据输入到读FIFO(First In First Out，先入先出队列)，由于读取到的源图像数据都是8×8大小的，所以读FIFO的深度不需要很大，节省了一部分FIFO资源。相应的，基于JPEG压缩处理完得到压缩数据后，也可以将压缩数据通过写FIFO输出给DDR。
可见，本实施例中，每当DDR中写入的源图像数据达到压缩条件对应的图像块的行数或者达到停止写入行数时，读取DDR中的源图像数据构成当前图像块进行压缩处理，而无需等待整帧源图像数据均写入DDR后再处理，相对于现有的整帧读写方案，本申请每次读写的数据量要少，从而降低数据传输延迟，同时本申请在压缩之前完成整帧图像的分块，减少了系统资源的浪费。
在上述实施例的基础上：
作为一种可选的实施例，基于当前图像块的读取首地址、行数及列数从DDR中读取当前图像块的源图像数据之前，该图像处理方法还包括：
若当前图像块为图像块矩阵中最后一列的图像块，判断DDR中源图像数据的列数是否为图像块的列数的整数倍；
若否，对DDR中源图像数据进行列边缘处理。
作为一种可选的实施例，列边缘处理包括以DDR中最后一列的源图像数据进行补齐操作，以使源图像数据的列数为图像块的列数的整数倍。
作为一种可选的实施例，基于当前图像块的读取首地址、行数及列数从DDR中读取当前图像块的源图像数据之前，该图像处理方法还包括：
若当前图像块为图像块矩阵中最后一行的图像块，判断DDR中源图像数据的行数是否为图像块的行数的整数倍；
若否，对DDR中源图像数据进行行边缘处理。
作为一种可选的实施例，行边缘处理包括以DDR中最后一行的源图像数据进行补齐操作，以使DDR中源图像数据的行数为图像块的行数的整数倍。
可以理解的是，DDR中的源图像数据的行数和列数可能并不是图像块的行数和列数的整数倍，因此，为了维持整帧图像的视觉连续性，需要对图像块矩阵中最后一列的图像块以及最后一行的图像块的源图像数据进行边缘处理。
具体的，当DDR中源图像数据写满预设行数，即满足与该预设行数对应的图像块矩阵中某一行图像块的读取条件，考虑到由于DDR中的源图像数据的列数可能不是图像块的列数的整数倍，因此，读取到的源图像数据的大小不满足图像块的尺寸，此时，对DDR中的源图像数据进行列边缘处理，通过DDR中最后一列的源图像数据或通过补0的方式补齐边缘部分，以使读取到的本行最后一个图像块的源图像数据满足压缩条件。
相应的，若整帧图像的源图像数据都写入DDR后，此时该读取图像块矩阵中最后一行的各图像块的源图像数据，考虑到存在DDR中的源图像数据的行数不是图像块行数的整数倍的情况，为保证整帧图像的视觉连续性，当判定DDR中的源图像数据的行数不是图像块行数的整数倍时，对DDR中的源图像数据进行行边缘处理，通过DDR中最后一行的源图像数据或通过补0的方式补齐边缘部分，以使读取的最后一行的图像块的源图像数据满足压缩条件。
下面以8×8的图像块为例，对本申请中基于AXI协议的源图像数据读取方案进行说明。
步骤201：系统初始化，通过检测AWADDR判断VGA写入DDR的源图像数据是否已经大于8行，如果没有大于8行，则等待，直到写入DDR的数据足够图像块读取。
步骤202：如果写入DDR的源图像数据大于8行，确定图像块矩阵的首行首列图像块的读取首地址。
步骤203：AXI主控读取图像块矩阵首行的源图像数据，首先读取图像块矩阵首行首列块的源图像数据，即m＝0，n＝0，按照从左到右的原则，确定图像块矩阵首行的下一个图像块的读取首地址，即m＝0，n++。判断下一个图像块是否是首行的最后一个图像块，如果不是首行的最后一个图像块，AXI主控按照该图像块的读取首地址从DDR中读取源图像数据，如果是首行的最后一个图像块，判断DDR中源图像数据的列数是否是8的整数倍，如果源图像数据的列数是8的倍数，则不需要对源图像数据进行列边缘处理，直接读取即可，如果不是，则需要对源图像数据进行列边缘处理，列边缘处理完后，AXI主控读取DDR中的源图像数据首行的最后一个图像块的源图像数据。
步骤204：判断AWADDR是否满足图像块矩阵下一行的源图像数据的读取条件，不满足则等待；
步骤205：如果满足读取条件，则产生图像块矩阵的下一行的第一图像块的读取首地址，即m++，n＝0。
步骤206：判断当前行是否为图像块矩阵的最后一行，如果此行不是图像块矩阵的最后一行，则AXI主控读取此行的第一图像块的源图像数据，依次确定此行的下一个图像块的读取首地址，判断当前图像块是否是此行的最后一个图像块，如果不是最后一个图像块，则AXI主控依次从DDR中读取此行的下一图像块的源图像数据，如果此图像块是此行的最后一个图像块，判断DDR中源图像数据的列数是否是8的倍数，如果是8的倍数，则不需要进行列边缘处理，如不是8的倍数，则进行列边缘处理，列边缘处理完毕后，AXI主控读取此行的最后一图像块的源图像数据。进入步骤204，依次读取图像块矩阵的下一行的源图像数据。
