13-01-2022 дата публикации
Номер: US20220012859A1
Автор:
FENG Xiaoxia,
FU Yuankun,
JIN Junwei,
Li Jingjing,
Li Meng,
LIU Dongjiang,
Liu Yuanyuan,
SHI Shimeng,
Shi Zhiyuan,
SI Ranran,
WANG Tianci,
Xie Zhifeng,
Zhang Kunpeng,
ZHANG LEI,
Zhao Liang,
Zhou Chuan
Принадлежит:
A method for parallel processing of retinal images includes: optimizing an objective function with a chaotic supply-demand algorithm to enhance a real retinal image; synthesizing a virtual retinal image by a hybrid image generation method; establishing a parallel multi-layer decomposed interval type-2 intuitionistic fuzzy convolutional neural network model based on the virtual retinal image and the enhanced real retinal image; and integrating outputs from a plurality of parallel multi-layer decomposed interval type-2 intuitionistic fuzzy convolutional neural network models as a final classification result. 1. A method , comprising:1) optimizing an objective function with a chaotic supply-demand algorithm to enhance a real retinal image;2) synthesizing a virtual retinal image by a hybrid image generation method;3) establishing a parallel multi-layer decomposed interval type-2 intuitionistic fuzzy convolutional neural network model based on the virtual retinal image and the real retinal image enhanced; and4) integrating outputs from a plurality of parallel multi-layer decomposed interval type-2 intuitionistic fuzzy convolutional neural network models as a final classification result.2. The method of claim 1 , wherein 1) is implemented as follows: {'br': None, 'i': EI', 'i,j', 'M', 'I', 'i,j, '()=(()|θ)\u2003\u2003(1)'}, '1.1) performing function transformation M on an input original real retinal image to obtain an enhanced output image{'o': [{'@ostyle': 'single', 'P'}, {'@ostyle': 'single', 'P'}, {'@ostyle': 'single', 'P'}, {'@ostyle': 'single', 'P'}], 'sup': σ', 'n', 'n', '2', 'n', 'n', '2', '1/2, 'sub': p=1', 'q=1', 'p=1', 'q=1, 'where I(i, j) denotes a part of the input original real retinal image, and EI(i, j) denotes a corresponding part of an enhanced retinal image, i, j=1, 2, . . . , n; θ=(r, β, ρ, σ) is an undetermined parameter of the function (1); M(I(i, j)|θ)=M(I(i, j)|r, β, ρ, σ)=rP/(s(i, j)+β)(I(i, j)−p×(i, j))+(i, j) denotes an image enhancement function ...
Подробнее
