Fix camera type issues

src/openflexure_microscope_server/things/camera/picamera.py

@@ -288,7 +288,7 @@ class StreamingPiCamera2(BaseCamera):
             "Sharpness": 1,
         }
 
-    _sensor_modes = None
+    _sensor_modes: Optional[list[SensorMode]] = None
 
     @lt.property
     def sensor_modes(self) -> list[SensorMode]:
@@ -363,6 +363,9 @@ class StreamingPiCamera2(BaseCamera):
                 camera_num=self._camera_num,
                 tuning=self.tuning,
             )
+            if self._picamera is None:
+                # Type narrow (error if failure)
+                raise RuntimeError("Failed to start Picamera")
             if check_sensor_model:
                 hw_sensor_model = self._picamera.camera_properties["Model"]
                 if hw_sensor_model != self._sensor_info.sensor_model:
@@ -861,7 +864,7 @@ class StreamingPiCamera2(BaseCamera):
         ]
 
     @lt.property
-    def lens_shading_tables(self) -> Optional[tf_utils.LensShading]:
+    def lens_shading_tables(self) -> Optional[tf_utils.LensShadingModel]:
         """The current lens shading (i.e. flat-field correction).
 
         Return the current lens shading correction, as three 2D lists each with
@@ -874,19 +877,6 @@ class StreamingPiCamera2(BaseCamera):
         """
         return tf_utils.get_lst(self.tuning)
 
-    @lens_shading_tables.setter
-    def _set_lens_shading_tables(self, lst: tf_utils.LensShading) -> None:
-        """Set the lens shading tables."""
-        with self._streaming_picamera(pause_stream=True):
-            self.tuning = tf_utils.set_lst(
-                self.tuning,
-                luminance=lst.luminance,
-                cr=lst.Cr,
-                cb=lst.Cb,
-                colour_temp=lst.colour_temp,
-            )
-            self._initialise_picamera()
-
     @lt.action
     def flat_lens_shading(self) -> None:
         """Disable flat-field correction.

src/openflexure_microscope_server/things/camera/picamera_recalibrate_utils.py

@@ -51,7 +51,6 @@ import numpy as np
 import picamera2
 from picamera2 import Picamera2
 from pydantic import BaseModel
-from scipy.ndimage import zoom
 
 LOGGER = logging.getLogger(__name__)
 
@@ -322,18 +321,7 @@ def _get_16x12_grid(chan: np.ndarray, dx: int, dy: int) -> np.ndarray:
     return np.reshape(np.array(grid), (12, 16))
 
 
-def _upsample_channels(grids: np.ndarray, shape: tuple[int]) -> np.ndarray:
-    """Zoom an image in the last two dimensions.
-
-    This is effectively the inverse operation of ``_get_16x12_grid``
-    """
-    zoom_factors = [
-        1,
-    ] + list(np.ceil(np.array(shape) / np.array(grids.shape[1:])))
-    return zoom(grids, zoom_factors, order=1)[:, : shape[0], : shape[1]]
-
-
-def _downsampled_channels(channels: np.ndarray, blacklevel: int) -> list[np.ndarray]:
+def _downsampled_channels(channels: np.ndarray, blacklevel: int) -> np.ndarray:
     """Generate a downsampled, un-normalised image from which to calculate the LST."""
     channel_shape = np.array(channels.shape[1:])
     lst_shape = np.array([12, 16])
@@ -398,9 +386,7 @@ def _grids_from_lst(lum: np.ndarray, Cr: np.ndarray, Cb: np.ndarray) -> np.ndarr
     return np.stack([B, G, G, R], axis=0)
 
 
-def _raw_channels_from_camera(
-    camera: Picamera2, sensor_info: SensorInfo
-) -> LensShadingTables:
+def _raw_channels_from_camera(camera: Picamera2, sensor_info: SensorInfo) -> np.ndarray:
     """Acquire a raw image and return a 4xNxM array of the colour channels."""
     if camera.started:
         camera.stop_recording()

src/openflexure_microscope_server/things/camera/picamera_tuning_file_utils.py

@@ -20,9 +20,12 @@ CALIBRATED_COLOUR_TEMP = 5000
 DEFAULT_COLOUR_TEMP = 1234
 
 
-class LensShading(BaseModel):
+class LensShadingModel(BaseModel):
     """A Pydantic model holding the lens shading tables.
 
