WO2021164324A1 - 一种图像垂直滤波处理的方法及芯片 - Google Patents

一种图像垂直滤波处理的方法及芯片 Download PDF

Info

Publication number
WO2021164324A1
WO2021164324A1 PCT/CN2020/126767 CN2020126767W WO2021164324A1 WO 2021164324 A1 WO2021164324 A1 WO 2021164324A1 CN 2020126767 W CN2020126767 W CN 2020126767W WO 2021164324 A1 WO2021164324 A1 WO 2021164324A1
Authority
WO
WIPO (PCT)
Prior art keywords
image data
line
filter
data
read
Prior art date
Application number
PCT/CN2020/126767
Other languages
English (en)
French (fr)
Inventor
王洪良
张德闪
牟奇
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 US17/780,248 priority Critical patent/US11659173B2/en
Publication of WO2021164324A1 publication Critical patent/WO2021164324A1/zh

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/80Camera processing pipelines; Components thereof
    • H04N23/81Camera processing pipelines; Components thereof for suppressing or minimising disturbance in the image signal generation
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102Methods 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
    • H04N19/117Filters, e.g. for pre-processing or post-processing
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102Methods 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
    • H04N19/124Quantisation
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N25/00Circuitry of solid-state image sensors [SSIS]; Control thereof
    • H04N25/70SSIS architectures; Circuits associated therewith
    • H04N25/71Charge-coupled device [CCD] sensors; Charge-transfer registers specially adapted for CCD sensors
    • H04N25/75Circuitry for providing, modifying or processing image signals from the pixel array
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/42Methods 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

