Cell image segmentation method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention discloses a cell image segmentation method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring a to-be-segmented cell image; inputting the to-be-segmented cell image into a preset generative adversarial network to obtain a mask image; obtaining an instance center mark of the to-be-segmented cell image according to the mask image; and according to the mask image and the instance center mark, based on a preset image segmentation algorithm, obtaining a target segmentation image of the to-be-segmented cell. The to-be-segmented cell image is input into the preset generative adversarial network model, the mask image of the to-be-segmented cell image is automatically generated, and the cell nucleus of the to-be-segmented cell image is labeled, so that cell nucleus segmentation is performed on the to-be-segmented cell image, manual cell nucleus labeling on the to-be-segmented cell image is avoided, manual operation steps are reduced, and the cell segmentation efficiency and precision are improved.

Description

technical field [0001] Embodiments of the present invention relate to image processing technology, and in particular to a cell image segmentation method, device, electronic equipment, and storage medium. Background technique [0002] CLSM (Confocal laser scanning microscope) images contain a large amount of information on cell and tissue structures, and are widely used in evaluating three-dimensional cell cultures based on biological scaffolds. Since the morphology, distribution, nucleoplasmic ratio, and structure of the nucleus are the basis for studying subcellular information, the segmentation of the nucleus in 3D cell culture is an important task in the study and analysis of 3D cell culture. [0003] The traditional method of cell nucleus segmentation is supervised learning using human-labeled data sets, which takes a lot of time to label mask images, and the labeling process varies from person to person, and the obtained data sets have expert bias, so the segmentation e...

AI Technical Summary

Problems solved by technology

[0003] The traditional method of cell nucleus segmentation is supervised learning using human-labeled data sets, which takes a lot of time to label mask images, and the labeling process varies from person to person, and the obtained data sets have expert bias, so the segmentation effect of this type of method is limited

And it requires a lot of manpower, it is easy to cause over-segmentation or under-segmentation, and the segmentation efficiency and accuracy of cell images are low

Images