步骤207：如果下一行是图像块矩阵的最后一行，则判断DDR中源图像数据的行数是否是8的倍数，如果不是8的倍数，则进行行边缘处理，处理完毕后，AXI主控读取末行第一个图像块的源图像数据，然后确定下一个图像块的读取首地址。判断下一个图像块是否是末行的最后一个图像块，如果不是末行的最后一个图像块，AXI主控读取当前图像块的源图像数据。如果是末行的最后一个图像块，判断DDR中源图像数据的列数是否是8的倍数， 如果此图像块是此行的最后一个图像块，判断DDR中源图像数据的列数是否是8的倍数，如果是8的倍数，则不需要进行列边缘处理，如不是8的倍数，则进行列边缘处理，列边缘处理完毕后，AXI主控读取此行的最后一图像块的源图像数据，一帧图像完成捕获。
综上所述，本申请不需要等待VGA模块存入DDR中的源图像数据整帧完成，在存入的源图像数据大于一个图像块的数据时，从DDR中读取源图像数据，以图像块数据为单位进行源图像数据捕获，以及后续的格式转换以及JPEG图像压缩。节省了源图像数据的捕获时间，极大的降低了数据传输的时间延迟，并且基于AXI协议的捕获方法代替了JPEG图像压缩前的分块单元，节省了系统资源。
请参照图2，图2为本申请所提供的一种图像处理装置的结构示意图，该图像处理装置包括：
判断模块1，用于判断DDR中的源图像数据的写入行数是否为预设行数或停止写入行数，若是，生成第一触发信号；预设行数为根据压缩条件确定的图像块的行数的整数倍，图像块为基于整帧图像确定的图像块矩阵中的任一图像块；
读取模块2，用于当接收到第一触发信号，基于当前图像块的读取首地址、行数及列数从DDR中读取当前图像块的源图像数据；
压缩模块3，用于对当前图像块的源图像数据进行压缩处理得到压缩数据，并将压缩数据写入DDR。
可见，本实施例中，每当DDR中写入的源图像数据达到压缩条件对应的图像块的行数或者达到停止写入行数时，读取DDR中的源图像数据构成当前图像块进行压缩处理，而无需等待整帧源图像数据均写入DDR后再处理，相对于现有的整帧读写方案，本申请每次读写的数据量要少，从而降低数据传输延迟，同时本申请在压缩之前完成整帧图像的分块，减少了系统资源的浪费。
作为一种可选的实施例，判断DDR中的源图像数据的写入行数是否为预设行数的过程包括：
检测DDR中的当前写地址是否为预设地址；
若是，判定DDR中的源图像数据的写入行数为预设行数；
若否，判定DDR中的源图像数据的写入行数不为预设行数。
作为一种可选的实施例，该图像处理装置还包括：
边缘处理模块4，用于若当前图像块为图像块矩阵中最后一列的图像块，判断DDR中源图像数据的列数是否为图像块的列数的整数倍，若否，对DDR中源图像数据进行列边缘处理。
作为一种可选的实施例，列边缘处理包括以DDR中最后一列的源图像数据进行补齐操作，以使源图像数据的列数为图像块的列数的整数倍。
作为一种可选的实施例，边缘处理模块4，还用于若当前图像块为图像块矩阵中最后一行的图像块，判断DDR中源图像数据的行数是否为图像块的行数的整数倍，若否，对DDR中源图像数据进行行边缘处理。
作为一种可选的实施例，行边缘处理包括以DDR中最后一行的源图像数据进行补齐操作，以使DDR中源图像数据的行数为图像块的行数的整数倍。
作为一种可选的实施例，压缩处理包括RGB转YUV格式转换操作、DCT变换操作、量化操作及熵编码操作。
作为一种优选的实施例，参照图3所示，该图像处理装置还包括：
块地址产生模块5，用于当接收到第一触发信号，依次生成图像块矩阵中各个图像块的首地址；
读FIFO 6，用于传输读取模块2获取到的源图像数据；
写FIFO 7，用于传输压缩模块3处理后的压缩数据。
另一方面，本申请还提供了一种电子设备，包括：
存储器，用于存储计算机程序；
处理器，用于执行计算机程序时实现如上文任意一个实施例所描述的图像处理方法的步骤。
本申请所提供的一种电子设备具有和上述图像处理方法相同的有益效果。
对于本申请所提供的一种电子设备的介绍请参照上述实施例，本申请在此不再赘述。
另一方面，本申请还提供了一种计算机可读存储介质，计算机可读存储介质上存储有计算机程序，计算机程序被处理器执行时实现如上文任意一个实施例所描述的图像处理方法的步骤。
本申请所提供的一种计算机可读存储介质具有和上述图像处理方法相同的有益效果。
对于本申请所提供的一种计算机可读存储介质的介绍请参照上述实施例，本申请在此不再赘述。
还需要说明的是，在本说明书中，诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的状况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。
对所公开的实施例的上述说明，使本领域专业技术人员能够实现或使用本申请。对这些实施例的多种修改对本领域的专业技术人员来说将是显而易见的，本文中所定义的一般原理可以在不脱离本申请的精神或范围的情况下，在其他实施例中实现。因此，本申请将不会被限制于本文所示的这些实施例，而是要符合与本文所公开的原理和新颖特点相一致的最宽的范围。

