Cell image segmentation tracing method based on U-shaped residual neural network

Abstract

The invention discloses a cell image segmentation tracing method based on a U-shaped residual neural network, and the method comprises the following steps: carrying out the preprocessing of a cell image, and recording the position of a cell nucleus according to a cell data set; generating a moving position vector, and marking a cell nucleus position and a time sequence change trend in the cell image; segmenting a cell image and predicting a moving position vector, and simultaneously performing feature extraction and prediction on the cell image and the moving position vector; post-processing the moving position vector, reducing over-predicted positions, and interpolating undetected positions; and tracking the segmentation result by the cell, and obtaining a final result by using a watershed algorithm on the segmentation result and the predicted position. According to the cell image tracing method, an efficient and easy-to-use segmentation mode is provided, accurate tracing of the cell image on a time sequence is realized, meanwhile, the cell segmentation capability is improved, the time of manual labeling of continuous frame cells is shortened, and overall observation in the cell division and differentiation process in biology is facilitated.

Description

technical field [0001] The invention belongs to the technical field of image processing and computer vision, in particular to a cell image segmentation and tracing method based on a U-shaped residual neural network. Background technique [0002] With the continuous development of microscopic cell image recognition in the field of medical image processing, the research on medical cell image processing methods can effectively find out the cause of pathology, facilitate the right medicine and the development of specific drugs. At the same time, cell image segmentation, as a key technology for medical cell image analysis and processing, is also an important part of machine vision. [0003] With the improvement of image processing technology and computer vision technology, the accuracy of cell segmentation has been continuously improved. Based on the identification and segmentation of cell outlines, new requirements have been put forward for the identification and positioning of ...

AI Technical Summary

Problems solved by technology

The invention solves the problem of low segmentation accuracy in the case of cell overlap and adhesion, and also performs high-precision tracking of cell division and cell disappearance events, and improves the calculation efficiency of cell tracking. The vector is filtered and analyzed, thereby avoiding the uncertainty of manual labeling, and at the same time, it has a high accuracy rate for various cell segmentation traces

Images