Image segmentation method, device, apparatus, and storage medium

By performing super-resolution segmentation and fusing the results of dual-path segmentation on medical images, the problem of low segmentation accuracy caused by blurred boundaries and low contrast in medical images is solved, thereby improving the recognition rate and reliability of the image segmentation model.

CN116894848BActive Publication Date: 2026-05-29LIANREN HEALTHCARE BIG DATA TECH CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
LIANREN HEALTHCARE BIG DATA TECH CO LTD
Filing Date
2023-08-28
Publication Date
2026-05-29

AI Technical Summary

Technical Problem

The blurry boundaries and low contrast of medical images result in low segmentation accuracy for AI image segmentation models.

Method used

By acquiring the image to be identified, super-resolution segmentation is performed using a simple linear iterative clustering algorithm to obtain the super-resolution segmented image to be identified. This super-resolution segmented image and the image to be identified are then input into a pre-trained target segmentation model to obtain dual-path segmentation results. The region of interest is determined based on these two results.

Benefits of technology

It improves the recognition rate and reliability of image segmentation models, reduces the learning difficulty of models, and performs fusion decision prediction through dual-modal input.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116894848B_ABST
    Figure CN116894848B_ABST
Patent Text Reader

Abstract

The application discloses an image segmentation method, device and equipment and a storage medium. The method comprises the following steps: obtaining an image to be recognized, performing super-resolution segmentation on the image to be recognized based on a simple linear iterative clustering algorithm to obtain a super-resolution segmentation image corresponding to the image to be recognized; inputting the image to be recognized and the super-resolution segmentation image into a pre-trained target segmentation model to obtain a first target segmentation result and a second target segmentation result, wherein the first target segmentation result is used for indicating the probability of each pixel point in the image to be recognized belonging to a region of interest, and the second target segmentation result is used for indicating the probability of each image segmentation region in the super-resolution segmentation image being the region of interest; and determining the region of interest in the image to be recognized based on the first segmentation result and the second segmentation result. The technical problem of low segmentation accuracy of an artificial intelligence image segmentation model caused by the characteristics of a blurred medical image boundary is solved.
Need to check novelty before this filing date? Find Prior Art