Claims (12)

  1. 一种图像处理方法，其特征在于，包括：
    判断DDR中的源图像数据的写入行数是否为预设行数或停止写入行数；所述预设行数为根据压缩条件确定的图像块的行数的整数倍，所述图像块为基于整帧图像确定的图像块矩阵中的任一图像块；
    若是，基于当前图像块的读取首地址、所述行数及列数从所述DDR中读取所述当前图像块的源图像数据；
    对所述当前图像块的源图像数据进行压缩处理得到压缩数据，并将所述压缩数据写入所述DDR。
  2. 根据权利要求1所述的图像处理方法，其特征在于，所述判断DDR中的源图像数据的写入行数是否为预设行数的过程包括：
    检测DDR中的当前写地址是否为预设地址；
    若是，判定所述DDR中的源图像数据的写入行数为预设行数；
    若否，判定所述DDR中的源图像数据的写入行数不为所述预设行数。
  3. 根据权利要求1所述的图像处理方法，其特征在于，所述基于当前图像块的读取首地址、所述行数及列数从所述DDR中读取所述当前图像块的源图像数据之前，该图像处理方法还包括：
    若所述当前图像块为所述图像块矩阵中最后一列的所述图像块，判断所述DDR中所述源图像数据的列数是否为所述图像块的列数的整数倍；
    若否，对所述DDR中所述源图像数据进行列边缘处理。
  4. 根据权利要求3所述的图像处理方法，其特征在于，所述列边缘处理包括以所述DDR中最后一列的所述源图像数据进行补齐操作，以使所述源图像数据的列数为所述图像块的列数的整数倍。
  5. 根据权利要求1所述的图像处理方法，其特征在于，所述基于当前图像块的读取首地址、所述行数及列数从所述DDR中读取所述当前图像块的源图像数据之前，该图像处理方法还包括：
    若所述当前图像块为所述图像块矩阵中最后一行的所述图像块，判断所述DDR中所述源图像数据的行数是否为所述图像块的行数的整数倍；
    若否，对所述DDR中所述源图像数据进行行边缘处理。
  6. 根据权利要求5所述的图像处理方法，其特征在于，所述行边缘处理包括以所述DDR中最后一行的所述源图像数据进行补齐操作，以使所述DDR中所述源图像数据的行数为所述图像块的行数的整数倍。
  7. 根据权利要求1-6任意一项所述的图像处理方法，其特征在于，所述压缩处理包括RGB转YUV格式转换操作、DCT变换操作、量化操作及熵编码操作。
  8. 根据权利要求1所述的图像处理方法，其特征在于，所述停止写入行数为判定所述整帧图像的所有所述源图像数据均写入所述DDR时对应的写入行数。
  9. 一种图像处理装置，其特征在于，包括：
    判断模块，用于判断DDR中的源图像数据的写入行数是否为预设行数或停止写入行数，若是，生成第一触发信号；所述预设行数为根据压缩条件确定的图像块的行数的整数倍，所述图像块为基于整帧图像确定的图像块矩阵中的任一图像块；
    读取模块，用于当接收到所述第一触发信号，基于当前图像块的读取首地址、所述行数及列数从所述DDR中读取所述当前图像块的源图像数据；
    压缩模块，用于对所述当前图像块的源图像数据进行压缩处理得到压缩数据，并将所述压缩数据写入所述DDR。
  10. 根据权利要求9所述的图像处理装置，其特征在于，所述停止写入行数为判定所述整帧图像的所有所述源图像数据均写入所述DDR时对应的写入行数。
  11. 一种电子设备，其特征在于，包括：
    存储器，用于存储计算机程序；
    处理器，用于执行所述计算机程序时实现如权利要求1-7任意一项所述的图像处理方法的步骤。
  12. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质上存储有计算机程序，所述计算机程序被处理器执行时实现如权利要求1-8任意一项所述的图像处理方法的步骤。
PCT/CN2021/143274 2021-09-22 2021-12-30 一种图像处理方法、装置及相关组件 WO2023045168A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US18/259,681 US11930198B2 (en) 2021-09-22 2021-12-30 Image processing method and apparatus, and related assembly

