公开(公告)号：US11409655B2

公开(公告)日：2022-08-09

申请号：US16799061

申请日：2020-02-24

Applicant: CANON KABUSHIKI KAISHA

Inventor： Tadayuki Ito

IPC: G06F12/084 ,  G06F12/0804 ,  G06F12/0875 ,  G06F12/0891

Abstract: There is provided with an interface apparatus. The interface apparatus provides a shared cache for a plurality of processing units. A first port acquires data from a first processing unit included in the plurality of processing units. A second port outputs the data acquired from the first processing unit to a second processing unit included in the plurality of processing units. A cache caches the data acquired from the first processing unit. A controller controls, based on information acquired from the second processing unit, whether to write back data written in the cache to a memory different from the cache.

公开(公告)号：US20190005633A1

公开(公告)日：2019-01-03

申请号：US16122208

申请日：2018-09-05

Applicant: CANON KABUSHIKI KAISHA

Inventor： Takayuki Nagashima ,  Tadayuki Ito ,  Yutaka Murata

IPC: G06T5/50 ,  H04N5/14 ,  H04N5/232

Abstract: An image processing apparatus includes: a filter unit configured to execute a filter process on each block area in the input image and output the filter processing result of each pixel in the input image; a supply unit configured to read out attribute information from the attribute map and supply attribute information for each pixel; and an image processing unit configured to perform the predetermined image processing for the filter processing result output by the filter unit based on the attribute information for each pixel supplied by the supply unit, wherein the pixel order in which the pixel value in the input image is received by the filter processing unit and the pixel order in which the filter processing unit outputs the filter processing result are different, and wherein the supply unit supplies the attribute information for each pixel according to the pixel order of the filter processing result.

公开(公告)号：US09055253B2

公开(公告)日：2015-06-09

申请号：US14221847

申请日：2014-03-21

Applicant: CANON KABUSHIKI KAISHA

Inventor： Tadayuki Ito

IPC: G06K9/34 ,  H04N1/409 ,  G06T1/60

CPC classification number: H04N1/4092 ,  G06T1/60 ,  H04N1/409

Abstract: When processing image data by referring to at least one of reference image data and correction data, an image is processed in a unit of band area by dividing image data of a target image in a plurality of band areas. In this case, the minimum transfer unit of the image data is set. The minimum transfer unit of at least one of the reference image data and the correction data is obtained to process the image data in the minimum transfer unit. The transfer rate of the image data is determined so that the image data to be buffered fits in a capacity of a buffer available in image processing. The height of the band area to be created by division is determined based on the ratio of the transfer rate of the image data to the minimum transfer unit of the image data.

Abstract translation: 通过参考参考图像数据和校正数据中的至少一个来处理图像数据时，通过划分多个带区域中的目标图像的图像数据，以带区域为单位处理图像。 在这种情况下，设定图像数据的最小传送单位。 获得参考图像数据和校正数据中的至少一个的最小传送单元，以处理最小传送单元中的图像数据。 确定图像数据的传送速率，使得要缓冲的图像数据适合图像处理中可用的缓冲器的容量。 基于图像数据的传送速率与图像数据的最小传送单位的比率来确定要由分割创建的带区域的高度。

公开(公告)号：US08724901B2

公开(公告)日：2014-05-13

申请号：US13912558

申请日：2013-06-07

Applicant: Canon Kabushiki Kaisha

Inventor： Tadayuki Ito

IPC: G06K9/34 ,  G06K9/20

CPC classification number: H04N1/4092 ,  G06T1/60 ,  H04N1/409

Abstract: When processing image data by referring to at least one of reference image data and correction data, an image is processed in a unit of band area by dividing image data of a target image in a plurality of band areas. In this case, the minimum transfer unit of the image data is set. The minimum transfer unit of at least one of the reference image data and the correction data is obtained to process the image data in the minimum transfer unit. The transfer rate of the image data is determined so that the image data to be buffered fits in a capacity of a buffer available in image processing. The height of the band area to be created by division is determined based on the ratio of the transfer rate of the image data to the minimum transfer unit of the image data.

Abstract translation: 通过参考参考图像数据和校正数据中的至少一个来处理图像数据时，通过划分多个带区域中的目标图像的图像数据，以带区域为单位处理图像。 在这种情况下，设定图像数据的最小传送单位。 获得参考图像数据和校正数据中的至少一个的最小传送单元，以处理最小传送单元中的图像数据。 确定图像数据的传送速率，使得要缓冲的图像数据适合图像处理中可用的缓冲器的容量。 基于图像数据的传送速率与图像数据的最小传送单位的比率来确定要由分割创建的带区域的高度。

公开(公告)号：US11119924B2

公开(公告)日：2021-09-14

申请号：US15919608

申请日：2018-03-13

Applicant: CANON KABUSHIKI KAISHA

Inventor： Tadayuki Ito

IPC: G06F12/0815 ,  G06F12/0855 ,  G06F12/0853 ,  G06F12/084 ,  G06F12/0846

Abstract: In order to allow efficient data communication, an interface device that includes N ports, comprises: a cache memory that is shared by the N ports and includes a plurality of cache tags each of which is allocated to one of the N ports; and N cache determination units corresponding to the N ports. Each of the N cache determination units comprises: a determiner configured to determine, based on all of the values of the plurality of cache tags, whether a cache miss has occurred in the cache memory, and an update unit configured to update, when the determiner determines that a cache miss has occurred, cache tag values allocated to a self-port.

