Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
VOXEL MODEL-BASED CHARACTERIZATION METHOD FOR RESPIRATORY CHARACTERISTICS
Document Type and Number:
WIPO Patent Application WO/2021/077515
Kind Code:
A1
Abstract:
A voxel model-based characterization method for respiratory characteristics, comprising: using a camera to continuously collect multi-frame depth images of thoracic and abdominal body surfaces of a human body, and performing three-dimensional modeling on the multi-frame depth images to obtain a multi-frame voxel model in a time series; traversing voxel units of the multi-frame voxel model to extract volume features and area features of the multi-frame voxel model; acquiring a common smallest voxel bounding box of the multi-frame voxel model; describing the spatial distribution of the multi-frame voxel model in the form of probability, and arranging the probabilities of the smallest voxel bounding box of each frame of the voxel model to form a sample space of an ultra-high-dimensional vector; reducing the dimensionality of the sample space to obtain essential parameters after dimensionality reduction; and according to the essential parameters, the volume features and the area features, obtaining feature variables that are able to characterize the voxel model. The described method is able to comprehensively characterize body surface breathing motion information, and is widely used.

Inventors:
YU SHUMEI (CN)
HOU PENGCHENG (CN)
SUN RONGCHUAN (CN)
KUANG SHAOLONG (CN)
SUN LINING (CN)
Application Number:
PCT/CN2019/119693
Publication Date:
April 29, 2021
Filing Date:
November 20, 2019
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
UNIV SOOCHOW (CN)
International Classes:
G16H20/40; G06T17/00; G16H30/20
Domestic Patent References:
WO2018172229A12018-09-27
Foreign References:
CN104395933A2015-03-04
CN107403407A2017-11-28
CN109069858A2018-12-21
CN103761745A2014-04-30
US20180315188A12018-11-01
Attorney, Agent or Firm:
CENTRAL SOUTH WELL INTELLECTUAL PROPERTY OFFICE (CN)
Download PDF: