Title:
FEATURE DISCRETIZATION METHOD FOR REMOTE SENSING IMAGE ON THE BASIS OF ROUGH FUZZY MODEL
Document Type and Number:
WIPO Patent Application WO/2022/127075
Kind Code:
A1
Abstract:
The invention provides a feature discretization method for a remote sensing image on the basis of a rough fuzzy model. Said method comprises the following steps: listing the image element values and categories of a selected sample in various wavebands in a remote sensing image, and building an image information decision table based on the image element values and the categories; initializing a category center of each category and the membership degree between a sample instance and the category center; iteratively updating the category center of each category and the membership degree between the sample instance and the category center, so as to obtain the final value of the category center of each category and the final value of the membership degree; and building a rough fuzzy set, calculating the average approximate precision of the rough fuzzy set, discretizing the image information decision table, evaluating the discretization result on the basis of the average approximate precision and a genetic algorithm, and selecting an optimal discretization scheme.
Inventors:
HUANG MENGXING (CN)
CHEN QIONG (CN)
FENG SILING (CN)
FENG WENLONG (CN)
ZHANG YU (CN)
WU DI (CN)
CHEN QIONG (CN)
FENG SILING (CN)
FENG WENLONG (CN)
ZHANG YU (CN)
WU DI (CN)
Application Number:
PCT/CN2021/103353
Publication Date:
June 23, 2022
Filing Date:
June 30, 2021
Export Citation:
Assignee:
UNIV HAINAN (CN)
International Classes:
G06K9/00
Foreign References:
| CN112580483A | 2021-03-30 | |||
| CN107909062A | 2018-04-13 | |||
| CN104268532A | 2015-01-07 | |||
| CN106971204A | 2017-07-21 | |||
| US10489713B1 | 2019-11-26 |
Attorney, Agent or Firm:
HAINAN HEIP INTELLECTUAL PROPERTY AGENCY CO., LTD (CN)
Download PDF: