From 8ecc14be9d9c7008e02635fd403c299aab07b29c Mon Sep 17 00:00:00 2001
From: Christopher Rhodes <christopher.rhodes@embl.de>
Date: Tue, 16 Jul 2024 19:06:21 +0200
Subject: [PATCH] Generally removing necessity of instantiating with a map of
object IDs
---
model_server/base/roiset.py | 187 ++++++++++++++++++++++--------------
tests/base/test_roiset.py | 6 +-
2 files changed, 118 insertions(+), 75 deletions(-)
diff --git a/model_server/base/roiset.py b/model_server/base/roiset.py
index 36f9e066..6afa4b1a 100644
--- a/model_server/base/roiset.py
+++ b/model_server/base/roiset.py
@@ -108,6 +108,19 @@ def _focus_metrics():
'moment': lambda x: moment(x.flatten(), moment=2),
}
+
+def _filter_df(df: pd.DataFrame, filters: RoiFilter = None) -> pd.DataFrame:
+ query_str = 'label > 0' # always true
+ if filters is not None: # parse filters
+ for k, val in filters.dict(exclude_unset=True).items():
+ assert k in ('area')
+ vmin = val['min']
+ vmax = val['max']
+ assert vmin >= 0
+ query_str = query_str + f' & {k} > {vmin} & {k} < {vmax}'
+ return df.loc[df.query(query_str).index, :]
+
+
# TODO: get overlapping bounding boxes
def _filter_overlap_bbox(df: pd.DataFrame) -> pd.DataFrame:
@@ -128,9 +141,50 @@ def _filter_overlap_bbox(df: pd.DataFrame) -> pd.DataFrame:
sdf['overlaps_with'] = second
return sdf
+def _make_df_from_object_ids(acc_raw, acc_obj_ids, expand_box_by) -> pd.DataFrame:
+ """
+ Build dataframe associate object IDs with summary stats
+ :param acc_raw: accessor to raw image data
+ :param acc_obj_ids: accessor to map of object IDs
+ :param expand_box_by: number of pixels to expand bounding box in all directions (without exceeding image boundary)
+ # :param deproject: assign object's z-position based on argmax of raw data if True
+ :return: pd.DataFrame
+ """
+ # build dataframe of objects, assign z index to each object
+
+ if acc_obj_ids.nz == 1: # deproject objects' z-coordinates from argmax of raw image
+ df = pd.DataFrame(regionprops_table(
+ acc_obj_ids.data[:, :, 0, 0],
+ intensity_image=acc_raw.data.argmax(axis=3, keepdims=True)[:, :, 0, 0].astype('uint16'),
+ properties=('label', 'area', 'intensity_mean', 'bbox')
+ )).rename(columns={'bbox-0': 'y0', 'bbox-1': 'x0', 'bbox-2': 'y1', 'bbox-3': 'x1'})
+ df['zi'] = df['intensity_mean'].round().astype('int')
+
+ else: # objects' z-coordinates come from arg of max count in object identities map
+ df = pd.DataFrame(regionprops_table(
+ acc_obj_ids.data[:, :, 0, :],
+ properties=('label', 'area', 'bbox')
+ )).rename(columns={
+ 'bbox-0': 'y0', 'bbox-1': 'x0', 'bbox-2': 'z0', 'bbox-3': 'y1', 'bbox-4': 'x1', 'bbox-5': 'z1'
+ })
+ df['zi'] = df['label'].apply(lambda x: (acc_obj_ids.data == x).sum(axis=(0, 1, 2)).argmax())
+
+ df = _df_insert_slices(df, acc_raw.shape_dict, expand_box_by)
+
+ # TODO: make this contingent on whether seg is included
+ df['binary_mask'] = df.apply(
+ lambda r: (acc_obj_ids.data == r.label).max(axis=-1)[r.y0: r.y1, r.x0: r.x1, 0],
+ axis=1,
+ result_type='reduce',
+ )
+ return df
+
+
+def _df_insert_slices(df: pd.DataFrame, sd: dict, expand_box_by) -> pd.DataFrame:
+ h = sd['Y']
+ w = sd['X']
+ nz = sd['Z']
-def _df_insert_slices(df: pd.DataFrame, shape: tuple, expand_box_by) -> pd.DataFrame:
- h, w, c, nz = shape
df['h'] = df['y1'] - df['y0']
df['w'] = df['x1'] - df['x0']
ebxy, ebz = expand_box_by
@@ -185,6 +239,18 @@ def _safe_add(a, g, b):
np.iinfo(a.dtype).max
).astype(a.dtype)
+def _make_object_ids_from_df(df: pd.DataFrame, sd: dict) -> InMemoryDataAccessor:
+ id_mask = np.zeros((sd['Y'], sd['X'], 1, sd['Z']), dtype='uint16')
+ if 'binary_mask' not in df.columns:
+ raise MissingSegmentationError('RoiSet dataframe does not contain segmentation')
+
