公开(公告)号：US20220414891A1

公开(公告)日：2022-12-29

申请号：US17641445

申请日：2020-10-31

Applicant: SOUTH CHINA UNIVERSITY OF TECHNOLOGY

Inventor： Junying CHEN ,  Haijun YOU

IPC: G06T7/149 ,  G06T7/12 ,  G06T7/62 ,  G06T7/11

Abstract: The present invention discloses a method for automatic segmentation of a fuzzy boundary image based on active contour and deep learning. In the method, firstly, a fuzzy boundary image is segmented using a deep convolutional neural network model to obtain an initial segmentation result; then, a contour of a region inside the image segmented using the deep convolutional neural network model is used as an initialized contour and a contour constraint of an active contour model; and the active contour model drives, through image characteristics of a surrounding region of each contour point, the contour to move towards a target edge to derive an accurate segmentation line between a target region and other background regions. The present invention introduces an active contour model on the basis of a deep convolutional neural network model to further refine a segmentation result of a fuzzy boundary image, which has the capability of segmenting a fuzzy boundary in the image, thus further improving the accuracy of segmentation of the fuzzy boundary image.

公开(公告)号：US20220343466A1

公开(公告)日：2022-10-27

申请号：US17626503

申请日：2019-10-25

Applicant: SOUTH CHINA UNIVERSITY OF TECHNOLOGY

Inventor： Junying CHEN ,  Renxin ZHUANG

IPC: G06T5/00 ,  G06N3/08 ,  G06N3/04

Abstract: Disclosed is a high-contrast minimum variance imaging method based on deep learning. For the problem of the poor performance of a traditional minimum variance imaging method in terms of ultrasonic image contrast, a deep neural network is applied in order to suppress an off-axis scattering signal in channel data received by an ultrasonic transducer, and after the deep neural network is combined with a minimum variance beamforming method, an ultrasonic image with a higher contrast can be obtained while the resolution performance of the minimum variance imaging method is maintained. In the present method, compared with the traditional minimum variance imaging method, after an apodization weight is calculated, channel data is first processed by using a deep neural network, and weighted stacking of the channel data is then carried out, so that the pixel value of a target imaging point is obtained, thereby forming a complete ultrasonic image.

公开(公告)号：US20210213436A1

公开(公告)日：2021-07-15

申请号：US16330103

申请日：2017-11-27

Applicant: SOUTH CHINA UNIVERSITY OF TECHNOLOGY

Inventor： Kui SHEN ,  Yingwei LI ,  Yonghai CAO ,  Junying CHEN

IPC: B01J31/16 ,  B01J37/06 ,  B01J37/02 ,  B01J37/00 ,  C07D233/58

Abstract: An ordered macroporous metal-organic framework single crystals and a preparation method therefor. In the method, a three-dimensional structure constructed by polymer microspheres is used as a template; 2-methylimidazole and zinc nitrate, precursors of MOFs, are firstly deposited in the three-dimensional template; the three-dimensional template containing the precursors is soaked in a mixed solution of ammonia water and methanol subsequently, and the three-dimensional template is taken out after crystallization; the three-dimensional template is soaked in an organic solvent to remove the macromolecular three-dimensional template, and the ordered macroporous MOF single crystals is obtained through centrifugal separation. The ordered macroporous MOF single crystals have a basic framework of zeolitic imidazolate framework-8, and structurally include highly-ordered macro-pores whose pore size may be controlled to be between 50 and 2000 nm based on a size of the used template.