公开(公告)号：US20180276126A1

公开(公告)日：2018-09-27

申请号：US15919608

申请日：2018-03-13

Applicant: CANON KABUSHIKI KAISHA

Inventor： Tadayuki Ito

IPC: G06F12/0815 ,  G06F12/0855

CPC classification number: G06F12/0815 ,  G06F12/084 ,  G06F12/0846 ,  G06F12/0853 ,  G06F12/0855 ,  G06F2212/1008 ,  G06F2212/1021

Abstract: In order to allow efficient data communication, an interface device that includes N ports, comprises: a cache memory that is shared by the N ports and includes a plurality of cache tags each of which is allocated to one of the N ports; and N cache determination units corresponding to the N ports. Each of the N cache determination units comprises: a determiner configured to determine, based on all of the values of the plurality of cache tags, whether a cache miss has occurred in the cache memory, and an update unit configured to update, when the determiner determines that a cache miss has occurred, cache tag values allocated to a self-port.

公开(公告)号：US20140204428A1

公开(公告)日：2014-07-24

申请号：US14221847

申请日：2014-03-21

Applicant: CANON KABUSHIKI KAISHA

Inventor： Tadayuki Ito

IPC: H04N1/409

CPC classification number: H04N1/4092 ,  G06T1/60 ,  H04N1/409

Abstract: When processing image data by referring to at least one of reference image data and correction data, an image is processed in a unit of band area by dividing image data of a target image in a plurality of band areas. In this case, the minimum transfer unit of the image data is set. The minimum transfer unit of at least one of the reference image data and the correction data is obtained to process the image data in the minimum transfer unit. The transfer rate of the image data is determined so that the image data to be buffered fits in a capacity of a buffer available in image processing. The height of the band area to be created by division is determined based on the ratio of the transfer rate of the image data to the minimum transfer unit of the image data.

Abstract translation: 通过参考参考图像数据和校正数据中的至少一个来处理图像数据时，通过划分多个带区域中的目标图像的图像数据，以带区域为单位处理图像。 在这种情况下，设定图像数据的最小传送单位。 获得参考图像数据和校正数据中的至少一个的最小传送单元，以处理最小传送单元中的图像数据。 确定图像数据的传送速率，使得要缓冲的图像数据适合图像处理中可用的缓冲器的容量。 基于图像数据的传送速率与图像数据的最小传送单位的比率来确定要由分割创建的带区域的高度。

公开(公告)号：US20130271797A1

公开(公告)日：2013-10-17

申请号：US13912558

申请日：2013-06-07

Applicant: CANON KABUSHIKI KAISHA

Inventor： Tadayuki Ito

IPC: H04N1/409

CPC classification number: H04N1/4092 ,  G06T1/60 ,  H04N1/409

Abstract: When processing image data by referring to at least one of reference image data and correction data, an image is processed in a unit of band area by dividing image data of a target image in a plurality of band areas. In this case, the minimum transfer unit of the image data is set. The minimum transfer unit of at least one of the reference image data and the correction data is obtained to process the image data in the minimum transfer unit. The transfer rate of the image data is determined so that the image data to be buffered fits in a capacity of a buffer available in image processing. The height of the band area to be created by division is determined based on the ratio of the transfer rate of the image data to the minimum transfer unit of the image data.

Abstract translation: 通过参考参考图像数据和校正数据中的至少一个来处理图像数据时，通过划分多个带区域中的目标图像的图像数据，以带区域为单位处理图像。 在这种情况下，设定图像数据的最小传送单位。 获得参考图像数据和校正数据中的至少一个的最小传送单元，以处理最小传送单元中的图像数据。 确定图像数据的传送速率，使得要缓冲的图像数据适合图像处理中可用的缓冲器的容量。 基于图像数据的传送速率与图像数据的最小传送单位的比率来确定要由分割创建的带区域的高度。

公开(公告)号：US09672874B2

公开(公告)日：2017-06-06

申请号：US13945799

申请日：2013-07-18

Applicant: CANON KABUSHIKI KAISHA

Inventor： Tadayuki Ito

IPC: G11C7/00 ,  G06T1/60

CPC classification number: G11C7/00 ,  G06T1/60

Abstract: In a storage device including a plurality of storage regions, in order to synchronize input/output control of data into/from the storage region, a writing sequence or a reading sequence of the data into or from the storage region is stored, one of the storage regions to be accessed is selected in accordance with the stored sequence, and the synchronization control of input/output processing into/from an intermediate buffer is carried out, which allows an intermediate buffer control mechanism to be applied to various intermediate buffers.

公开(公告)号：US09569703B2

公开(公告)日：2017-02-14

申请号：US14715991

申请日：2015-05-19

Applicant: CANON KABUSHIKI KAISHA

Inventor： Tadayuki Ito

IPC: G06K15/02 ,  G06F3/12 ,  G06F12/1081 ,  G06T1/60

CPC classification number: G06K15/181 ,  G06K15/1817 ,  G06T1/60

Abstract: A reading process is performed to read data of each area, which is divided from image data, from a buffer storing the image data to be performed on a predetermined image process. A transmitting process is performed to transmit the read data of the area to an image processor. A reception process is performed to receive data generated in the predetermined image process from the image processor. A parameter for a writing process is set based on an area size of the received data. A writing process is performed to write the received data to the buffer by data transfer using direct memory access (DMA).

Abstract translation: 执行读取处理，以从存储要在预定图像处理中执行的图像数据的缓冲器中读取与图像数据分割的每个区域的数据。 执行发送处理以将该区域的读取数据发送到图像处理器。 执行接收处理以从图像处理器接收在预定图像处理中生成的数据。 基于接收到的数据的面积来设定写入处理的参数。 执行写入处理，以通过使用直接存储器访问（DMA）的数据传送将接收的数据写入缓冲器。