Image processing method, device, user terminal, server and storage medium

Abstract

The embodiment of the invention provides a picture processing method, a device, a user terminal, a server and a storage medium, wherein, the method can include: obtaining an original pathological picture to be processed, a target pathological picture corresponding to the original pathological picture and a slice area of the target pathological picture; obtaining a target pathological picture corresponding to the target pathological picture. Sending the target pathological picture and the slice area of the target pathological picture to a target user terminal matched with the original pathological picture, so that the target user terminal labels the label data of the target pathological picture in the slice area of the target pathological picture; receiving annotation data of the target pathology picture submitted by the target user terminal; converting the annotation data of the target pathological picture to obtain the annotation data of the original pathological picture. The invention can realize the labeling of the pathological picture, and can improve the labeling efficiency and the accuracy of the pathological picture.

Description

technical field [0001] The present invention relates to the technical field of the Internet, specifically to the technical field of image processing, and in particular to an image processing method, an image processing device, a user terminal for image processing, a server for image processing, and a computer storage media. Background technique [0002] Annotation is a crucial work in the data preparation stage in the field of machine learning. The purpose of labeling pictures is to find entities that can be used for subsequent machine learning from the pictures, for example: a common picture with a cup is taken, and the labeling of the picture is to determine the labeling object in the picture, which is the cup, and in the picture Mark the coordinates of the cup, and also mark some attributes of the cup, such as shape, height, width, color and so on. In the prior art, the picture labeling scheme is mainly aimed at ordinary pictures, and the characteristics of this type of...

AI Technical Summary

Problems solved by technology

At present, there is no specific labeling scheme for pathological pictures. If the above-mentioned existing picture labeling schemes are directly applied to the field of pathological picture labeling, there will be major technical defects: due to the various private formats, large sizes and low resolution of pathological pictures Advanced features make pathological images unsuitable for direct loading and labeling. Most of the current terminal devices cannot load and run pathological images. The concurrency of the terminal is also extremely limited, so the annotation of pathological images cannot be achieved by using the existing image annotation scheme

Images