Image tagging method, device, computer system and readable storage medium

Abstract

The invention discloses an image labeling method, device, computer system and readable storage medium, comprising the following steps: receiving an image and performing grayscale processing on the image to generate a processed image; receiving a brush trajectory, and obtaining starting point coordinates from the brush trajectory , end point coordinates, and sampling points of the line, and store the sampling points in the line stack; draw a straight line from the starting point coordinates to the end point coordinates to generate a closed line according to the starting point coordinates, end point coordinates, line sampling points and closing rules, and extract the line's Line sampling points are stored in the line stack, and the sampling points in the line stack are output to the client; or according to the starting point coordinates, end point coordinates, line sampling points and closure rules, the line is set as an unclosed line, and the line The sampling points in the stack are output to the client; receive line selection information and line movement information, and adjust the position of the line. The invention improves the labeling efficiency and the accuracy of labeling the lesion area.

Description

technical field [0001] The invention relates to the field of communication technology, in particular to an image labeling method, device, computer system and readable storage medium. Background technique [0002] At present, when doctors mark the lesion area of ​​the image, they usually use the brush function of the common drawing tool to draw lines on the outer contour of the lesion area, so as to achieve the purpose of marking the lesion area; [0003] However, the current brush function only supports drawing lines on the image, but the labeling of the lesion area usually requires the end point of the brush to coincide with the starting point, so that the doctor can only complete it by zooming in on the image and slowly moving the brush to its starting point; Not only is the efficiency low, but it also brings great inconvenience to the doctor and increases the fatigue strength of the doctor; at the same time, because once the line is drawn, only a single boundary point can...

AI Technical Summary

Problems solved by technology

[0003] However, the current brush function only supports drawing lines on the image, but the labeling of the lesion area usually requires the end point of the brush to coincide with the starting point, so that the doctor can only complete it by zooming in on the image and slowly moving the brush to its starting point; Not only is the efficiency low, but it also brings great inconvenience to the doctor and increases the fatigue strength of the doctor; at the same time, because once the line is drawn, only a single boundary point can be moved, resulting in uneven boundaries, thus reducing the need for Labeling accuracy of lesion area

Images