+    Note this shouldn't be confused with the typehint for LensShadingTables in
+    recalibrate utils which is for the arrays.
+
     PiCamera needs three numpy arrays for lens shading correction. Each array is
     (12, 16) in size. The arrays are luminance, red-difference chroma (Cr), and
     blue-difference chroma (Cb).
@@ -165,7 +168,7 @@ def flatten_lst(tuning: dict, keep_luminance: bool = False) -> dict:
     )
 
 
-def get_lst(tuning: dict) -> LensShading:
+def get_lst(tuning: dict) -> LensShadingModel:
     """Return the lens shading as a LenSading Base Model."""
     # Note "alsc" is the Picamera2 term for "Automatic Lens Shading Correction"
     alsc = find_tuning_algo(tuning, "rpi.alsc")
@@ -175,7 +178,7 @@ def get_lst(tuning: dict) -> LensShading:
         w, h = 16, 12
         return [lin[w * i : w * (i + 1)] for i in range(h)]
 
-    return LensShading(
+    return LensShadingModel(
         luminance=reshape_lst(alsc["luminance_lut"]),
         Cr=reshape_lst(alsc["calibrations_Cr"][0]["table"]),
         Cb=reshape_lst(alsc["calibrations_Cb"][0]["table"]),

src/openflexure_microscope_server/things/camera/simulation.py

@@ -33,7 +33,7 @@ from . import BaseCamera
 
 LOGGER = logging.getLogger(__name__)
 
-# The ratio between "motor" steps and pixels
+# The ratio between "motor" steps and pixels in (x, y, z)
 # higher related to a faster movement
 RATIO = (2, 2, 0.2)
 
@@ -203,10 +203,11 @@ class SimulatedCamera(BaseCamera):
         """
         canvas_width, canvas_height, _ = self.canvas_shape
         image_width, image_height, _ = self.shape
-        pos = tuple(x * s for x, s in zip(pos, RATIO, strict=True))
+        # Scale position by RATIO to get position in base image.
+        im_pos = tuple(x * ratio for x, ratio in zip(pos, RATIO, strict=True))
         top_left = (
-            int(pos[0]) - image_width // 2 + self.sample_limits[0] // 2,
-            int(pos[1]) - image_height // 2 + self.sample_limits[1] // 2,
+            int(im_pos[0]) - image_width // 2 + self.sample_limits[0] // 2,
+            int(im_pos[1]) - image_height // 2 + self.sample_limits[1] // 2,
         )
         # Create index list with modulo rather than slicing to handle wrapping at the
         # canvas edge.
@@ -217,7 +218,7 @@ class SimulatedCamera(BaseCamera):
         # Use npx to make each 1d index list 3D
         focused_image = canvas[np.ix_(x_indices, y_indices, z_indices)]
 
-        image = fast_pil_blur(focused_image, sigma=np.abs(pos[2]) / 5)
+        image = fast_pil_blur(focused_image, sigma=np.abs(im_pos[2]) / 5)
 
         if image.shape != self.shape:
             raise ValueError(
@@ -251,7 +252,7 @@ class SimulatedCamera(BaseCamera):
         _traceback: Optional[TracebackType],
     ) -> None:
         """Close the capture thread when the Thing context manager is closed."""
-        if self.stream_active:
+        if self._capture_thread is not None and self._capture_thread.is_alive():
             self._capture_enabled = False
             self._capture_thread.join()
 
@@ -300,7 +301,7 @@ class SimulatedCamera(BaseCamera):
             try:
                 frame = self.generate_frame()
                 self.mjpeg_stream.add_frame(_frame2bytes(frame))
-                ds_frame = frame.resize((320, 240), resample=Image.NEAREST)
+                ds_frame = frame.resize((320, 240), resample=Image.Resampling.NEAREST)
                 self.lores_mjpeg_stream.add_frame(_frame2bytes(ds_frame))
 
             except Exception as e: