System and method for identifying and marking erythrocytes in peripheral blood

Abstract

The invention discloses a method and system for identifying and marking erythrocytes in peripheral blood, and the method comprises the following steps of 1) identifying an image to obtain all cells contained in the image; 2) obtaining the contour of the identified cell, calculating the roundness of the contour of the cell, fitting the long axis length and the edge position of an ellipse, and respectively marking all cells as large red blood cells, small red blood cells, platelets, edge cells, normal cells and abnormal cells; 3) displaying the contours, diameters and corresponding cell types ofall cells; 4) carrying out manual audit on the cell image, and if errors, label leakage or redundant labels exist, modifying, adding or deleting the cell image; and if abnormal cells and impurities are found, re-marking or deleting. The red blood cells are identified and marked by adopting the target detection technology, so that the problems of image missing and large calculated amount in the prior art are solved.

Description

technical field [0001] The invention relates to the field of cell identification, in particular to a system and method for identifying and marking red blood cells in peripheral blood. Background technique [0002] The patent application document with the application number of 201410337904.3 discloses a method for identifying red blood cells in feces. By performing reverse binarization on feces microscopic images, a type of impurity image is identified, its location is recorded, and then in the original image Exclude the impurity; then perform reverse binarization processing on the original image that excludes one impurity, and then identify an impurity to exclude, repeat this step to exclude multiple impurities and perform template matching, and the image required by the composite matching is identified as red blood cells. [0003] The disadvantages of the above method are: [0004] 1) Due to the influence of microscope illumination, the brightness of the image is not unif...

AI Technical Summary

Problems solved by technology

[0004] 1) Due to the influence of microscope illumination, the brightness of the image is not uniform. When many images are processed by binarization, the edges will be missing

[0005] 2) Due to the large amount of impurities in feces, it is necessary to obtain red blood cells by removing impurities one by one, resulting in a large amount of calculation in the real-time detection of red blood cells

Images