chagned training resolution to dictionaries

Author: lufre1

synaptic_reconstruction/tools/distance_measure_widget.py

@@ -105,11 +105,7 @@ def on_measure_seg_to_object(self):
         resolution = segmentation.shape
         print("on_measure_seg_to_object resolution", resolution)
         # get image metadata
-        scale = None
-        if "Image data" in self.viewer.layers:
-            if "voxel_size" in self.viewer.layers["Image data"].metadata.keys():
-                voxel_size = self.viewer.layers["Image data"].metadata["voxel_size"]
-                scale = compute_scale_from_voxel_size(voxel_size)
+
         (distances,
          endpoints1,
          endpoints2,
@@ -122,7 +118,6 @@ def on_measure_seg_to_object(self):
             endpoints1=endpoints1,
             endpoints2=endpoints2,
             seg_ids=seg_ids,
-            scale=scale
         )
         table_data = self._to_table_data(distances, seg_ids, endpoints1, endpoints2)
         self._add_lines_and_table(lines, properties, table_data, name="distances")
@@ -134,12 +129,7 @@ def on_measure_pairwise(self):
             return
         resolution = segmentation.shape
         # get image metadata
-        scale = None
-        if "Image data" in self.viewer.layers:
-            print("image data in viewer layers")
-            if "voxel_size" in self.viewer.layers["Image data"].metadata.keys():
-                voxel_size = self.viewer.layers["Image data"].metadata["voxel_size"]
-                scale = compute_scale_from_voxel_size(voxel_size)
+
         (distances,
          endpoints1,
          endpoints2,
@@ -148,7 +138,6 @@ def on_measure_pairwise(self):
         )
         lines, properties = distance_measurements.create_pairwise_distance_lines(
             distances=distances, endpoints1=endpoints1, endpoints2=endpoints2, seg_ids=seg_ids.tolist(),
-            scale=scale
         )
         table_data = self._to_table_data(
             distances=properties["distance"],

synaptic_reconstruction/tools/segmentation_widget.py

@@ -74,24 +74,17 @@ def on_predict(self):
         # TODO: Use scale derived from the image resolution.
         # get avg image shape from training of the selected model
         # wichmann data avg voxel size = 17.53
-        training_voxel_size = {
-            "x": 17.48,
-            "y": 17.48,
-            "z": 17.48
-        }
+
         metadata = self._get_layer_selector_data(self.image_selector_name, return_metadata=True)
-        if "voxel_size" in metadata.keys():
-            voxel_size = metadata["voxel_size"]
-        else:
-            voxel_size = None
+        voxel_size = metadata.get("voxel_size", None)
 
         if self.scale_param.value() != 1.0:  # changed from default
             scale = []
             for k in range(len(image.shape)):
                 scale.append(self.scale_param.value())
         elif voxel_size:
             # calculate scale so voxel_size is the same as in training
-            scale = compute_scale_from_voxel_size(voxel_size)
+            scale = compute_scale_from_voxel_size(voxel_size, model_type)
         else:
             scale = None
         print(f"Rescaled the image by {scale} to optimize for the selected model.")

synaptic_reconstruction/tools/util.py

@@ -10,13 +10,6 @@
 from ..inference.mitochondria import segment_mitochondria
 
 
-_TRAINING_VOXEL_SIZE = {
-            "x": 17.48,
-            "y": 17.48,
-            "z": 17.48
-        }
-
-
 def get_model(model_type: str, device: Optional[Union[str, torch.device]] = None) -> torch.nn.Module:
     """Get the model for the given segmentation type.
 
@@ -86,12 +79,12 @@ def get_cache_dir():
 
 def get_model_training_resolution(model_type):
     resolutions = {
-        "active_zone": 1.44,
-        "compartments": 3.47,
+        "active_zone": {"x": 1.44, "y": 1.44, "z": 1.44},
+        "compartments": {"x": 3.47, "y": 3.47, "z": 3.47},
         "mitochondria": 1.0,  # FIXME: this is a dummy value, we need to determine the real one
-        "vesicles_2d": 1.35,
-        "vesicles_3d": 1.35,
-        "vesicles_cryo": 0.88,
+        "vesicles_2d": {"x": 1.35, "y": 1.35},
+        "vesicles_3d": {"x": 1.35, "y": 1.35, "z": 1.35},
+        "vesicles_cryo": {"x": 1.35, "y": 1.35, "z": 0.88},
     }
     return resolutions[model_type]
 
@@ -233,8 +226,9 @@ def compute_average_voxel_size(voxel_size: dict) -> float:
 
 def compute_scale_from_voxel_size(
     voxel_size: dict,
-    training_voxel_size=_TRAINING_VOXEL_SIZE
+    model_type: str
 ) -> List[float]:
+    training_voxel_size = get_model_training_resolution(model_type)
     scale = [
         voxel_size["x"] / training_voxel_size["x"],
         voxel_size["y"] / training_voxel_size["y"],

synaptic_reconstruction/tools/volume_reader.py

@@ -75,17 +75,19 @@ def read_image_volume(path: PathOrPaths) -> List[LayerData]:
 
 
 def read_voxel_size(input_path: str, layer_attributes: dict) -> None:
-    with mrcfile.open(input_path) as mrc:
+    """_summary_
+
+    Args:
+        input_path (str): _description_
+        layer_attributes (dict): _description_
+    """
+    with mrcfile.open(input_path, permissive=True) as mrc:
         try:
             voxel_size = mrc.voxel_size
             layer_attributes["voxel_size"] = {
-                "x": voxel_size.x,
-                "y": voxel_size.y,
-                "z": voxel_size.z,
+                "x": voxel_size.x / 10,
+                "y": voxel_size.y / 10,
+                "z": voxel_size.z / 10,
             }
         except Exception as e:
             print(f"Failed to read voxel size: {e}")
-        # return mrc.voxel_size
-
-# if __name__ == "__main__":
-#     read_image_volume("/home/freckmann15/data/mitochondria/fidi_orig/20240722_WT/36859_J1_66K_TS_PS_01_rec_2kb1dawbp_crop_raw.mrc")