+ def _label_obj(r):
+ sl = np.s_[r.y0:r.y1, r.x0:r.x1, :, r.zi:r.zi + 1]
+ id_mask[sl] = id_mask[sl] + r.label * r.binary_mask
+
+ df.apply(_label_obj, axis=1)
+ return InMemoryDataAccessor(id_mask)
+
class RoiSet(object):
@@ -192,7 +258,8 @@ class RoiSet(object):
def __init__(
self,
acc_raw: GenericImageDataAccessor,
- acc_obj_ids: GenericImageDataAccessor,
+ # acc_obj_ids: GenericImageDataAccessor,
+ df: pd.DataFrame,
params: RoiSetMetaParams = RoiSetMetaParams(),
):
"""
@@ -203,19 +270,20 @@ class RoiSet(object):
labels its membership in a connected object
:param params: optional arguments that influence the definition and representation of ROIs
"""
- assert acc_obj_ids.chroma == 1
- self.acc_obj_ids = acc_obj_ids
+ # assert acc_obj_ids.chroma == 1
+ # self.acc_obj_ids = acc_obj_ids
self.acc_raw = acc_raw
self.accs_derived = []
self.params = params
- self._df = self.filter_df(
- self.make_df_from_object_ids(
- self.acc_raw, self.acc_obj_ids, expand_box_by=params.expand_box_by
- ),
- params.filters,
- )
+ # self._df = self.filter_df(
+ # self.make_df_from_object_ids(
+ # self.acc_raw, self.acc_obj_ids, expand_box_by=params.expand_box_by
+ # ),
+ # params.filters,
+ # )
+ self._df = df
self.count = len(self._df)
self.object_class_maps = {} # classification results
@@ -223,6 +291,24 @@ class RoiSet(object):
"""Expose ROI meta information via the Pandas.DataFrame API"""
return self._df.itertuples(name='Roi')
+ @staticmethod
+ def from_object_ids(
+ acc_raw: GenericImageDataAccessor,
+ acc_obj_ids: GenericImageDataAccessor,
+ params: RoiSetMetaParams = RoiSetMetaParams(),
+ ):
+ assert acc_obj_ids.chroma == 1
+
+ df = _filter_df(
+ _make_df_from_object_ids(
+ acc_raw, acc_obj_ids, expand_box_by=params.expand_box_by
+ ),
+ params.filters,
+ )
+
+ return RoiSet(acc_raw, df, params)
+
+
# TODO: add or overload for object detection case
@staticmethod
def from_binary_mask(
@@ -241,50 +327,16 @@ class RoiSet(object):
:param params: optional arguments that influence the definition and representation of ROIs
:return: object identities map
"""
- return RoiSet(acc_raw, _get_label_ids(acc_seg, allow_3d=allow_3d, connect_3d=connect_3d), params)
+ return __class__.from_object_ids(acc_raw, _get_label_ids(acc_seg, allow_3d=allow_3d, connect_3d=connect_3d), params)
# TODO: generate overlapping RoiSet from multiple masks
# call e.g. static adder
+
@staticmethod
- def make_df_from_object_ids(acc_raw, acc_obj_ids, expand_box_by) -> pd.DataFrame:
- """
- Build dataframe associate object IDs with summary stats
- :param acc_raw: accessor to raw image data
- :param acc_obj_ids: accessor to map of object IDs
- :param expand_box_by: number of pixels to expand bounding box in all directions (without exceeding image boundary)
- # :param deproject: assign object's z-position based on argmax of raw data if True
- :return: pd.DataFrame
- """
- # build dataframe of objects, assign z index to each object
-
- if acc_obj_ids.nz == 1: # deproject objects' z-coordinates from argmax of raw image
- df = pd.DataFrame(regionprops_table(
- acc_obj_ids.data[:, :, 0, 0],
- intensity_image=acc_raw.data.argmax(axis=3, keepdims=True)[:, :, 0, 0].astype('uint16'),
- properties=('label', 'area', 'intensity_mean', 'bbox')
- )).rename(columns={'bbox-0': 'y0', 'bbox-1': 'x0', 'bbox-2': 'y1', 'bbox-3': 'x1'})
- df['zi'] = df['intensity_mean'].round().astype('int')
-
- else: # objects' z-coordinates come from arg of max count in object identities map
- df = pd.DataFrame(regionprops_table(
- acc_obj_ids.data[:, :, 0, :],
- properties=('label', 'area', 'bbox')
- )).rename(columns={
- 'bbox-0': 'y0', 'bbox-1': 'x0', 'bbox-2': 'z0', 'bbox-3': 'y1', 'bbox-4': 'x1', 'bbox-5': 'z1'
- })