Definitions

  • This application relates to the field of image vertical filtering processing, and in particular to a method and chip for image vertical filtering processing.
  • the main function of the image signal processing (Image Signal Processing, ISP) chip is to process the signal output by the front-end image sensor in real time, such as noise reduction and color correction.
  • the ISP chip also has a series of functions such as accelerating the focusing speed, digital anti-shake, and extracting image feature values to realize face recognition.
  • ISP chip is favored by major manufacturers because of its speed of processing video images, that is, the ISP chip can implement traditional software image processing algorithms through hardware chips, greatly improving the processing speed of video images. By processing the video image through the ISP chip, the user will not feel any delay at all.
  • the output and input of the recursive filter in the ISP chip have a coupling feedback relationship.
  • the recursive filter is divided into horizontal direction filtering and vertical direction filtering, and the principles and architectures of the two implementations are different.
  • the filtered output image data needs to be fed back point by point.
  • image data is input line by point.
  • the recursive filter in the horizontal direction can be designed.
  • the purpose of this application is to provide a method and chip for image vertical filtering processing, which are used to increase the speed of the chip for vertical filtering processing of the image and reduce the cost of the chip.
  • the present application provides a method for image vertical filtering processing, which includes:
  • Cache a preset number of lines of image data, and read the image data of the line to be read; wherein the line to be read is the first line of the image data that is not cached;
  • the image data is updated to obtain the image data after the vertical filtering process.
  • the invoking filter performs vertical filtering processing on the image data of the preset number of lines and the image data of the line to be read to obtain the first filtered data, including:
  • the image data of the front line is generated according to the image data of the preset number of caches, and the image data of the post line is determined according to the image data of the preset number of caches and the image data of the line to be read ;
  • the updating the image data according to the first filtering data includes:
  • the line image data to be filtered is updated to filter data of the line image data to be filtered.
  • the invoking the filter to perform vertical filtering processing on the image data of the preset number of lines and the image data of the line to be read to obtain the second filtered data includes:
  • the updating the image data according to the second filtering data to obtain the image data after the vertical filtering process includes:
  • update the image data of the line to be filtered to the second filter data update the line to be filtered to the next line of the line to be filtered, and return to execute the preset line according to the cache
  • the method before calling a filter to perform vertical filtering processing on the image data of the preset number of lines and the image data of the line to be read to obtain the first filtered data, the method further includes:
  • the method further includes:
  • control beat of the filter is modified according to the control beat modification command.
  • This application also provides a chip for image vertical filtering processing, the chip including:
  • the cache module is used to cache a preset number of lines of image data and read the image data of the line to be read; wherein the line to be read is the first line of the image data that is not cached;
  • the first calling module is configured to call a filter to perform vertical filtering processing on the image data of the preset number of lines and the image data of the line to be read to obtain the first filtered data;
  • a judging module configured to update the image data according to the first filter data, and judge whether the row to be read is the last row of the image data;
  • the update module is used to update the row to be read to the next row of the row to be read when the row to be read is not the last row of the image data, and return to execute the cache preset The image data of the number of rows, and the step of reading the image data of the row to be read;
  • the second calling module is used to call the filter to perform vertical processing on the image data of the preset number of rows and the image data of the row to be read when the row to be read is the last row of the image data. Filter processing to obtain second filter data, and update the image data according to the second filter data to obtain the image data after the vertical filter processing.
  • the first calling module includes:
  • An obtaining unit configured to obtain the filter order, and determine the pre-line number and the post-line number of the image data of the line to be filtered according to the filter order;
  • the first determining unit is configured to generate image data of the pre-line according to the cached image data of the preset number of lines, and according to the cached image data of the preset number of lines and the image data of the line to be read The image data determines the image data of the rear row;
  • the first calling unit is configured to call a filter to perform vertical filtering processing on the image data of the front line and the image data of the back line to obtain the filtered data of the image data of the line to be filtered as the first filter data.
  • the judgment module includes:
  • the first update unit is configured to update the line image data to be filtered to the filtered data of the line image data to be filtered.
  • the method for image vertical filtering processing includes: buffering a preset number of lines of image data, and reading the image data of the line to be read; wherein the line to be read is the first line of the image data that is not cached; Call the filter to perform vertical filtering processing on the image data of the preset number of lines and the image data of the line to be read to obtain the first filter data; update the image data according to the first filter data, and determine whether the line to be read is an image The last line of data; if not, update the line to be read to the next line of the line to be read, and return to the step of caching the image data of the preset number of lines and reading the image data of the line to be read; if yes , The filter is called to perform vertical filtering processing on the image data of the preset number of lines and the image data of the line to be read to obtain the second filtering data, and the image data is updated according to the second filtering data to obtain the vertical filtering process Image data.
  • the technical solution provided by this application performs vertical filtering processing on the image data of the preset number of lines and the image data of the line to be read by calling the filter to obtain the first filtering data, and then updating the image data according to the first filtering data
  • the algorithm principle of the digital recursive filter and the characteristics of the image chip the point-by-point feedback is changed to the line-by-line feedback, the algorithm implementation architecture is optimized, the development efficiency is improved, and the maintenance time is reduced, which greatly improves the image of the chip.
  • the speed of vertical filtering processing reduces chip cost.
  • the present application also provides a chip for image vertical filtering processing, which has the above-mentioned beneficial effects, and will not be repeated here.
  • FIG. 1 is a flowchart of an image vertical filtering method provided by an embodiment of the application
  • FIG. 2 is a flowchart of an actual performance of S102 in the method for image vertical filtering provided in FIG. 1;
  • FIG. 3 is a flowchart of an actual expression of S105 in the method for image vertical filtering processing provided in FIG. 1;
  • FIG. 4 is a structural diagram of an image vertical filtering processing chip provided by an embodiment of the application.
  • FIG. 5 is a structural diagram of another image vertical filtering processing chip provided by an embodiment of the application.
  • the core of this application is to provide a method and chip for image vertical filtering processing, which are used to increase the speed of the chip for vertical filtering processing of the image and reduce the cost of the chip.
  • FIG. 1 is a flowchart of a method for image vertical filtering processing according to an embodiment of the application.
  • S101 Cache the image data of a preset number of lines, and read the image data of the line to be read;
  • the line to be read mentioned here is the first line in the image data that is not cached.
  • the chip Before performing vertical filtering on the image data, the chip will cache a preset number of lines of image data, and then read the uncached Image data, because the image data is read line by point, in order to improve the processing speed, when the image data is subjected to vertical filtering processing, you can read one point and process one column, of course, you can also first read the line to be read.
  • the vertical filtering process is performed after all the image data is read, which is not specifically limited in this application.
  • S102 Invoke a filter to perform vertical filtering processing on the image data of the preset number of lines and the image data of the line to be read to obtain first filtered data;
  • the image data of the preset number of lines mentioned here may specifically be the entire image data, that is, the entire image is cached.
  • Fig. 2 is a flowchart of an actual expression of S102 in the method for image vertical filtering provided in Fig. 1, which specifically includes the following steps:
  • S201 Obtain the filter order, and determine the pre-line number and the post-line number of the image data of the line to be filtered according to the filter order;
  • the filtered data obtained is the filter data of the image data of the fifth line
  • the line to be filtered mentioned here is the image data of the fifth line.
  • Rows, the preceding rows of the rows to be filtered are rows 1-4, and the rear rows of the rows to be filtered are rows 6-9.
  • S202 Generate the image data of the front line according to the image data of the preset number of cached lines, and determine the image data of the post line according to the image data of the preset number of cached lines and the image data of the line to be read;
  • the line to be filtered is the boundary line or adjacent to the boundary line, there is a problem that the image data is insufficient and the filtering cannot be completed.
  • the post line is line 2-5. While the front line is empty, this application can generate the image data of the front line according to the image data of the preset number of cached lines, that is, all the image data of the front line is replaced with the image data of the first line.
  • S203 Invoke the filter to perform vertical filtering processing on the image data of the preceding line and the image data of the succeeding line, to obtain the filter data of the image data of the line to be filtered as the first filter data.