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111103935.9A CN113573072B (zh) 2021-09-22 2021-09-22 一种图像处理方法、装置及相关组件
CN202111103935.9 2021-09-22

Publications (1)

Publication Number Publication Date
WO2023045168A1 true WO2023045168A1 (zh) 2023-03-30

Family

ID=78173829

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/143274 WO2023045168A1 (zh) 2021-09-22 2021-12-30 一种图像处理方法、装置及相关组件

Country Status (3)

Country Link
US (1) US11930198B2 (zh)
CN (1) CN113573072B (zh)
WO (1) WO2023045168A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113573072B (zh) * 2021-09-22 2021-12-17 苏州浪潮智能科技有限公司 一种图像处理方法、装置及相关组件
CN114219700B (zh) * 2022-02-23 2022-06-03 广州匠芯创科技有限公司 一种图像处理方法、系统、设备及介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6473527B1 (en) * 1999-06-01 2002-10-29 Mustek Systems Inc. Module and method for interfacing analog/digital converting means and JPEG compression means
CN105260983A (zh) * 2015-11-11 2016-01-20 西安电子科技大学 基于HyperX平台的5/3小波变换方法
CN113068040A (zh) * 2021-06-02 2021-07-02 苏州浪潮智能科技有限公司 一种图像压缩方法、装置、电子设备及可读存储介质
CN113421184A (zh) * 2021-06-21 2021-09-21 上海爱观视觉科技有限公司 一种基于fpga的实时图像处理方法
CN113573072A (zh) * 2021-09-22 2021-10-29 苏州浪潮智能科技有限公司 一种图像处理方法、装置及相关组件

