Fig 1.
Sample HE stained images from different organs in MoNuSeg [9] dataset, in which high quality stained images are shown in the first row, and the second row contains low quality samples.
Table 1.
Sample images from multiple datasets for image segmentation.
Fig 2.
Construction of sample patches and corresponding pseudo labels for training.
Fig 3.
Comparison between mixing based on RGB and HSV space.
Fig 4.
Workflow of image segmentation through FGDC-net.
Fig 5.
Information transmission flows in FGDC modules.
Fig 6.
Details of ResBlock module.
Table 2.
Division of original images for training and testing datasets.
Table 3.
Division of original images for training and testing datasets.
Table 4.
Ablation experimental results of multiple network structures.
Table 5.
Ablation experimental results of multiple training strategies.
Table 6.
Comparison experimental results on Kumar dataset.
Fig 7.
Samples segmentation results of multiple organs from three datasets.
Table 7.
Comparison experimental results on MoNuSeg dataset.