Self-adaptive automatic focusing method based on gray gradient value

Abstract

The invention provides a self-adaptive automatic focusing method based on a gray gradient value, relates to the automatic focusing problem of a microscope, and discloses the self-adaptive automatic focusing method based on the gray gradient value. The method comprises the following steps: firstly taking an out-of-focus blank image as a reference object, comparing the blank image with a current image to obtain an importance degree factor to measure calculation participation degree of difference values of gray values of pixel points; then evaluating whether the current image is clear according to the number of the difference values of the gray values of adjacent pixel points; then automatically positioning image focal points of the microscope by using a self-adaptive staged variable-step-size searching mode. Experiment tests show that a focusing evaluation algorithm has the advantages of low calculation complexity, good single-peak performance and good timeliness, and is capable of achieving rapid, accurate and automatic focusing of the microscope.

Description

technical field [0001] The invention designs an adaptive auto-focus method based on the gray gradient value. Background technique [0002] The traditional cytopathological diagnosis method is difficult to meet the needs of practical applications by collecting human cell samples, staining, and then observing under a microscope. With the development of technology, automatic film reading technology came into being. This technology automatically controls the movement and focus of the microscope, automatically analyzes and identifies abnormal cells after collecting images under the microscope, effectively assists doctors in diagnosis of diseases, and improves the sensitivity and specificity of doctors' diagnosis while reducing the labor intensity of doctors. [0003] DNA ploidy analysis technology uses image analysis and recognition technology to measure the relative content of DNA in the nucleus, and identify cells with DNA exceeding the normal range for further confirmation by...

Images