From 940b79410abfa86f13fce1ea706e8f4ae622d0f9 Mon Sep 17 00:00:00 2001
From: Christopher Rhodes <christopher.rhodes@embl.de>
Date: Thu, 12 Sep 2024 13:03:36 +0200
Subject: [PATCH] Tested segmentation from two dataframes
---
model_server/base/roiset.py | 24 +++++++++++++++------
tests/base/test_roiset.py | 43 +++++++++++++++++++++++++++++++++++--
2 files changed, 59 insertions(+), 8 deletions(-)
diff --git a/model_server/base/roiset.py b/model_server/base/roiset.py
index b5618abe..fbda1cf1 100644
--- a/model_server/base/roiset.py
+++ b/model_server/base/roiset.py
@@ -169,17 +169,28 @@ def filter_df_overlap_bbox(df1: pd.DataFrame, df2: pd.DataFrame = None) -> pd.Da
sdf.loc[:, 'bbox_intersec'] = intersec
return sdf
-# TODO: option to quantify overlap e.g. IOU
-def filter_df_overlap_seg(df: pd.DataFrame) -> pd.DataFrame:
+
+def filter_df_overlap_seg(df1: pd.DataFrame, df2: pd.DataFrame = None) -> pd.DataFrame:
"""
- Return subset of DataFrame whose segmentations overlap in 3D space.
+ If passed a single DataFrame, return the subset whose segmentations overlap in 3D space. If passed two DataFrames,
+ return the subset where a ROI in the first overlaps a ROI in the second. May return duplicates entries where a ROI
+ overlaps with multiple neighbors.
+ :param df1: DataFrame with potentially overlapping bounding boxes
+ :param df2: (optional) second DataFrame
+ :return DataFrame describing subset of overlapping ROIs
+ seg_overlaps_with: index of ROI that overlaps
+ seg_intersec: pixel area of intersecting region
+ seg_iou: intersection over union
"""
- dfbb = filter_df_overlap_bbox(df)
+ dfbb = filter_df_overlap_bbox(df1, df2)
def _overlap_seg(r):
- roi1 = df.loc[r.name]
- roi2 = df.loc[r.bbox_overlaps_with]
+ roi1 = df1.loc[r.name]
+ if df2 is not None:
+ roi2 = df2.loc[r.bbox_overlaps_with]
+ else:
+ roi2 = df1.loc[r.bbox_overlaps_with]
ex0 = min(roi1.x0, roi2.x0, roi1.x1, roi2.x1)
ew = max(roi1.x0, roi2.x0, roi1.x1, roi2.x1) - ex0
ey0 = min(roi1.y0, roi2.y0, roi1.y1, roi2.y1)
@@ -197,6 +208,7 @@ def filter_df_overlap_seg(df: pd.DataFrame) -> pd.DataFrame:
dfbb['seg_iou'] = emasks.apply(lambda x: (x == 2).sum() / (x > 0).sum())
return dfbb
+
def make_df_from_object_ids(acc_raw, acc_obj_ids, expand_box_by) -> pd.DataFrame:
"""
Build dataframe associate object IDs with summary stats
diff --git a/tests/base/test_roiset.py b/tests/base/test_roiset.py
index 5d14bfab..c4d98df8 100644
--- a/tests/base/test_roiset.py
+++ b/tests/base/test_roiset.py
@@ -647,7 +647,7 @@ class TestRoiSetPolygons(BaseTestRoiSetMonoProducts, unittest.TestCase):
self.assertTrue((res.bbox_intersec == 2).all())
return res
- # TODO: test overloaded condition comparing two DFs
+
def test_overlap_bbox_multiple(self):
df1 = pd.DataFrame({
'x0': [0, 1],
@@ -668,6 +668,7 @@ class TestRoiSetPolygons(BaseTestRoiSetMonoProducts, unittest.TestCase):
self.assertEqual(len(res), 1)
self.assertTrue((res.bbox_intersec == 2).all())
+
def test_overlap_seg(self):
df = pd.DataFrame({
'x0': [0, 1, 2],
@@ -693,4 +694,42 @@ class TestRoiSetPolygons(BaseTestRoiSetMonoProducts, unittest.TestCase):
res = filter_df_overlap_seg(df)
self.assertTrue((res.loc[res.seg_overlaps, :].index == [1, 2]).all())
- self.assertTrue((res.loc[res.seg_overlaps, 'seg_iou'] == 0.4).all())
\ No newline at end of file
+ self.assertTrue((res.loc[res.seg_overlaps, 'seg_iou'] == 0.4).all())
+
+ def test_overlap_seg_multiple(self):
+ df1 = pd.DataFrame({
+ 'x0': [0, 1],
+ 'x1': [2, 3],
+ 'y0': [0, 0],
+ 'y1': [2, 2],
+ 'zi': [0, 0],
+ 'binary_mask': [
+ [
+ [1, 1],
+ [1, 0]
+ ],
+ [
+ [0, 1],
+ [1, 1]
+ ],
+ ]
+ })
+ df2 = pd.DataFrame({
+ 'x0': [2],
+ 'x1': [4],
+ 'y0': [0],
+ 'y1': [2],
+ 'zi': [0],
+ 'binary_mask': [
+ [
+ [1, 1],
+ [1, 1]
+ ],
+ ]
+ })
+ res = filter_df_overlap_seg(df1, df2)
+ self.assertTrue((res.loc[1, 'bbox_overlaps_with'] == [0]).all())
+ self.assertEqual(len(res), 1)
+ self.assertTrue((res.bbox_intersec == 2).all())
+ self.assertTrue((res.loc[res.seg_overlaps, :].index == [1]).all())
+ self.assertTrue((res.loc[res.seg_overlaps, 'seg_iou'] == 0.4).all())
--
GitLab