Invention Grant
- Patent Title: Image processing method and image receiving apparatus
-
Application No.: US16500535Application Date: 2018-04-09
-
Publication No.: US11238559B2Publication Date: 2022-02-01
- Inventor: Masataka Shiokawa , Kei Takahashi , Yuji Iwaki , Hiromichi Godo
- Applicant: Semiconductor Energy Laboratory Co., Ltd.
- Applicant Address: JP Atsugi
- Assignee: Semiconductor Energy Laboratory Co., Ltd.
- Current Assignee: Semiconductor Energy Laboratory Co., Ltd.
- Current Assignee Address: JP Atsugi
- Agency: Robinson Intellectual Property Law Office
- Agent Eric J. Robinson
- Priority: JPJP2017-084603 20170421
- International Application: PCT/IB2018/052435 WO 20180409
- International Announcement: WO2018/193333 WO 20181025
- Main IPC: G06T3/40
- IPC: G06T3/40 ; G06T5/00

Abstract:
An image processing method that generates high-resolution image data from low-resolution image data and an image receiving apparatus that is operated by the image processing method are provided.
In the image processing method that generates high-resolution image data from low-resolution image data, the low-resolution image data is divided to generate a plurality of first image data, among the plurality of first image data, one of two adjacent image data is second image data, and the other is third image data. Surroundings of the second image data are supplemented with pixel data to generate fourth image data. The pixel data includes part of the third image data. A convolutional neural network using the fourth image data as an input is implemented, fifth image data is output from the convolutional neural network, and a plurality of the fifth image data is combined to generate high-resolution image data. The image receiving apparatus is operated by the image processing method.
In the image processing method that generates high-resolution image data from low-resolution image data, the low-resolution image data is divided to generate a plurality of first image data, among the plurality of first image data, one of two adjacent image data is second image data, and the other is third image data. Surroundings of the second image data are supplemented with pixel data to generate fourth image data. The pixel data includes part of the third image data. A convolutional neural network using the fourth image data as an input is implemented, fifth image data is output from the convolutional neural network, and a plurality of the fifth image data is combined to generate high-resolution image data. The image receiving apparatus is operated by the image processing method.
Public/Granted literature
- US20200051211A1 IMAGE PROCESSING METHOD AND IMAGE RECEIVING APPARATUS Public/Granted day:2020-02-13
Information query