  • the parameters of the filter may also be obtained, and the parameters of the filter may be obtained. Quantization processing to adjust the accuracy of the filter.
  • control beats of the filter can be modified according to the control beat modification command, so as to Realize the adjustment of the filter control shots, and then facilitate the digital chip engineer to debug the image quality filter.
  • S103 Update the image data according to the first filter data, and determine whether the row to be read is the last row of the image data;
  • step S104 If not, go to step S104; if yes, go to step S105;
  • the image data mentioned here is updated according to the first filtering data, which may specifically be:
  • the line image data to be filtered is updated to the filtering data of the line image data to be filtered.
  • the vertical filtering process requires vertical processing of the image data, and the required line buffer is extremely large. Take the 9th order filter as an example, if you want to perform vertical filtering on the image data, you need to cache 9 lines of image data at the same time. On the basis, if the point-by-point feedback is performed again, the chip cost and the occupied area will be greatly increased. Therefore, this application performs vertical filtering processing on the image data of the preset number of lines and the image data of the line to be read by calling the filter.
  • the first filter data is obtained, and the first filter data is the filter data of the line to be filtered, that is, the recursive vertical filter processing of the image data is realized by way of line-by-line feedback, which greatly reduces the chip cost and occupied area.
  • step S101 is returned to continue the calculation.
  • a filter data of the line image data to be filtered is returned to continue the calculation.
  • S105 Call the filter to perform vertical filtering processing on the image data of the preset number of lines and the image data of the line to be read to obtain the second filtering data, and update the image data according to the second filtering data to obtain the vertical filtering process.
  • Image data
  • the filter When the line to be read is the last line of the image data, it indicates that there is no data to read at this time.
  • the filter is called to perform vertical filtering processing on the image data of the preset number of lines and the image data of the line to be read, and the first line is obtained. Secondly filter the data, and update the image data according to the second filter data to obtain the image data after the vertical filtering process, and complete the vertical filtering process of the image this time.
  • a method for image vertical filtering processing provided by the present application is to perform vertical filtering processing on a preset number of lines of image data and image data of the line to be read by calling a filter to obtain first filtered data, Then the image data is updated according to the first filter data.
  • the point-by-point feedback is changed to the line-by-line feedback, which optimizes the algorithm implementation architecture, improves development efficiency, and reduces
  • the maintenance time greatly improves the speed of the chip's vertical filtering of the image and reduces the chip cost.
  • the described calling filter performs vertical filtering processing on the image data of the preset number of lines and the image data of the line to be read to obtain the second filtered data, which is described below with reference to FIG. 3 .
  • FIG. 3 is a flowchart of an actual performance of S105 in the method for image vertical filtering provided in FIG. 1.
  • S301 Obtain the filter order, and determine the pre-line number and the post-line number of the image data of the line to be filtered according to the filter order;
  • S302 Generate the image data of the front line according to the image data of the preset number of cached lines, and determine the image data of the post line according to the image data of the preset number of cached and the last line of the image data;
  • S303 Invoke the filter to perform vertical filtering processing on the image data of the preceding line and the image data of the succeeding line to obtain the filtered data of the image data of the line to be filtered as the second filtering data.
  • the line to be filtered is the boundary line or adjacent to the boundary line, there is a problem that the image data is insufficient and the filtering cannot be completed.
  • the 9th order filter Take the 9th order filter as an example, when the line to be filtered is the last line, the front line is the 5th-2th line from the bottom , And the rear row is empty. At this time, this application can generate the image data of the rear row according to the cached preset number of image data, that is, replace all the image data of the rear row with the image data of the last row.
  • step S105 the image data mentioned in step S105 is updated according to the second filtering data to obtain the image data after the vertical filtering process, which may specifically be:
  • step S302 If not, update the image data of the line to be filtered as the second filter data, and update the line to be filtered as the next line of the line to be filtered, and return to step S302;
  • FIG. 4 is a structural diagram of an image vertical filtering processing chip provided by an embodiment of the application.
  • the chip may include:
  • the cache module 100 is configured to cache a preset number of lines of image data and read the image data of the line to be read; wherein the line to be read is the first line of the image data that is not cached;
  • the first calling module 200 is configured to call a filter to perform vertical filtering processing on the image data of the preset number of lines and the image data of the line to be read to obtain the first filtered data;
  • the judging module 300 is used to update the image data according to the first filter data, and to judge whether the row to be read is the last row of the image data;
  • the update module 400 is used to update the line to be read to the next line of the line to be read when the line to be read is not the last line of the image data, and return to the cache module 100 to cache the preset number of lines of image data, And the step of reading the image data of the row to be read;
  • the second calling module 500 is used to call the filter to perform vertical filtering processing on the image data of the preset number of rows and the image data of the row to be read when the line to be read is the last line of the image data to obtain the second filtered data,
  • the image data is updated according to the second filter data to obtain the image data processed by the vertical filter.
  • the first invoking module 200 may include:
  • the acquiring unit is used to acquire the filter order, and determine the pre-line number and the post-line number of the image data of the line to be filtered according to the filter order;
  • the first determining unit is used to generate the image data of the pre-line according to the image data of the preset number of caches, and determine the image of the post-line according to the image data of the preset number of caches and the image data of the line to be read data;
  • the first calling unit is used for calling the filter to perform vertical filtering processing on the image data of the front line and the image data of the back line to obtain the filtered data of the image data of the line to be filtered as the first filtering data.
  • the judgment module 300 may include:
  • the first update unit is used to update the line image data to be filtered into the filter data of the line image data to be filtered.
  • the second invoking module 500 may include:
  • the second determining unit is used to obtain the filter order, and determine the pre-line number and the post-line number of the image data of the line to be filtered according to the filter order;
  • the third determining unit is used to generate the image data of the front line according to the image data of the preset number of cached, and determine the image data of the post line according to the image data of the preset number of caches and the last line of the image data;
  • the calling unit is used for calling the filter to perform vertical filtering processing on the image data of the front line and the image data of the back line to obtain the filtered data of the image data of the line to be filtered as the second filtering data.
  • the second invoking module 500 may include:
  • the judging unit is used to judge whether the line to be filtered is the last line of the image data
  • the second update unit is used to update the image data of the line to be filtered to the second filter data when the line to be filtered is not the last line of the image data, and update the line to be filtered to the next line of the line to be filtered, and return to the first line 3.
  • the determining unit executes the step of generating image data of the pre-line according to the image data of the preset number of cached rows, and determining the image data of the post-line according to the image data of the preset number of cached and the last line of the image data;
  • the third update unit is configured to update the image data of the last line to the second filter data when the line to be filtered is the last line of the image data, and output the updated image data.
  • the chip may further include:
  • the quantization processing module is used to obtain the parameters of the filter and perform quantization processing on the parameters of the filter.
  • the chip may further include:
  • the modification module is used to modify the control beat of the filter according to the control beat modification command when the control beat modification command is received.
  • FIG. 5 is a structural diagram of another image vertical filtering processing chip provided by an embodiment of the application.
  • the chip may include a synchronization module, a data flow control module, and a calculation module, where:
  • the synchronization module is used to process the ready/need handshake signal.
  • the ready signal indicates that the upper-level module tells the lower-level module that the data is ready and is highly effective; the lower-level need signal tells the upper-level module that it can receive data and is highly effective.
  • a data transfer is completed, that is, a handshake.
  • each algorithm will have a large amount of control and calculation. If each processing carries a handshake signal, the design will become extremely complicated and difficult to maintain.
  • this application treats this part as a
  • the module processes separately, determines the parameter information according to the width and height of the image, converts the ready/need handshake signal into a valid signal, a horizontal count (h_cnt) signal and a vertical count (v_cnt) signal, and uses the above 3 signals to control the data flow And calculation can greatly reduce the complexity of design and control;
  • Data flow control module used to complete data buffer control, data expansion control, data selection output control and other functions
  • the calculation module is used to complete the multiplication and addition calculation of image data and coefficients.

Abstract

本申请公开了一种图像垂直滤波处理的方法,包括:缓存并读取图像数据;调用滤波器对图像数据进行垂直滤波处理,得到第一滤波数据;根据第一滤波数据对图像数据进行更新,并判断待读取行是否为图像数据的最后一行;若否,则将待读取行更新为待读取行的下一行,并返回执行缓存图像数据步骤;若是,则确定垂直滤波处理后的图像数据。本申请依据数字递归滤波器的算法原理以及图像芯片的特性,将逐点反馈改为逐行反馈,优化了算法实现架构,提高了开发效率,并减少了维护时间,极大的提高了芯片对图像进行垂直滤波处理的速度,降低芯片成本。本申请同时还提供了一种图像垂直滤波处理的芯片,具有上述有益效果。

Description

一种图像垂直滤波处理的方法及芯片
本申请要求于2020年02月21日提交中国国家知识产权局,申请号为202010108903.7,发明名称为“一种图像垂直滤波处理的方法及芯片”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
本申请涉及图像垂直滤波处理领域,特别涉及一种图像垂直滤波处理的方法及芯片。
背景技术
图像信号处理(Image Signal Processing,ISP)芯片主要作用是对前端图像传感器输出的信号实时处理,如降噪、颜色校正等。同时,ISP芯片还具有加快对焦速度,数字防抖,提取图像特征值实现人脸识别等一系列功能。目前,市面上几乎所有的智能手机以及监控芯片的处理器中都集成了ISP算法单元,少数厂商则采用定制的独立ISP芯片。ISP芯片之所以能得到各大厂商的青睐得益于其处理视频图像的速度,即ISP芯片可以把传统的软件图像处理算法通过硬件芯片实现,大大提高视频图像的处理速度。通过ISP芯片处理视频图像,用户完全不会感觉到有任何延迟。
目前,在ISP芯片中递归滤波器的输出与输入具有耦合反馈关系,递归滤波器分为水平方向滤波与垂直方向滤波,两者实现的原理与架构均有所不同。依据数字递归滤波器的原理,滤波输出的图像数据需要逐点反馈。在视频图像芯片处理领域,图像数据均是逐行逐点输入,对于水平方向的递归滤波,只要设计出具有反馈功能的水平累窗模块便能设计出水平方向上的递归滤波 器。但是,对于垂直方向的递归滤波,若是实现逐点反馈,会大大增加设计难度、降低处理速度(芯片中均具有反压功能),降低芯片可靠性,增加芯片成本。
因此,如何提高芯片对图像进行垂直滤波处理的速度,降低芯片成本是本领域技术人员目前需要解决的技术问题。
发明内容
本申请的目的是提供一种图像垂直滤波处理的方法及芯片,用于提高芯片对图像进行垂直滤波处理的速度,降低芯片成本。
为解决上述技术问题,本申请提供一种图像垂直滤波处理的方法,该方法包括:
缓存预设行数的图像数据,并读取待读取行的图像数据;其中,所述待读取行为所述图像数据中未被缓存的第一行;
调用滤波器对所述预设行数的图像数据和所述待读取行的图像数据进行垂直滤波处理,得到第一滤波数据;
根据所述第一滤波数据对所述图像数据进行更新,并判断所述待读取行是否为所述图像数据的最后一行;
若否,则将所述待读取行更新为所述待读取行的下一行,并返回执行缓存所述预设行数的图像数据,并读取待读取行的图像数据的步骤;
若是,则调用所述滤波器对所述预设行数的图像数据和所述待读取行的图像数据进行垂直滤波处理,得到第二滤波数据,并根据所述第二滤波数据对所述图像数据进行更新,得到垂直滤波处理后的所述图像数据。
可选的,所述调用滤波器对所述预设行数的图像数据和所述待读取行的图像数据进行垂直滤波处理,得到第一滤波数据,包括:
获取滤波器阶数,并根据所述滤波器阶数确定待滤波行图像数据的前置行数及后置行数;
根据缓存的所述预设行数的图像数据生成前置行的图像数据,并根据缓存的所述预设行数的图像数据及所述待读取行的图像数据确定后置行的图像数据;
调用滤波器对所述前置行的图像数据及所述后置行的图像数据进行垂直滤波处理,得到所述待滤波行图像数据的滤波数据作为所述第一滤波数据。
可选的,所述根据所述第一滤波数据对所述图像数据进行更新,包括:
将所述待滤波行图像数据更新为所述待滤波行图像数据的滤波数据。
可选的,所述调用所述滤波器对所述预设行数的图像数据和所述待读取行的图像数据进行垂直滤波处理,得到第二滤波数据,包括:
获取滤波器阶数,并根据所述滤波器阶数确定待滤波行图像数据的前置行数及后置行数;
根据缓存的所述预设行数的图像数据生成前置行的图像数据,并根据缓存的所述预设行数的图像数据及所述图像数据的最后一行确定后置行的图像数据;
调用滤波器对所述前置行的图像数据及所述后置行的图像数据进行垂直滤波处理,得到所述待滤波行图像数据的滤波数据作为所述第二滤波数据。
可选的,所述根据所述第二滤波数据对所述图像数据进行更新,得到垂直滤波处理后的所述图像数据,包括:
判断所述待滤波行是否为所述图像数据的最后一行;
若否,则将所述待滤波行的图像数据更新为所述第二滤波数据,并将所述待滤波行更新为所述待滤波行的下一行,返回执行根据缓存的所述预设行 数的图像数据生成所述前置行的图像数据,并根据缓存的所述预设行数的图像数据及所述图像数据的最后一行确定所述后置行的图像数据的步骤;
若是,则将最后一行的图像数据更新为所述第二滤波数据,并输出更新后的所述图像数据。
可选的,在调用滤波器对所述预设行数的图像数据和所述待读取行的图像数据进行垂直滤波处理,得到第一滤波数据之前,还包括:
获取所述滤波器的参数,并对所述滤波器的参数进行量化处理。
可选的,在获取所述滤波器的参数,并对所述滤波器的参数进行量化处理之后,还包括:
当接收到控制拍数修改命令时,根据所述控制拍数修改命令对所述滤波器的控制拍数进行修改。
本申请还提供一种图像垂直滤波处理的芯片,该芯片包括:
缓存模块,用于缓存预设行数的图像数据,并读取待读取行的图像数据;其中,所述待读取行为所述图像数据中未被缓存的第一行;
第一调用模块,用于调用滤波器对所述预设行数的图像数据和所述待读取行的图像数据进行垂直滤波处理,得到第一滤波数据;
判断模块,用于根据所述第一滤波数据对所述图像数据进行更新,并判断所述待读取行是否为所述图像数据的最后一行;
更新模块,用于当所述待读取行不为所述图像数据的最后一行时,将所述待读取行更新为所述待读取行的下一行,并返回执行缓存所述预设行数的图像数据,并读取待读取行的图像数据的步骤;
第二调用模块,用于当所述待读取行为所述图像数据的最后一行时,调用所述滤波器对所述预设行数的图像数据和所述待读取行的图像数据进行垂 直滤波处理,得到第二滤波数据,并根据所述第二滤波数据对所述图像数据进行更新,得到垂直滤波处理后的所述图像数据。
可选的,所述第一调用模块包括:
获取单元,用于获取滤波器阶数,并根据所述滤波器阶数确定待滤波行图像数据的前置行数及后置行数;
第一确定单元,用于根据缓存的所述预设行数的图像数据生成所述前置行的图像数据,并根据缓存的所述预设行数的图像数据及所述待读取行的图像数据确定所述后置行的图像数据;
第一调用单元,用于调用滤波器对所述前置行的图像数据及所述后置行的图像数据进行垂直滤波处理,得到所述待滤波行图像数据的滤波数据作为所述第一滤波数据。
可选的,所述判断模块包括:
