Merge pull request #19 from brown-ccv/feat-combine-binary-mask

[Feat] Combine binary masks

jaworski_cellseg/binary_maps_widget.py

@@ -5,6 +5,15 @@
 from skimage.morphology import remove_small_objects
 
 
+def create_binary_map(cloud_channel, binary_map_threshold):
+    cloud_percentile_float = np.percentile(
+        cloud_channel.astype(np.float64), binary_map_threshold
+    )
+
+    # Create a binary mask using the percentile as a threshold
+    return cloud_channel > cloud_percentile_float
+
+
 def create_binary_map_widget(
     viewer: napari.Viewer,
 ):
@@ -20,23 +29,40 @@ def create_binary_map_widget(
             "widget_type": "ComboBox",
             "choices": [],
         },
-        binary_map_threshold={"label": "Threshold", "min": 0.0, "max": 100.0},
+        binary_map_1_threshold={"label": "Threshold", "min": 0.0, "max": 100.0},
+        channel_2_name={
+            "widget_type": "ComboBox",
+            "choices": [],
+        },
+        binary_map_2_threshold={"label": "Threshold", "min": 0.0, "max": 100.0},
     )
     def binary_map_widget(
         channel_name: Optional[str] = None,
-        binary_map_threshold: float = 90.0,
+        binary_map_1_threshold: float = 90.0,
+        channel_2_name: Optional[str] = None,
+        binary_map_2_threshold: float = 90.0,
     ):
         cloud_channel = viewer.layers[channel_name].data
-        cloud_percentile_float = np.percentile(
-            cloud_channel.astype(np.float64), binary_map_threshold
+        binary_mask = create_binary_map(cloud_channel, binary_map_1_threshold)
+        viewer_channel_name = (
+            f"{channel_name} binary mask perc={binary_map_1_threshold}"
         )
 
-        # Create a binary mask using the percentile as a threshold
-        binary_mask = cloud_channel > cloud_percentile_float
+        # If a second channel is selected, combine the binary masks
+        if channel_2_name and channel_2_name in viewer.layers:
+            cloud_channel_2 = viewer.layers[channel_2_name].data
+            binary_mask_2 = create_binary_map(cloud_channel_2, binary_map_2_threshold)
+
+            # Combine the two masks (logical AND)
+            binary_mask = np.logical_and(binary_mask, binary_mask_2)
+            viewer_channel_name = (
+                f"Combinend {channel_name}-{channel_2_name}"
+                f"binary mask"
+                f"perc={binary_map_1_threshold}-{binary_map_2_threshold}"
+            )
+
         binary_mask = remove_small_objects(binary_mask, min_size=100)
-        viewer.add_image(
-            binary_mask, name=f"{channel_name} binary mask perc={binary_map_threshold}"
-        )
+        viewer.add_image(binary_mask, name=viewer_channel_name)
 
     def update_channel_names_dropdown(event=None):
         channel_layer_names = []
@@ -45,6 +71,7 @@ def update_channel_names_dropdown(event=None):
                 channel_layer_names.append(layer.name)
 
         binary_map_widget.channel_name.choices = channel_layer_names
+        binary_map_widget.channel_2_name.choices = channel_layer_names
 
     viewer.layers.events.inserted.connect(update_channel_names_dropdown)
     return binary_map_widget