6007 reverse_indexing for PILReader (#6008)

Fixes #6007

### Description
- reverse_indexing = False:
to support consistency with PIL/torchvision
```py
img = LoadImage(image_only=True, ensure_channel_first=True, reverse_indexing=False)("MONAI-logo_color.png")  # PILReader
torchvision.utils.save_image(img, "MONAI-logo_color_torchvision.png", normalize=True)
```
- reverse_indexing = True:
to support consistency with other backends in monai
```py
img = LoadImage(image_only=True, ensure_channel_first=True, reader="PILReader", reverse_indexing=True)(filename)  # PIL backend
img_1 = LoadImage(image_only=True, ensure_channel_first=True, reader="ITKReader")(filename)  # itk backend
np.testing.assert_allclose(img, img_1)  # true
```

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Wenqi Li <wenqil@nvidia.com>

Author: wyli

monai/data/image_reader.py

@@ -1141,13 +1141,17 @@ class PILReader(ImageReader):
     Args:
         converter: additional function to convert the image data after `read()`.
             for example, use `converter=lambda image: image.convert("LA")` to convert image format.
+        reverse_indexing: whether to swap axis 0 and 1 after loading the array, this is enabled by default,
+            so that output of the reader is consistent with the other readers. Set this option to ``False`` to use
+            the PIL backend's original spatial axes convention.
         kwargs: additional args for `Image.open` API in `read()`, mode details about available args:
             https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.open
     """
 
-    def __init__(self, converter: Callable | None = None, **kwargs):
+    def __init__(self, converter: Callable | None = None, reverse_indexing: bool = True, **kwargs):
         super().__init__()
         self.converter = converter
+        self.reverse_indexing = reverse_indexing
         self.kwargs = kwargs
 
     def verify_suffix(self, filename: Sequence[PathLike] | PathLike) -> bool:
@@ -1194,8 +1198,8 @@ def get_data(self, img) -> tuple[np.ndarray, dict]:
         It computes `spatial_shape` and stores it in meta dict.
         When loading a list of files, they are stacked together at a new dimension as the first dimension,
         and the metadata of the first image is used to represent the output metadata.
-        Note that it will swap axis 0 and 1 after loading the array because the `HW` definition in PIL
-        is different from other common medical packages.
+        Note that by default `self.reverse_indexing` is set to ``True``, which swaps axis 0 and 1 after loading
+        the array because the spatial axes definition in PIL is different from other common medical packages.
 
         Args:
             img: a PIL Image object loaded from a file or a list of PIL Image objects.
@@ -1207,7 +1211,7 @@ def get_data(self, img) -> tuple[np.ndarray, dict]:
         for i in ensure_tuple(img):
             header = self._get_meta_dict(i)
             header[MetaKeys.SPATIAL_SHAPE] = self._get_spatial_shape(i)
-            data = np.moveaxis(np.asarray(i), 0, 1)
+            data = np.moveaxis(np.asarray(i), 0, 1) if self.reverse_indexing else np.asarray(i)
             img_array.append(data)
             header[MetaKeys.ORIGINAL_CHANNEL_DIM] = (
                 "no_channel" if len(data.shape) == len(header[MetaKeys.SPATIAL_SHAPE]) else -1

monai/transforms/io/array.py

@@ -116,9 +116,9 @@ class LoadImage(Transform):
         - Current default readers: (nii, nii.gz -> NibabelReader), (png, jpg, bmp -> PILReader),
           (npz, npy -> NumpyReader), (nrrd -> NrrdReader), (DICOM file -> ITKReader).
 
-    Please note that for png, jpg, bmp, and other 2D formats, readers often swap axis 0 and 1 after
-    loading the array because the `HW` definition for non-medical specific file formats is different
-    from other common medical packages.
+    Please note that for png, jpg, bmp, and other 2D formats, readers by default swap axis 0 and 1 after
+    loading the array with ``reverse_indexing`` set to ``True`` because the spatial axes definition
+    for non-medical specific file formats is different from other common medical packages.
 
     See also:
 

monai/transforms/io/dictionary.py

@@ -53,9 +53,9 @@ class LoadImaged(MapTransform):
         - Current default readers: (nii, nii.gz -> NibabelReader), (png, jpg, bmp -> PILReader),
           (npz, npy -> NumpyReader), (dcm, DICOM series and others -> ITKReader).
 
-    Please note that for png, jpg, bmp, and other 2D formats, readers often swap axis 0 and 1 after
-    loading the array because the `HW` definition for non-medical specific file formats is different
-    from other common medical packages.
+    Please note that for png, jpg, bmp, and other 2D formats, readers by default swap axis 0 and 1 after
+    loading the array with ``reverse_indexing`` set to ``True`` because the spatial axes definition
+    for non-medical specific file formats is different from other common medical packages.
 
     Note:
 

tests/test_pil_reader.py

@@ -25,7 +25,7 @@
 
 TEST_CASE_2 = [(128, 128, 3), ["test_image.png"], (128, 128, 3), (128, 128)]
 
-TEST_CASE_3 = [(128, 128, 4), ["test_image.png"], (128, 128, 4), (128, 128)]
+TEST_CASE_3 = [(128, 128, 4), ["test_image.png"], (128, 128, 4), (128, 128), False]
 
 TEST_CASE_4 = [(128, 128), ["test_image1.png", "test_image2.png", "test_image3.png"], (3, 128, 128), (128, 128)]
 
@@ -38,20 +38,21 @@
 
 class TestPNGReader(unittest.TestCase):
     @parameterized.expand([TEST_CASE_1, TEST_CASE_2, TEST_CASE_3, TEST_CASE_4, TEST_CASE_5, TEST_CASE_6])
-    def test_shape_value(self, data_shape, filenames, expected_shape, meta_shape):
+    def test_shape_value(self, data_shape, filenames, expected_shape, meta_shape, reverse=True):
         test_image = np.random.randint(0, 256, size=data_shape)
         with tempfile.TemporaryDirectory() as tempdir:
             for i, name in enumerate(filenames):
                 filenames[i] = os.path.join(tempdir, name)
                 Image.fromarray(test_image.astype("uint8")).save(filenames[i])
-            reader = PILReader(mode="r")
+            reader = PILReader(mode="r", reverse_indexing=reverse)
             result = reader.get_data(reader.read(filenames))
             # load image by PIL and compare the result
             test_image = np.asarray(Image.open(filenames[0]))
 
         self.assertTupleEqual(tuple(result[1]["spatial_shape"]), meta_shape)
         self.assertTupleEqual(result[0].shape, expected_shape)
-        test_image = np.moveaxis(test_image, 0, 1)
+        if reverse:
+            test_image = np.moveaxis(test_image, 0, 1)
         if result[0].shape == test_image.shape:
             np.testing.assert_allclose(result[0], test_image)
         else: