Merge pull request #21 from LaboratoryOpticsBiosciences/value_filter

Author: ClementCaporal

src/napari_brainbow_diagnose/_utils_channel_space.py

@@ -67,7 +67,7 @@ def get_channels_ranges(a: np.ndarray) -> np.ndarray:
 
 
 def image_mask_of_wheel_selection(
-    image: np.ndarray, wheel_selection: np.ndarray
+    image: np.ndarray, wheel_selection: np.ndarray, value_threshold: float
 ):
     """
     Returns a boolean mask of the wheel selection.
@@ -79,6 +79,9 @@ def image_mask_of_wheel_selection(
         Shape: (c, z, y, x) or (c, y, x)
     wheel_selection : np.ndarray
         The wheel selection to apply.
+    value_threshold : float
+        The value threshold to apply to the image voxel.
+        Only voxels with a value above the threshold will be selected.
 
     Returns
     -------
@@ -103,6 +106,9 @@ def image_mask_of_wheel_selection(
     selection_mask = selected_points.reshape(image.shape[1:])
     selection_mask = selection_mask.astype(bool)
 
+    print(selection_mask.shape, (hsv[2] < value_threshold).shape)
+    selection_mask[hsv[2] < value_threshold] = False
+
     return selection_mask
 
 

src/napari_brainbow_diagnose/_widget.py

@@ -97,7 +97,12 @@ def create_mask_on_image(self):
             bool
         )
         mask_corrected = self.density_figure.selection_mask.astype(bool)
-        mask_on_image = image_mask_of_wheel_selection(channels, mask_corrected)
+
+        value_threshold = self.density_resolution_widget.value_threshold.value
+
+        mask_on_image = image_mask_of_wheel_selection(
+            channels, mask_corrected, value_threshold
+        )
 
         self.viewer.add_labels(mask_on_image, name="mask_on_image")