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8473673B2 (en) * 2005-06-24 2013-06-25 Hewlett-Packard Development Company, L.P. Memory controller based (DE)compression
JP2008146581A (ja) * 2006-12-13 2008-06-26 Texas Instr Japan Ltd メモリバス共有システム
WO2015025237A2 (en) * 2013-07-26 2015-02-26 Riversilica Technologies Pvt Ltd A method and system for transcoding a digital video
CN105120286B (zh) * 2015-06-26 2019-01-18 福州瑞芯微电子股份有限公司 一种视频解码参考帧取数方法及设备
US10277904B2 (en) * 2015-08-28 2019-04-30 Qualcomm Incorporated Channel line buffer data packing scheme for video codecs
CN106528456B (zh) * 2016-11-16 2020-02-21 湖南国科微电子股份有限公司 一种提高视频解码显示中系统总线效率的方法及系统
US11449577B2 (en) * 2019-11-20 2022-09-20 Micron Technology, Inc. Methods and apparatus for performing video processing matrix operations within a memory array

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6473527B1 (en) * 1999-06-01 2002-10-29 Mustek Systems Inc. Module and method for interfacing analog/digital converting means and JPEG compression means
CN105260983A (zh) * 2015-11-11 2016-01-20 西安电子科技大学 基于HyperX平台的5/3小波变换方法
CN113068040A (zh) * 2021-06-02 2021-07-02 苏州浪潮智能科技有限公司 一种图像压缩方法、装置、电子设备及可读存储介质
CN113421184A (zh) * 2021-06-21 2021-09-21 上海爱观视觉科技有限公司 一种基于fpga的实时图像处理方法
CN113573072A (zh) * 2021-09-22 2021-10-29 苏州浪潮智能科技有限公司 一种图像处理方法、装置及相关组件

Also Published As

Publication number Publication date
CN113573072A (zh) 2021-10-29
CN113573072B (zh) 2021-12-17
US11930198B2 (en) 2024-03-12
US20230396786A1 (en) 2023-12-07

Similar Documents

Publication Publication Date Title
WO2023045168A1 (zh) 一种图像处理方法、装置及相关组件
CN101729840A (zh) 一种利用视频图像特征序列的存储处理方法
WO2022161227A1 (zh) 图像处理方法、装置、图像处理芯片和电子设备
CN115460414B (zh) 一种基板管理控制芯片的视频压缩方法、系统及相关组件
TWI637631B (zh) 影像處理裝置、影視子系統與影視處理電路
WO2023134128A1 (zh) 一种视频压缩处理方法、装置及介质
WO2021248936A1 (zh) 一种视频存储方法、装置、soc系统、介质
WO2022062719A1 (zh) 视频处理方法、计算机可读存储介质及电子设备
CN103248797A (zh) 一种基于fpga的视频分辨率增强方法及模块
WO2022095797A1 (zh) 图像的压缩方法、装置、智能终端及计算机可读存储介质
WO2023103336A1 (zh) 一种视频数据传输方法、视频数据解码方法及相关装置
WO2023197507A1 (zh) 视频数据处理方法、系统、装置及计算机可读存储介质
CN115209145A (zh) 一种视频压缩方法、系统、装置及可读存储介质
CN106412588B (zh) 影像帧处理方法
CN211909012U (zh) 拼接图像处理设备
WO2023193701A1 (zh) 图像编码方法及装置
WO2024119812A1 (zh) 图片处理方法、系统、装置、图像压缩装置及存储介质
CN108881923B (zh) 减少jpeg编解码行缓冲容量的方法
CN103730097B (zh) 超高分辨率图像的显示方法与系统
CN104581175B (zh) 影像处理装置与方法
CN117061789B (zh) 一种视频传输框架、方法、设备和存储介质
US20220130010A1 (en) Image Processing Method and Device
CN114327338B (zh) 一种超高分图像显示系统及图像显示方法
WO2024045026A1 (zh) 一种显示方法、电子设备、显示设备、传屏器及介质
US9179156B2 (en) Memory controller for video analytics and encoding

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: 21958278

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE