Microvascular segmentation method based on video jitter removal

By combining the technology of low-resolution preliminary matching and high-resolution precise matching in microvascular video processing, the problems of microvascular video jitter and contrast enhancement are solved, and the image stability and segmentation accuracy are improved.

CN119648724AActive Publication Date: 2025-03-18SHANGHAI DENDRITIC PRECISION INSTR CO LTD
6 Cites 0 Cited by

Patent Information

Application Number
CN202411769971.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-12-04
Publication Date
2025-03-18
Estimated Expiration
2044-12-04

AI Technical Summary

Technical Problem

The prior art is difficult to effectively eliminate the jitter problem in microvascular videos, and the existing image processing methods have limitations in enhancing contrast and segmenting microvascular details, resulting in low accuracy in microvascular lesions detection.

Method used

Microvascular fluorescence video was collected through microscope equipment, combined with low-resolution preliminary matching and high-resolution precise matching, and matched using normalized square square matching (NSSD) algorithm. After debounce processing, the images were input into the deep learning model for automatic segmentation.

Benefits of technology

It effectively eliminates the jitter phenomenon in the video, improves the stability and contrast of the image, makes the microvascular structure clearer and more obvious, improves the observation effect of the microvascular area, and improves the segmentation accuracy and robustness through automated image segmentation.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a capillary segmentation method based on video jitter removal, and relates to the technical field of image processing, and the method comprises the steps: collecting a capillary video, dividing the capillary video into n frames of images, carrying out the preprocessing of each frame of image, and generating processed image data; selecting a frame of image of the microvascular structure according to a division sequence of the image, marking a region of interest (ROI) containing the microvascular structure, and taking the ROI as a template image; scaling the template image and the current frame image to low resolution, and outputting a low-resolution matching result; returning the template image and the current frame image to high resolution, obtaining a high-resolution matching result according to the matching logic of the low-resolution matching result, and performing de-jitter processing to obtain a stable image; the stable image is used as input data, the region of interest containing the microvessels is output through the deep learning model to serve as a segmentation region, automatic image segmentation is carried out through the deep learning model, and the region of interest containing the microvessels is efficiently segmented.
Need to check novelty before this filing date? Find Prior Art