第一更新单元,用于将所述待滤波行图像数据更新为所述待滤波行图像数据的滤波数据。
本申请所提供图像垂直滤波处理的方法,包括:缓存预设行数的图像数据,并读取待读取行的图像数据;其中,待读取行为图像数据中未被缓存的第一行;调用滤波器对预设行数的图像数据和待读取行的图像数据进行垂直滤波处理,得到第一滤波数据;根据第一滤波数据对图像数据进行更新,并判断待读取行是否为图像数据的最后一行;若否,则将待读取行更新为待读取行的下一行,并返回执行缓存预设行数的图像数据,并读取待读取行的图像数据的步骤;若是,则调用滤波器对预设行数的图像数据和待读取行的图像数据进行垂直滤波处理,得到第二滤波数据,并根据第二滤波数据对图像数据进行更新,得到垂直滤波处理后的图像数据。
本申请所提供的技术方案,通过调用滤波器对预设行数的图像数据和待读取行的图像数据进行垂直滤波处理,得到第一滤波数据,然后根据第一滤波数据对图像数据进行更新,依据数字递归滤波器的算法原理以及图像芯片的特性,将逐点反馈改为逐行反馈,优化了算法实现架构,提高了开发效率,并减少了维护时间,极大的提高了芯片对图像进行垂直滤波处理的速度,降低芯片成本。本申请同时还提供了一种图像垂直滤波处理的芯片,具有上述有益效果,在此不再赘述。
附图说明
为了更清楚地说明本申请实施例或现有技术中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本申请的实施例,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,还可以根据提供的附图获得其他的附图。
图1为本申请实施例所提供的一种图像垂直滤波处理的方法的流程图;
图2为图1所提供的一种图像垂直滤波处理的方法中S102的一种实际表现方式的流程图;
图3为图1所提供的一种图像垂直滤波处理的方法中S105的一种实际表现方式的流程图;
图4为本申请实施例所提供的一种图像垂直滤波处理的芯片的结构图;
图5为本申请实施例所提供的另一种图像垂直滤波处理的芯片的结构图。
具体实施方式
本申请的核心是提供一种图像垂直滤波处理的方法及芯片,用于提高芯片对图像进行垂直滤波处理的速度,降低芯片成本。
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。
目前,在数字芯片领域图像处理算法大多都是选用非递归滤波器,该类型的滤波器没有递归反馈,数据流容易控制,便于设计与实现。但是如果达到同等效果,非递归滤波器所需的阶数是递归滤波器阶数的5~10倍,尤其是处理垂直方向上的滤波,需要的行缓存就会成倍增加,这就意味着芯片成本的大大增加。此外,目前在该领域,大多数在设计与实现滤波器时架构不够优化,尤其是在垂直方向上,本身需要行缓存,当面对递归滤波器这种具有反馈信号的设计需求时,就会变得特别复杂,维护成本也会直线上升。因此,本申请提供了一种图像垂直滤波处理的方法,用于解决上述问题。
请参考图1,图1为本申请实施例所提供的一种图像垂直滤波处理的方法的流程图。
其具体包括如下步骤:
S101:缓存预设行数的图像数据,并读取待读取行的图像数据;
这里提到的待读取行即为图像数据中未被缓存的第一行,在对图像数据进行垂直滤波处理之前,芯片会先缓存预设行数的图像数据,然后读取未被缓存的图像数据,由于图像数据是按行逐点读取的,因此为提高处理速度,在对图像数据进行垂直滤波处理的时候,可以读取一个点处理一列,当然也可先将待读取行的图像数据全部读取之后再进行垂直滤波处理,本申请对此不做具体限定。
S102:调用滤波器对预设行数的图像数据和待读取行的图像数据进行垂直滤波处理,得到第一滤波数据;
可选的,这里提到的预设行数的图像数据,其具体可以为整张图像数据,即将整张图片全部缓存。
优选的,为降低图像数据所占用的缓存,这里提到的调用滤波器对预设行数的图像数据和待读取行的图像数据进行垂直滤波处理,得到第一滤波数据,其具体也可以为如图2所示的步骤,下面请参考图2,图2为图1所提供的一种图像垂直滤波处理的方法中S102的一种实际表现方式的流程图,其具体包括以下步骤:
S201:获取滤波器阶数,并根据滤波器阶数确定待滤波行图像数据的前置行数及后置行数;
以9阶滤波器为例,当9阶滤波器的输入行为1-9行的图像数据时,得到的滤波数据为第5行图像数据的滤波数据,这里提到的待滤波行即为第5行,待滤波行的前置行即为第1-4行,待滤波行的后置行即为第6-9行。
S202:根据缓存的预设行数的图像数据生成前置行的图像数据,并根据缓存的预设行数的图像数据及待读取行的图像数据确定后置行的图像数据;
当待滤波行为边界行或邻近边界行时,存在着图像数据不足,无法完成滤波的问题,以9阶滤波器为例,当待滤波行为第1行时,后置行为第2-5行,而前置行为空,此时本申请可以根据缓存的预设行数的图像数据生成前置行的图像数据,即将前置行的图像数据全部用第一行的图像数据进行替代。
S203:调用滤波器对根据前置行的图像数据及后置行的图像数据进行垂直滤波处理,得到待滤波行图像数据的滤波数据作为第一滤波数据。
优选的,在调用滤波器对预设行数的图像数据和待读取行的图像数据进行垂直滤波处理,得到第一滤波数据之前,还可以获取滤波器的参数,并对滤波器的参数进行量化处理,以实现对滤波器精度的调整。
进一步的,在获取滤波器的参数,并对滤波器的参数进行量化处理之后,还可以在接收到控制拍数修改命令时,根据控制拍数修改命令对滤波器的控制拍数进行修改,以实现对滤波器控制拍数的调整,进而方便数字芯片工程师对图像质量滤波器进行调试。
S103:根据第一滤波数据对图像数据进行更新,并判断待读取行是否为图像数据的最后一行;
若否,则进入步骤S104;若是,则进入步骤S105;
可选的,这里提到的根据第一滤波数据对图像数据进行更新,其具体可以为:
将待滤波行图像数据更新为待滤波行图像数据的滤波数据。
垂直滤波处理需要对图像数据进行垂直方向的处理,需要的行缓存极大,以9阶滤波器为例,若要对图像数据进行垂直滤波处理时,则需要同时缓存9行图像数据,在此基础上,如果再进行逐点反馈,则会极大的增加芯片成本及占用面积,因此本申请对调用滤波器对预设行数的图像数据和待读取行的图像数据进行垂直滤波处理,得到第一滤波数据,该第一滤波数据即为待滤波行的滤波数据,即通过逐行反馈的方式来实现图像数据的递归垂直滤波处理,极大的减少了芯片成本及占用面积。
S104:将待读取行更新为待读取行的下一行;
当待读取行不为图像数据的最后一行时,则表明此时还有数据可读,此时将待读取行更新为待读取行的下一行,并返回执行步骤S101,继续计算下一个待滤波行图像数据的滤波数据。
S105:调用滤波器对预设行数的图像数据和待读取行的图像数据进行垂直滤波处理,得到第二滤波数据,并根据第二滤波数据对图像数据进行更新,得到垂直滤波处理后的图像数据。
当待读取行为图像数据的最后一行时,则表明此时已无数据可读,此时调用滤波器对预设行数的图像数据和待读取行的图像数据进行垂直滤波处理,得到第二滤波数据,并根据第二滤波数据对图像数据进行更新,得到垂直滤波处理后的图像数据,完成此次图像垂直滤波处理。
