Cell fluorescence image discrimination method and system

Abstract

The invention discloses a cell fluorescence image discrimination method and system. The cell fluorescence image discrimination system comprises a format conversion module, an image binarization module, a fluorescence pattern position collection module, a cell fluorescence image cutting module, a cell classification module, a data set classification module, a sample equalization module, a data preprocessing module, a mathematical model construction module and a mathematical model training module. And a mathematical model test module and a detection module. According to the cell fluorescence image discrimination method and system provided by the invention, the efficiency and accuracy of fluorescence cell recognition can be improved. In a use scene of the invention, cell segmentation is carried out by using a fluorescence image in an HSV format, then positive and negative detection is carried out on segmented cells by using a deep learning network model, and finally positive cell quantity statistics is carried out, so that classification and recognition of the fluorescence staining image are realized.

Description

technical field [0001] The invention belongs to the technical field of intelligent learning, and relates to a cell recognition system, in particular to a cell fluorescence image discrimination method and system. Background technique [0002] Observing cell fluorescent slides under a microscope usually takes more than an hour through manual identification, and it is easy to cause visual fatigue; in addition, due to the subjective factors of various experts, the same slide may have different reviews result. [0003] Most of the existing artificial intelligence analysis methods in the industry simply perform threshold segmentation, then perform feature extraction, feature dimensionality reduction, and finally perform machine learning model classification. This method cannot obtain all the features of the entire fluorescent cell, resulting in low accuracy. [0004] In view of this, there is an urgent need to design a new cell identification method to overcome at least some of ...

AI Technical Summary

Problems solved by technology

This method cannot obtain all the features of the entire fluorescent cell, resulting in low accuracy

Images