Distinguishing Method of Fluorescence Microscopy Image Result

Abstract

The invention discloses a method for discriminating the results of fluorescence microscope images. First, color-separate the red, green and blue channels of a sample image taken by a fluorescence microscope to obtain a red channel image, a green channel image, and a blue channel image; Perform grayscale processing and binary processing on the red and green channel images to obtain black and white binary images R1 and G1; 3. Extract the contours of the binary images R1 and G1 respectively to obtain multiple irregular polygonal images. Multiple irregular polygonal images are filtered; 4. When the number of cell images presented on the filtered binary image G1 is ≥ 1, and the presented cell images are similar to the cell images presented on the filtered binary image R1 When overlapping, it is judged that the sample image is positive; otherwise, it is judged that the sample image is negative. The invention has the advantages of realizing program batching and intelligently distinguishing negative and positive of tested samples, which greatly improves the working efficiency of doctors.

Description

technical field [0001] The invention relates to fluorescence microscope images, in particular to a method for judging the results of fluorescence microscope images. Background technique [0002] During the detection of pathogen markers, the sample is diluted and centrifuged, then placed in the detection instrument, and the processing steps such as slide loading, incubation, washing, and drying are automatically completed, and finally the slide is placed under a fluorescence microscope to take images , there are multiple holes (usually ten holes) on a glass slide, about ten fluorescence images need to be taken for each hole, and through image recognition of the ten fluorescence images of each hole, it is judged that the sample on each image is In the end, the judgment results of the ten images are summed up to obtain the negative and positive values ​​of the samples in the hole. At present, the recognition of fluorescent images is mostly done by doctors through naked eye obs...

AI Technical Summary

Problems solved by technology

At present, the recognition of fluorescent images is mostly done by doctors through naked eye observation to determine whether the samples are negative or positive. This not only has human-made visual errors, doctors' work experience and other factors, resulting in low accuracy of image recognition and judgment, but also the labor of doctors. High strength, low work efficiency

Images