基于上述技术方案,本申请所提供的一种图像垂直滤波处理的方法,通过调用滤波器对预设行数的图像数据和待读取行的图像数据进行垂直滤波处理,得到第一滤波数据,然后根据第一滤波数据对图像数据进行更新,依据数字递归滤波器的算法原理以及图像芯片的特性,将逐点反馈改为逐行反馈,优化了算法实现架构,提高了开发效率,并减少了维护时间,极大的提高了芯片对图像进行垂直滤波处理的速度,降低芯片成本。
针对于上一实施例的步骤S105,其中所描述的调用滤波器对预设行数的图像数据和待读取行的图像数据进行垂直滤波处理,得到第二滤波数据,下面结合图3进行说明。
请参考图3,图3为图1所提供的一种图像垂直滤波处理的方法中S105的一种实际表现方式的流程图。
其具体包括以下步骤:
S301:获取滤波器阶数,并根据滤波器阶数确定待滤波行图像数据的前置行数及后置行数;
S302:根据缓存的预设行数的图像数据生成前置行的图像数据,并根据缓存的预设行数的图像数据及图像数据的最后一行确定后置行的图像数据;
S303:调用滤波器对根据前置行的图像数据及后置行的图像数据进行垂直滤波处理,得到待滤波行图像数据的滤波数据作为第二滤波数据。
当待滤波行为边界行或邻近边界行时,存在着图像数据不足,无法完成滤波的问题,以9阶滤波器为例,当待滤波行为最后1行时,前置行为倒数第5-2行,而后置行为空,此时本申请可以根据缓存的预设行数的图像数据生成后置行的图像数据,即将后置行的图像数据全部用最后1行的图像数据进行替代。
在此基础上,步骤S105中提到的根据第二滤波数据对图像数据进行更新,得到垂直滤波处理后的图像数据,其具体可以为:
判断待滤波行是否为图像数据的最后一行;
若否,则将待滤波行的图像数据更新为第二滤波数据,并将待滤波行更新为待滤波行的下一行,返回执行步骤S302;
若是,则将最后一行的图像数据更新为第二滤波数据,并输出更新后的图像数据。
请参考图4,图4为本申请实施例所提供的一种图像垂直滤波处理的芯片的结构图。
该芯片可以包括:
缓存模块100,用于缓存预设行数的图像数据,并读取待读取行的图像数据;其中,待读取行为图像数据中未被缓存的第一行;
第一调用模块200,用于调用滤波器对预设行数的图像数据和待读取行的图像数据进行垂直滤波处理,得到第一滤波数据;
判断模块300,用于根据第一滤波数据对图像数据进行更新,并判断待读取行是否为图像数据的最后一行;
更新模块400,用于当待读取行不为图像数据的最后一行时,将待读取行更新为待读取行的下一行,并返回缓存模块100执行缓存预设行数的图像数据,并读取待读取行的图像数据的步骤;
第二调用模块500,用于当待读取行为图像数据的最后一行时,调用滤波器对预设行数的图像数据和待读取行的图像数据进行垂直滤波处理,得到第二滤波数据,并根据第二滤波数据对图像数据进行更新,得到垂直滤波处理后的图像数据。
在上述实施例的基础上,作为一种优选实施方式,该第一调用模块200可以包括:
获取单元,用于获取滤波器阶数,并根据滤波器阶数确定待滤波行图像数据的前置行数及后置行数;
第一确定单元,用于根据缓存的预设行数的图像数据生成前置行的图像数据,并根据缓存的预设行数的图像数据及待读取行的图像数据确定后置行的图像数据;
第一调用单元,用于调用滤波器对根据前置行的图像数据及后置行的图像数据进行垂直滤波处理,得到待滤波行图像数据的滤波数据作为第一滤波数据。
在上述实施例的基础上,作为一种优选实施方式,该判断模块300可以包括:
第一更新单元,用于将待滤波行图像数据更新为待滤波行图像数据的滤波数据。
在上述实施例的基础上,作为一种优选实施方式,该第二调用模块500可以包括:
第二确定单元,用于获取滤波器阶数,并根据滤波器阶数确定待滤波行图像数据的前置行数及后置行数;
第三确定单元,用于根据缓存的预设行数的图像数据生成前置行的图像数据,并根据缓存的预设行数的图像数据及图像数据的最后一行确定后置行的图像数据;
调用单元,用于调用滤波器对根据前置行的图像数据及后置行的图像数据进行垂直滤波处理,得到待滤波行图像数据的滤波数据作为第二滤波数据。
在上述实施例的基础上,作为一种优选实施方式,该第二调用模块500可以包括:
判断单元,用于判断待滤波行是否为图像数据的最后一行;
第二更新单元,用于当待滤波行不为图像数据的最后一行时,将待滤波行的图像数据更新为第二滤波数据,并将待滤波行更新为待滤波行的下一行,返回第三确定单元执行根据缓存的预设行数的图像数据生成前置行的图像数据,并根据缓存的预设行数的图像数据及图像数据的最后一行确定后置行的图像数据的步骤;
第三更新单元,用于当待滤波行为图像数据的最后一行时,将最后一行的图像数据更新为第二滤波数据,并输出更新后的图像数据。
在上述实施例的基础上,作为一种优选实施方式,该芯片还可以包括:
量化处理模块,用于获取滤波器的参数,并对滤波器的参数进行量化处理。
在上述实施例的基础上,作为一种优选实施方式,该芯片还可以包括:
修改模块,用于当接收到控制拍数修改命令时,根据控制拍数修改命令对滤波器的控制拍数进行修改。
由于芯片部分的实施例与方法部分的实施例相互对应,因此芯片部分的实施例请参见方法部分的实施例的描述,这里暂不赘述。
优选的,请参考图5,图5为本申请实施例所提供的另一种图像垂直滤波处理的芯片的结构图。
如图5所示,该芯片可以包括同步模块、数据流控制模块及计算模块,其中:
同步模块,用于处理准备/需求(ready/need)握手信号,ready信号代表上级模块告诉下级模块数据已准备好,高有效;need信号下级代表告诉上级模块可以接收数据,高有效。当ready与need均为高时,就完成了一次数据传递,即握手。在数字芯片图像处理算法中,每个算法都会有大量的控制与计算,如果每次处理都带着握手信号,就会使设计变得极为复杂,也难以维护,故本申请将该部分作为一个模块单独处理,根据图像的宽高等确定参数信息,将ready/need握手信号转化为有效(valid)信号、水平计数(h_cnt)信号以及垂直计数(v_cnt)信号,利用以上3个信号去控制数据流以及计算可以大大降低设计与控制的复杂度;
数据流控制模块,用于完成数据缓存控制、数据扩边控制、数据选择输出控制等功能;
计算模块,用于完成图像数据与系数的乘加计算。
由于芯片部分的实施例与方法部分的实施例相互对应,因此芯片部分的实施例请参见方法部分的实施例的描述,这里暂不赘述。
说明书中各个实施例采用递进的方式描述,每个实施例重点说明的都是与其他实施例的不同之处,各个实施例之间相同相似部分互相参见即可。
以上对本发明所提供的一种图像垂直滤波处理的方法及芯片进行了详细介绍。本文中应用了具体个例对本发明的原理及实施方式进行了阐述,以上 实施例的说明只是用于帮助理解本发明的方法及其核心思想。应当指出,对于本技术领域的普通技术人员来说,在不脱离本发明原理的前提下,还可以对本发明进行若干改进和修饰,这些改进和修饰也落入本发明权利要求的保护范围内。
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。在没有更多限制的情况下,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。

