Title:
DEEP-LEARNING-BASED SUPER-RESOLUTION RECONSTRUCTION METHOD FOR MICROSCOPIC IMAGE, AND MEDIUM AND ELECTRONIC DEVICE
Document Type and Number:
WIPO Patent Application WO/2022/111368
Kind Code:
A1
Abstract:
A deep-learning-based super-resolution reconstruction method for a microscopic image, and a medium and an electronic device. The method comprises the following steps: performing wavelet decomposition on a low-resolution image to obtain a first coefficient sub-image; inputting the first coefficient sub-image into a variational auto-encoder to obtain a second coefficient sub-image that has a high resolution; performing inverse wavelet transformation on the basis of the second coefficient sub-image to reconstruct and obtain a high-resolution image; and inputting the low-resolution image into a trained convolutional neural network, adding the reconstructed high-resolution image into the convolutional neural network in a residual manner, and obtaining, by means of upsampling, a super-resolution image having the same size as the high-resolution image. Compared with the prior art, the method has the advantages of high reconstruction precision, convenience, etc.
Inventors:
HUANG GANG (CN)
YE XUEHUA (CN)
NIE SHENGDONG (CN)
YE XUEHUA (CN)
NIE SHENGDONG (CN)
Application Number:
PCT/CN2021/131344
Publication Date:
June 02, 2022
Filing Date:
November 18, 2021
Export Citation:
Assignee:
SHANGHAI UNIV OF MEDICINE&HEALTH SCIENCES (CN)
UNIV OF SHANGHAI FOR SCIENCE AND TECHNOLOGY (CN)
UNIV OF SHANGHAI FOR SCIENCE AND TECHNOLOGY (CN)
International Classes:
G06T3/40; G06N3/04
Foreign References:
| CN112465701A | 2021-03-09 | |||
| CN108734660A | 2018-11-02 | |||
| CN109636716A | 2019-04-16 | |||
| EP3493149A1 | 2019-06-05 |
Attorney, Agent or Firm:
SHANGHAI KESHENG INTELLECTUAL PROPERTY AGENCY LTD. (CN)
Download PDF: