-
公开(公告)号:US20220245413A1
公开(公告)日:2022-08-04
申请号:US17163512
申请日:2021-01-31
Applicant: KYOCERA Document Solutions Inc.
Inventor: Kunihiko SHIMAMOTO , Norifumi MIYAHARA , Kendrick Esperanza Wong
Abstract: Provided is an image processing apparatus that improves the performance when performing image processing accompanying an analysis. An image acquisition unit acquires processing image data for performing processing. A normal image processing unit performs normal image processing for normal output on the processing image data acquired by the image acquisition unit and outputs the normal image data. An analysis image processing unit performs analysis image processing accompanied by analysis in parallel with the normal image processing by the normal image processing unit and outputs the analysis image data. A merge processing unit merges the normal image data output by the normal image processing unit and the analysis image data output by the analysis image processing unit and outputs print data for printing.
-
公开(公告)号:US10977531B1
公开(公告)日:2021-04-13
申请号:US16699493
申请日:2019-11-29
Applicant: KYOCERA Document Solutions Inc.
Inventor: Masayoshi Nakamura , Kendrick Esperanza Wong
Abstract: Provided is an image processing apparatus that improves performance of reading data of a RIP process. A multi-channel DMA (Direct Memory Access) controller performs reading or writing of stored data related to a RIP (Raster Image Processor). A FIFO (First In, First Out) is able to change a number of entries for each of the channels of the multi-channel DMA controller. An image processing control unit interprets the RIP order, and according to that order, changes the configuration of the number of entries for each of the channels, and controls processing of the data.
-
公开(公告)号:US20210104011A1
公开(公告)日:2021-04-08
申请号:US16595985
申请日:2019-10-08
Applicant: KYOCERA Document Solutions Inc.
Inventor: Kendrick Esperanza Wong , Masayoshi Nakamura
IPC: G06T1/60
Abstract: Embodiments improve processing of data by determining if a read-modify-write operation on a frame is necessary or not. Some frames may be converted into a block of 8 bpp data. There may be no need to read the destination since the unnecessary pixels may be protected by the byte-enables. The burst write transfer may be performed for the entire frame when it is 8 bpp depth. An original transfer frame may be split into smaller portions of the frame. One or more of the smaller frame portions may be converted into byte alignment thus obviating the need for the read function to be performed on the smaller frame portions. Accordingly, significant bits of data are no longer processed under this operation which speeds up the overall processing of data. Portions of transfer frames that may not be converted to 8 bpp may be processed with read-modify-write operations.
-
-