- df['zi'] = df['label'].apply(lambda x: (acc_obj_ids.data == x).sum(axis=(0, 1, 2)).argmax())
-
- df = _df_insert_slices(df, acc_raw.shape, expand_box_by)
-
- # TODO: make this contingent on whether seg is included
- df['binary_mask'] = df.apply(
- lambda r: (acc_obj_ids.data == r.label).max(axis=-1)[r.y0: r.y1, r.x0: r.x1, 0],
- axis=1,
- result_type='reduce',
- )
- return df
+ def make_df_from_patches():
+ pass
# TODO: get overlapping segments
@@ -298,21 +350,6 @@ class RoiSet(object):
dfbb['iou'] = dfbb.apply()
-
- # TODO: test if overlaps exist
-
- @staticmethod
- def filter_df(df: pd.DataFrame, filters: RoiFilter = None) -> pd.DataFrame:
- query_str = 'label > 0' # always true
- if filters is not None: # parse filters
- for k, val in filters.dict(exclude_unset=True).items():
- assert k in ('area')
- vmin = val['min']
- vmax = val['max']
- assert vmin >= 0
- query_str = query_str + f' & {k} > {vmin} & {k} < {vmax}'
- return df.loc[df.query(query_str).index, :]
-
def get_df(self) -> pd.DataFrame:
return self._df
@@ -731,26 +768,29 @@ class RoiSet(object):
return {}
+ @property
+ def acc_obj_ids(self):
+ return _make_object_ids_from_df(self._df, self.acc_raw.shape_dict)
+
# TODO: add docstring
# TODO: make this work with obj det dataset
@staticmethod
def deserialize(acc_raw: GenericImageDataAccessor, where: Path, prefix=''):
df = pd.read_csv(where / 'dataframe' / (prefix + '.csv'))[['label', 'zi', 'y0', 'y1', 'x0', 'x1']]
- id_mask = np.zeros((*acc_raw.hw, 1, acc_raw.nz), dtype='uint16')
- def _label_obj(r):
- sl = np.s_[r.y0:r.y1, r.x0:r.x1, :, r.zi:r.zi + 1]
+ def _read_binary_mask(r):
ext = 'png'
fname = f'{prefix}-la{r.label:04d}-zi{r.zi:04d}.{ext}'
try:
ma_acc = generate_file_accessor(where / 'tight_patch_masks' / fname)
- bool_mask = ma_acc.data / np.iinfo(ma_acc.data.dtype).max
- id_mask[sl] = id_mask[sl] + r.label * bool_mask
+ assert ma_acc.chroma == 1 and ma_acc.nz == 1
+ mask_data = ma_acc.data / np.iinfo(ma_acc.data.dtype).max
+ return mask_data
except Exception as e:
raise DeserializeRoiSet(e)
-
- df.apply(_label_obj, axis=1)
- return RoiSet(acc_raw, InMemoryDataAccessor(id_mask))
+ df['binary_mask'] = df.apply(_read_binary_mask, axis=1)
+ id_mask = _make_object_ids_from_df(df, acc_raw.shape_dict)
+ return RoiSet.from_object_ids(acc_raw, id_mask)
def project_stack_from_focal_points(
@@ -808,4 +848,7 @@ class DeserializeRoiSet(Error):
pass
class DerivedChannelError(Error):
+ pass
+
+class MissingSegmentationError(Error):
pass
\ No newline at end of file
diff --git a/tests/base/test_roiset.py b/tests/base/test_roiset.py
index dcf849bd..8e577f50 100644
--- a/tests/base/test_roiset.py
+++ b/tests/base/test_roiset.py
@@ -137,7 +137,7 @@ class TestRoiSetMonoProducts(BaseTestRoiSetMonoProducts, unittest.TestCase):
def test_create_roiset_with_no_objects(self):
zero_obmap = InMemoryDataAccessor(np.zeros(self.seg_mask.shape, self.seg_mask.dtype))
- roiset = RoiSet(self.stack_ch_pa, zero_obmap)
+ roiset = RoiSet.from_object_ids(self.stack_ch_pa, zero_obmap)
self.assertEqual(roiset.count, 0)
def test_slices_are_valid(self):
@@ -592,7 +592,7 @@ class TestRoiSetSerialization(unittest.TestCase):
id_map = _get_label_ids(self.seg_mask_3d, allow_3d=True, connect_3d=False)
self.assertEqual(self.stack_ch_pa.shape, id_map.shape)
- roiset = RoiSet(
+ roiset = RoiSet.from_object_ids(
self.stack_ch_pa,
id_map,
params=RoiSetMetaParams(mask_type='contours')
@@ -605,7 +605,7 @@ class TestRoiSetSerialization(unittest.TestCase):
self.assertEqual(self.stack_ch_pa.shape[0:3], id_map.shape[0:3])
self.assertEqual(id_map.nz, 1)
- roiset = RoiSet(
+ roiset = RoiSet.from_object_ids(
self.stack_ch_pa,
id_map,
params=RoiSetMetaParams(mask_type='contours')
--
GitLab