stereo.tools.cell_cut.CellCut.cell_cut#
- CellCut.cell_cut(bgef_path=None, gem_path=None, mask_path=None, image_path=None, model_path=None, mask_save=True, model_type='deep-learning', depp_cro_size=20000, overlap=100, gen_mask_on_gpu='-1', tissue_seg_model_path=None, tissue_seg_method=None, post_processing_workers=10)[source]#
- Generate CGEF resutl via following combinations:
BGEF and mask
BGEF and ssDNA image
- Parameters:
bgef_path (
Optional
[str
]) – the path to BGEF file.gem_path (
Optional
[str
]) – the path to GEM file.mask_path (
Optional
[str
]) – the path to mask file.image_path (
Optional
[str
]) – the path to ssDNA image file.model_path (
Optional
[str
]) – the path to model file.mask_save (
bool
) – whether to save mask file after correction, generated from ssDNA image.model_type (
str
) – the type of model to generate mask, whcih only could be set to deep learning model and deep cell model.depp_cro_size (
int
) – deep crop size.overlap (
int
) – overlap size.gen_mask_on_gpu (
str
) – specify gpu id to predict when generate mask, if'-1'
, use cpu for prediction.tissue_seg_model_path (
Optional
[str
]) – the path of deep-learning model of tissue segmentation, if set it to None, it would use OpenCV to process.tissue_seg_method (
Optional
[str
]) – the method of tissue segmentation, 0 is deep-learning and 1 is OpenCV.post_processing_workers (
int
) – the number of processes for post-processing.
- Returns:
Path to CGEF result.