公开(公告)号：US11030715B2

公开(公告)日：2021-06-08

申请号：US16544066

申请日：2019-08-19

Applicant: Huawei Technologies Co., Ltd.

Inventor： Chenglin Zheng

IPC: G06T3/40 ,  G06T7/13

Abstract: An image processing method includes obtaining a difference map of an original image, obtaining an edge feature parameter of a first pixel in an intermediate image based on a pixel value of a pixel in the difference map, obtaining an edge feature parameter of a second pixel in a target image based on the edge feature parameter, and obtaining the target image based on the edge feature parameter of the second pixel.

公开(公告)号：US09471958B2

公开(公告)日：2016-10-18

申请号：US14570693

申请日：2014-12-15

Applicant: Huawei Technologies Co., Ltd.

Inventor： Chenglin Zheng

IPC: G06T3/40

CPC classification number: G06T3/4076 ,  G06T3/4053

Abstract: Embodiments of the present invention provide an image processing method and apparatus, where the image processing method includes: performing image magnification processing on a low-resolution image to obtain an initial high-resolution image; selecting a central similar block corresponding to a first pixel from the initial high-resolution image; selecting a search block corresponding to the first pixel from the low-resolution image, and determining each similar block in the search block, where the central similar block and the search block have a same block size; and determining a pixel value of the first pixel according to the central similar block and each similar block in the search block. By using technical solutions in the embodiments of the present invention, calculation complexity and an occupied memory resource can be effectively reduced.

Abstract translation: 本发明的实施例提供一种图像处理方法和装置，其中图像处理方法包括：对低分辨率图像执行图像放大处理以获得初始高分辨率图像; 从初始高分辨率图像中选择对应于第一像素的中央相似块; 从所述低分辨率图像中选择与所述第一像素相对应的搜索块，并且确定所述搜索块中的每个相似块，其中所述中央相似块和所述搜索块具有相同的块大小; 以及根据所述搜索块中的中央相似块和每个相似块确定所述第一像素的像素值。 通过在本发明的实施例中使用技术方案，可以有效地减少计算复杂度和占用的存储器资源。

公开(公告)号：US20140072232A1

公开(公告)日：2014-03-13

申请号：US14016705

申请日：2013-09-03

Applicant: Huawei Technologies Co., LTD

Inventor： Chenglin Zheng ,  Hai Chen

IPC: G06T3/40

CPC classification number: G06T3/4053

Abstract: A super-resolution method and apparatus for a video image are disclosed. The method includes performing super-resolution processing based on fuzzy motion estimation by integral multiple times on an original low-resolution video image, to obtain an intermediate high-resolution video image. When resolution of the intermediate high-resolution video image is different from resolution of a target high-resolution video image, a non-integral multiple between the resolution of the target high-resolution video image and the resolution of the intermediate high-resolution video image is calculated. Super-resolution processing based on interpolation by the non-integral is performed multiple times on the intermediate high-resolution video image, to obtain the target high-resolution video image.

Abstract translation: 公开了一种用于视频图像的超分辨率方法和装置。 该方法包括在原始低分辨率视频图像上基于模糊运动估计多次进行超分辨率处理，以获得中间高分辨率视频图像。 当中间高分辨率视频图像的分辨率与目标高分辨率视频图像的分辨率不同时，目标高分辨率视频图像的分辨率与中间高分辨率视频图像的分辨率之间的非整数倍 被计算。 在中间高分辨率视频图像上多次执行基于非积分的内插的超分辨率处理，以获得目标高分辨率视频图像。

公开(公告)号：US09076234B2

公开(公告)日：2015-07-07

申请号：US14016705

申请日：2013-09-03

Applicant: Huawei Technologies Co., Ltd

Inventor： Chenglin Zheng ,  Hai Chen

IPC: G06K9/32 ,  G06T3/40

CPC classification number: G06T3/4053

Abstract: A super-resolution method and apparatus for a video image are disclosed. The method includes performing super-resolution processing based on fuzzy motion estimation by integral multiple times on an original low-resolution video image, to obtain an intermediate high-resolution video image. When resolution of the intermediate high-resolution video image is different from resolution of a target high-resolution video image, a non-integral multiple between the resolution of the target high-resolution video image and the resolution of the intermediate high-resolution video image is calculated. Super-resolution processing based on interpolation by the non-integral is performed multiple times on the intermediate high-resolution video image, to obtain the target high-resolution video image.

Abstract translation: 公开了一种用于视频图像的超分辨率方法和装置。 该方法包括在原始低分辨率视频图像上基于模糊运动估计多次进行超分辨率处理，以获得中间高分辨率视频图像。 当中间高分辨率视频图像的分辨率与目标高分辨率视频图像的分辨率不同时，目标高分辨率视频图像的分辨率与中间高分辨率视频图像的分辨率之间的非整数倍 被计算。 在中间高分辨率视频图像上多次执行基于非积分的内插的超分辨率处理，以获得目标高分辨率视频图像。

公开(公告)号：US20210027162A1

公开(公告)日：2021-01-28

申请号：US17068358

申请日：2020-10-12

Applicant: Huawei Technologies Co., Ltd.

Inventor： Fan Yang ,  Chenglin Zheng

IPC: G06N3/08 ,  G06N3/04 ,  G06F7/544

Abstract: A neural network model of M network layers, a data processing method, and a processing apparatus configured to execute N tasks, where an ith network layer has a shared weight value to execute each of the N tasks and N groups of dedicated weight values, where each of the N groups of dedicated weight values executes one of the N tasks, all the groups of dedicated weight values are in a one-to-one correspondence with the N tasks, M is a positive integer and 1≤i≤M, when executing a first task, the ith network layer is configured to obtain input data, obtain output data based on a tth group of dedicated weight values, the shared weight value, and the input data, when 1≤i≤M, transmit the output data to an (i+1)th network, where the tth group of dedicated weight values corresponds to the first task, and when i=M, output the output data.