cellmap_flow.ai_annotate.mask_extraction

From an EM plane to the RGB image a model is sent, and from its reply to a mask.

Adapted from ask-to-mask (_normalize_to_uint8/_slice_to_rgb and extract_mask’s colour-difference mode): the model is asked to paint the structure one colour and everything else black, and the mask is the pixels that came back that colour. The arithmetic is cellmap_flow.portable.ai_annotate’s (normalize_to_uint8, extract_mask; model_mask lines a reply up with the plane first), which the browser runs too; here are the Pillow images the server’s backends and staging pass around.

Functions

slice_to_rgb(data_2d) → PIL.Image.Image	A 2D plane as a grey RGB image, the form image models take.
extract_mask(output_image, target_rgb[, threshold]) → numpy.ndarray	The pixels of the model's reply (a Pillow image or an RGB array)

Module Contents

cellmap_flow.ai_annotate.mask_extraction.slice_to_rgb(data_2d: numpy.ndarray) → PIL.Image.Image

A 2D plane as a grey RGB image, the form image models take.

cellmap_flow.ai_annotate.mask_extraction.extract_mask(output_image, target_rgb: tuple[int, int, int], threshold: float = 200.0) → numpy.ndarray

The pixels of the model’s reply (a Pillow image or an RGB array) painted target_rgb, as a bool (H, W) mask (portable.ai_annotate.extract_mask says how a pixel is scored).