Claims (10)

  1. 一种图像垂直滤波处理的方法,其特征在于,包括:
    缓存预设行数的图像数据,并读取待读取行的图像数据;其中,所述待读取行为所述图像数据中未被缓存的第一行;
    调用滤波器对所述预设行数的图像数据和所述待读取行的图像数据进行垂直滤波处理,得到第一滤波数据;
    根据所述第一滤波数据对所述图像数据进行更新,并判断所述待读取行是否为所述图像数据的最后一行;
    若否,则将所述待读取行更新为所述待读取行的下一行,并返回执行缓存所述预设行数的图像数据,并读取待读取行的图像数据的步骤;
    若是,则调用所述滤波器对所述预设行数的图像数据和所述待读取行的图像数据进行垂直滤波处理,得到第二滤波数据,并根据所述第二滤波数据对所述图像数据进行更新,得到垂直滤波处理后的所述图像数据。
  2. 根据权利要求1所述的方法,其特征在于,所述调用滤波器对所述预设行数的图像数据和所述待读取行的图像数据进行垂直滤波处理,得到第一滤波数据,包括:
    获取滤波器阶数,并根据所述滤波器阶数确定待滤波行图像数据的前置行数及后置行数;
    根据缓存的所述预设行数的图像数据生成前置行的图像数据,并根据缓存的所述预设行数的图像数据及所述待读取行的图像数据确定后置行的图像数据;
    调用滤波器对所述前置行的图像数据及所述后置行的图像数据进行垂直滤波处理,得到所述待滤波行图像数据的滤波数据作为所述第一滤波数据。
  3. 根据权利要求2所述的方法,其特征在于,所述根据所述第一滤波数据对所述图像数据进行更新,包括:
    将所述待滤波行图像数据更新为所述待滤波行图像数据的滤波数据。
  4. 根据权利要求1所述的方法,其特征在于,所述调用所述滤波器对所述预设行数的图像数据和所述待读取行的图像数据进行垂直滤波处理,得到第二滤波数据,包括:
    获取滤波器阶数,并根据所述滤波器阶数确定待滤波行图像数据的前置行数及后置行数;
    根据缓存的所述预设行数的图像数据生成前置行的图像数据,并根据缓存的所述预设行数的图像数据及所述图像数据的最后一行确定后置行的图像数据;
    调用滤波器对所述前置行的图像数据及所述后置行的图像数据进行垂直滤波处理,得到所述待滤波行图像数据的滤波数据作为所述第二滤波数据。
  5. 根据权利要求4所述的方法,其特征在于,所述根据所述第二滤波数据对所述图像数据进行更新,得到垂直滤波处理后的所述图像数据,包括:
    判断所述待滤波行是否为所述图像数据的最后一行;
    若否,则将所述待滤波行的图像数据更新为所述第二滤波数据,并将所述待滤波行更新为所述待滤波行的下一行,返回执行根据缓存的所述预设行数的图像数据生成所述前置行的图像数据,并根据缓存的所述预设行数的图像数据及所述图像数据的最后一行确定所述后置行的图像数据的步骤;
    若是,则将最后一行的图像数据更新为所述第二滤波数据,并输出更新后的所述图像数据。
  6. 根据权利要求1所述的方法,其特征在于,在调用滤波器对所述预设行数的图像数据和所述待读取行的图像数据进行垂直滤波处理,得到第一滤波数据之前,还包括:
    获取所述滤波器的参数,并对所述滤波器的参数进行量化处理。
  7. 根据权利要求6所述的方法,其特征在于,在获取所述滤波器的参数,并对所述滤波器的参数进行量化处理之后,还包括:
    当接收到控制拍数修改命令时,根据所述控制拍数修改命令对所述滤波器的控制拍数进行修改。
  8. 一种图像垂直滤波处理的芯片,其特征在于,包括:
    缓存模块,用于缓存预设行数的图像数据,并读取待读取行的图像数据;其中,所述待读取行为所述图像数据中未被缓存的第一行;
    第一调用模块,用于调用滤波器对所述预设行数的图像数据和所述待读取行的图像数据进行垂直滤波处理,得到第一滤波数据;
    判断模块,用于根据所述第一滤波数据对所述图像数据进行更新,并判断所述待读取行是否为所述图像数据的最后一行;
    更新模块,用于当所述待读取行不为所述图像数据的最后一行时,将所述待读取行更新为所述待读取行的下一行,并返回所述缓存模块执行缓存所述预设行数的图像数据,并读取待读取行的图像数据的步骤;
    第二调用模块,用于当所述待读取行为所述图像数据的最后一行时,调用所述滤波器对所述预设行数的图像数据和所述待读取行的图像数据进行垂直滤波处理,得到第二滤波数据,并根据所述第二滤波数据对所述图像数据进行更新,得到垂直滤波处理后的所述图像数据。
  9. 根据权利要求8所述的芯片,其特征在于,所述第一调用模块包括:
    获取单元,用于获取滤波器阶数,并根据所述滤波器阶数确定待滤波行图像数据的前置行数及后置行数;
    第一确定单元,用于根据缓存的所述预设行数的图像数据生成所述前置行的图像数据,并根据缓存的所述预设行数的图像数据及所述待读取行的图像数据确定所述后置行的图像数据;
    第一调用单元,用于调用滤波器对所述前置行的图像数据及所述后置行的图像数据进行垂直滤波处理,得到所述待滤波行图像数据的滤波数据作为所述第一滤波数据。
  10. 根据权利要求9所述的芯片,其特征在于,所述判断模块包括:
    第一更新单元,用于将所述待滤波行图像数据更新为所述待滤波行图像数据的滤波数据。
PCT/CN2020/126767 2020-02-21 2020-11-05 一种图像垂直滤波处理的方法及芯片 WO2021164324A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/780,248 US11659173B2 (en) 2020-02-21 2020-11-05 Method and chip for vertically filtering image

