OBJECT-ORIENTED FUZZY CLUSTERING IMAGE SEGMENTATION METHOD AND SYSTEM, TERMINAL, AND STORAGE MEDIUM

Title:

OBJECT-ORIENTED FUZZY CLUSTERING IMAGE SEGMENTATION METHOD AND SYSTEM, TERMINAL, AND STORAGE MEDIUM

Document Type and Number:

WIPO Patent Application WO/2023/109619

Kind Code:

A1

Abstract:

Provided in the present application are an object-oriented fuzzy clustering image segmentation method and system, a terminal, and a storage medium. Several generation points are randomly determined in an image domain to divide the image domain into several irregular regions, so as to form an object; a feature field Markov model of a pixel with respect to a neighborhood pixel is constructed to depict the fuzzy non-similarity between the object and a cluster; an a priori probability of the object with respect to the cluster is constructed on the basis of a label field; an object-oriented fuzzy clustering objective function based on a feature field MRF is accordingly constructed; the generation points are optimized by means of using objective function minimization as a criterion in combination with an MH algorithm; and the coverage range of the object is adjusted, so as to further accurately fit the boundaries of homogeneous regions and obtain an optimal segmentation result. By means of the object-oriented fuzzy clustering image segmentation method and system, the terminal, and the storage medium provided in the present application, a mis-segmentation problem caused by the similarity in spectral features between objects is avoided and the relatively strong noise resistance of an algorithm is ensured, thereby further improving the image segmentation precision.

More Like This:

WO/2021/121631	CORRECTING SEGMENTATION OF MEDICAL IMAGES USING A STATISTICAL ANALYSIS OF HISTORIC CORRECTIONS
JP2020087037	IDENTIFICATION APPARATUS AND IDENTIFICATION METHOD
WO/2017/163759	SYSTEM AND COMPUTER-IMPLEMENTED METHOD FOR SEMANTIC SEGMENTATION OF IMAGE, AND NON-TRANSITORY COMPUTER-READABLE MEDIUM

Inventors:

LI XIAOLI (CN)
ZHAO LONGLONG (CN)
CHEN JINSONG (CN)
HAN YU (CN)
JIANG XIAOLI (CN)

Application Number:

PCT/CN2022/137366

Publication Date:

June 22, 2023

Filing Date:

December 07, 2022

Export Citation:

Click for automatic bibliography generation Help

Assignee:

SHENZHEN INST ADV TECH (CN)

International Classes:

G06T7/143; G06T7/11; G06T7/46; G06V10/762

Foreign References:

CN114219822A	2022-03-22
CN112116610A	2020-12-22
US20050111730A1	2005-05-26

Other References:

LI XIAOLI: "Remote sensing image segmentation with regionalized fuzzy clustering based on Voronoi tessellation", ACTA GEODAETICA ET CARTOGRAPHICA SINICA, vol. 49, no. 6, 1 June 2020 (2020-06-01), pages 799, XP093072660, DOI: 10.11947/j.AGCS.2020.20190328

Attorney, Agent or Firm:

BEIJING ZHONG XUN TONG DA INTELLECTUAL PROPERTY AGENCY CO., LTD. (CN)

Download PDF:

View/Download PDF PDF Help

Previous Patent: ANTI-PANCREATIC CANCER POLYPEPTIDE DRUG CONJUGATE, AND CARRIER-FREE SELF-ASSEMBLY NANO DRUG THEREOF,...

Next Patent: STACKER