delta.pipeline.ROI#
- class delta.pipeline.ROI(img_stack, fluo_stack, roi_nb, first_frame, box, config)[source]#
ROI processor object.
- __init__(img_stack, fluo_stack, roi_nb, first_frame, box, config)[source]#
Initialize ROI.
- Parameters
- img_stackList[Image]
Image stack.
- fluo_stackList[List[Image]]
Fluo stack. Shape: (frames, fluo_channels, Y, X).
- roi_nbint
ROI index.
- first_frameint
Index of the first frame (in general 0 or 1).
- boxCroppingBox
CroppingBox for ROI.
- configConfig
DeLTA configuration.
- Returns
- None.
Methods
__init__
(img_stack, fluo_stack, roi_nb, ...)Initialize ROI.
chunks_predict
(inputs, model[, batch_size])Run keras model, but by splitting the input data into "chunks".
compare
(other[, level])Compare this ROI with another and print the differences.
from_xarray
(dataset)Create a ROI from an
xarray.Dataset
.get_fluo
(frame)Return the ROI fluo images at a given frame.
get_img
(frame)Return the ROI image at a given frame.
get_labels
(frame)Return the ROI labels at a given frame.
get_seg
(frame)Return the ROI segmentation mask at a given frame.
get_segmentation_inputs
(frame)Compile segmentation inputs for ROI.
get_tracking_inputs
(frame)Compile tracking inputs for ROI from seg_stack.
load_netcdf
(filename[, group])Load a ROI from a netCDF file.
process_segmentation_outputs
(logits, frame)Process outputs after they have been segmented.
process_tracking_outputs
(logits, frame, boxes)Process output from tracking U-Net.
segment
(frames[, model])Segment img_stack and store the results in seg_stack.
to_netcdf
(filename, **kwargs)Save the ROI as a netCDF file.
to_xarray
()Convert the ROI into a xarray.Dataset.
track
(frames[, model, progress_bar])Track cells in the ROI.
Attributes
roi_nb
The ROI index number
box
ROI crop box
first_frame
Index of the first frame
img_stack
Input images stack
seg_stack
Segmentation images stack
lineage
Lineage object for ROI
label_stack
Labelled images stack
config
Configuration parameters object
scaling
Resizing ratios along Y and X