Skip to contents

Fits a layout-initialized mixture to colour and texture features relative to agar. Expected target positions seed one component; the centre seeds the dense partner core and an adjacent region seeds peripheral central growth. For a fungal partner this peripheral growth represents candidate hyphal tissue, not inhibition or a diffusible pigment boundary. All image classes require visual review; the photograph does not identify species or resolve individual microscopic hyphae. Similar colours or overlapping cultures may remain ambiguous. Seeded segmentation separates the target colonies later.

Usage

classify_pixels(
  plate,
  layout = plate_layout(),
  classifier = NULL,
  posterior_smooth_mm = 0.15,
  context_mm = 0.5,
  max_em_pixels = 60000L
)

Arguments

plate

A plate processed by model_background().

layout

A plate_layout().

classifier

Optional funmel_classifier from train_classifier().

posterior_smooth_mm

Smoothing of class posteriors for confident pixels (mm).

context_mm

Neighbourhood (mm) from which low-confidence (faint) pixels borrow their class.

max_em_pixels

Maximum number of pixels used to fit the mixture.

Value

The plate with element classes: an integer matrix (0 agar/outside, 1 target-like tissue, 2 central core, 3 peripheral central growth, 4 other) and the fitted model.

Details

The internal labels fungus and bacteria denote target-like and central-core components; the biological partner identity comes from the layout.

Examples

sim <- simulate_plate(width = 300, height = 400, seed = 4)
p <- read_plate(sim$image) |> detect_plate() |> model_background()
p <- classify_pixels(p, plate_layout())
table(p$classes$map)
#> 
#>     0     1     2     3     4 
#> 91861 12415  1749 13869   106