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010108903.7 2020-02-21
CN202010108903.7A CN111314579B (zh) 2020-02-21 2020-02-21 一种图像垂直滤波处理的方法及芯片

Publications (1)

Publication Number Publication Date
WO2021164324A1 true WO2021164324A1 (zh) 2021-08-26

Family

ID=71147519

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/126767 WO2021164324A1 (zh) 2020-02-21 2020-11-05 一种图像垂直滤波处理的方法及芯片

Country Status (3)

Country Link
US (1) US11659173B2 (zh)
CN (1) CN111314579B (zh)
WO (1) WO2021164324A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111314579B (zh) * 2020-02-21 2021-10-15 苏州浪潮智能科技有限公司 一种图像垂直滤波处理的方法及芯片

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2073553A1 (en) * 2007-12-21 2009-06-24 Thomson Licensing Method and apparatus for performing de-blocking filtering of a video picture
CN104240181A (zh) * 2014-08-29 2014-12-24 中国航天科工集团第三研究院第八三五七研究所 一种垂直滤波器的图像垂直尺寸缩小方法
CN109064424A (zh) * 2018-07-24 2018-12-21 郑州云海信息技术有限公司 图像滤波数据存取的控制方法及装置
CN110517201A (zh) * 2019-08-29 2019-11-29 北京迈格威科技有限公司 循环保边平滑滤波的方法、装置和电子设备
CN111314579A (zh) * 2020-02-21 2020-06-19 苏州浪潮智能科技有限公司 一种图像垂直滤波处理的方法及芯片

Family Cites Families (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6973127B1 (en) * 1999-12-23 2005-12-06 Xvd Corporation Apparatus and method for memory saving wavelet based video coding
WO2003001449A2 (en) * 2001-06-11 2003-01-03 Silicon Optix Inc. Method and system for processing two-dimensional image data
US8009729B2 (en) * 2005-11-30 2011-08-30 Qualcomm Incorporated Scaler architecture for image and video processing
US7595805B2 (en) * 2006-04-11 2009-09-29 Qualcomm Incorporated Techniques to facilitate use of small line buffers for processing of small or large images
CN101409833B (zh) * 2007-10-12 2012-10-03 昆山杰得微电子有限公司 去块效应滤波装置及方法
CN101286226B (zh) * 2008-05-23 2010-07-07 杭州华三通信技术有限公司 一种缩小图像的数字滤波方法和装置
CN101600049A (zh) * 2008-06-06 2009-12-09 联咏科技股份有限公司 图像处理装置与方法
CN102456214B (zh) * 2010-10-21 2013-10-30 珠海全志科技股份有限公司 图像缩放中线缓存的使用方法和装置
PH12018501811A1 (en) * 2010-12-07 2019-02-18 Sony Corp Image processing device and image processing method
AU2013273830A1 (en) * 2013-12-23 2015-07-09 Canon Kabushiki Kaisha Post-processed bokeh rendering using asymmetric recursive Gaussian filters
CN104050635B (zh) * 2014-05-30 2017-02-15 北京理工大学 一种模板尺寸可变的图像非线性滤波实时处理系统及方法
CN108369725A (zh) * 2017-03-13 2018-08-03 深圳市大疆创新科技有限公司 处理图像的方法、芯片、处理器、计算机系统和移动设备

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2073553A1 (en) * 2007-12-21 2009-06-24 Thomson Licensing Method and apparatus for performing de-blocking filtering of a video picture
CN104240181A (zh) * 2014-08-29 2014-12-24 中国航天科工集团第三研究院第八三五七研究所 一种垂直滤波器的图像垂直尺寸缩小方法
CN109064424A (zh) * 2018-07-24 2018-12-21 郑州云海信息技术有限公司 图像滤波数据存取的控制方法及装置
CN110517201A (zh) * 2019-08-29 2019-11-29 北京迈格威科技有限公司 循环保边平滑滤波的方法、装置和电子设备
CN111314579A (zh) * 2020-02-21 2020-06-19 苏州浪潮智能科技有限公司 一种图像垂直滤波处理的方法及芯片

Also Published As

Publication number Publication date
CN111314579B (zh) 2021-10-15
US11659173B2 (en) 2023-05-23
CN111314579A (zh) 2020-06-19
US20220417508A1 (en) 2022-12-29

Similar Documents

Publication Publication Date Title
CN100576882C (zh) Cmos图像传感器固定模式噪声消除电路
WO2021164324A1 (zh) 一种图像垂直滤波处理的方法及芯片
JP4389766B2 (ja) 撮像装置
CN109993722A (zh) 图像处理方法、装置、存储介质及电子设备
TW202213269A (zh) 影像處理方法、裝置、電腦設備及儲存媒體
JP5602532B2 (ja) 画像処理装置および画像処理方法
JP2015144435A (ja) 小さな画像又は大きな画像を処理する小さなラインバッファを容易に用いる技術
CN101753820A (zh) 信息处理装置、缓冲器控制方法和计算机程序
KR930010027B1 (ko) 화상의 움직임벡터검출장치
JP6985450B2 (ja) 信号対雑音比調整回路、信号対雑音比調整方法および信号対雑音比調整プログラム
CN109299030B (zh) 基于ZYNQ的cameralink转帕尔制的方法
CN111428560A (zh) 视频图像处理方法、装置、电子设备及存储介质
US20100110213A1 (en) Image processing processor, image processing method, and imaging apparatus
TWI377521B (en) Image processing apparatus and method
KR20090004574A (ko) 연산 유닛 및 화상 필터링 장치
CN110996005B (zh) 一种实时数字图像增强方法及系统
CN1327690C (zh) 一种视频图像缩放过程中的清晰度补偿方法
CN115942128A (zh) 一种基于异构平台的isp系统设计与实现方法
US20080049120A1 (en) Video frame buffer
CN104967796B (zh) 超分辨率智能图像传感器芯片
CN100399797C (zh) 影像增显系统
CN103813054A (zh) 图像处理设备、成像设备、图像处理方法和程序
CN101115153A (zh) 模拟前端装置和图像拾取装置
US8527683B2 (en) System and method for processing digital data
WO2024093396A1 (zh) 一种图像处理方法、装置及电子设备和非易失性可读存储介质

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: 20919723

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: 20919723

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 20919723

Country of ref document: EP

Kind code of ref document: A1