Convert unsigned J2K data to signed (#53)

Author: scaramallion

pylibjpeg/pydicom/pixel_data_handler.py

@@ -42,8 +42,9 @@
 
 from pydicom.encaps import generate_pixel_data_frame
 from pydicom.pixel_data_handlers.util import pixel_dtype, get_expected_length
+from pydicom.uid import JPEG2000, JPEG2000Lossless
 
-from .utils import get_pixel_data_decoders
+from .utils import get_pixel_data_decoders, get_j2k_parameters
 
 
 LOGGER = logging.getLogger(__name__)
@@ -54,6 +55,8 @@
     'numpy': ('http://www.numpy.org/', 'NumPy'),
 }
 
+APPLY_J2K_CORRECTIONS = True
+
 _DECODERS = get_pixel_data_decoders()
 SUPPORTED_TRANSFER_SYNTAXES = list(_DECODERS.keys())
 
@@ -162,10 +165,28 @@ def get_pixeldata(ds):
     # Generators for the encoded JPEG image frame(s) and insertion offsets
     generate_frames = generate_pixel_data_frame(ds.PixelData, nr_frames)
     generate_offsets = range(0, expected_len, frame_len)
+    pixel_module = ds.group_dataset(0x0028)
     for frame, offset in zip(generate_frames, generate_offsets):
         # Encoded JPEG data to be sent to the decoder
-        arr[offset:offset + frame_len] = decoder(
-            frame, ds.group_dataset(0x0028)
-        )
+        arr[offset:offset + frame_len] = decoder(frame, pixel_module)
+
+    if tsyntax in [JPEG2000, JPEG2000Lossless] and APPLY_J2K_CORRECTIONS:
+        j2k_parameters = get_j2k_parameters(frame)
+        if j2k_parameters:
+            shift = ds.BitsAllocated - j2k_parameters['precision']
+            if (
+                shift
+                and not j2k_parameters['is_signed']
+                and bool(ds.PixelRepresentation)
+            ):
+                # Correct for a mismatch between J2K and Pixel Representation
+                #  by converting unsigned data to signed (2's complement)
+                pixel_module.PixelRepresentation = 0
+                # This probably isn't very efficient
+                arr = arr.view(pixel_dtype(pixel_module))
+                np.left_shift(arr, shift, out=arr)
+                arr = arr.astype(pixel_dtype(ds))
+
+                return np.right_shift(arr, shift)
 
     return arr.view(pixel_dtype(ds))

pylibjpeg/pydicom/utils.py

@@ -149,3 +149,43 @@ def reshape_frame(ds, arr):
             arr = arr.transpose(1, 2, 0)
 
     return arr
+
+
+def get_j2k_parameters(codestream):
+    """Return some of the JPEG 2000 component sample's parameters in `stream`.
+
+    Parameters
+    ----------
+    codestream : bytes
+        The JPEG 2000 (ISO/IEC 15444-1) codestream data to be parsed.
+
+    Returns
+    -------
+    dict
+        A dict containing the JPEG 2000 parameters for the first component
+        sample, will be empty if `codestream` doesn't contain JPEG 2000 data or
+        if unable to parse the data.
+    """
+    try:
+        # First 2 bytes must be the SOC marker - if not then wrong format
+        if codestream[0:2] != b'\xff\x4f':
+            return {}
+
+        # SIZ is required to be the second marker - Figure A-3 in 15444-1
+        if codestream[2:4] != b'\xff\x51':
+            return {}
+
+        # See 15444-1 A.5.1 for format of the SIZ box and contents
+        ssiz = ord(codestream[42:43])
+        parameters = {}
+        if ssiz & 0x80:
+            parameters['precision'] = (ssiz & 0x7F) + 1
+            parameters['is_signed'] = True
+        else:
+            parameters['precision'] = ssiz + 1
+            parameters['is_signed'] = False
+
+        return parameters
+
+    except (IndexError, TypeError):
+        return {}

pylibjpeg/tests/test_pydicom.py

@@ -17,7 +17,7 @@
 
 from pylibjpeg.data import get_indexed_datasets
 from pylibjpeg.pydicom.utils import (
-    generate_frames, reshape_frame, get_pixel_data_decoders
+    generate_frames, reshape_frame, get_pixel_data_decoders, get_j2k_parameters
 )
 from pylibjpeg.utils import add_handler, remove_handler
 
@@ -219,6 +219,24 @@ def test_pixel_array(self):
             [165,  11,   0],
             [175,  17,   0]] == arr[175:195, 28, :].tolist()
 
+    def test_pixel_representation_mismatch(self):
+        """Test mismatch between Pixel Representation and the J2K data."""
+        index = get_indexed_datasets(self.uid)
+        ds = index['J2K_pixelrep_mismatch.dcm']['ds']
+
+        arr = ds.pixel_array
+        assert arr.flags.writeable
+        assert 'int16' == arr.dtype
+        assert (512, 512) == arr.shape
+
+        assert -2000 == arr[0, 0]
+        assert [621, 412, 138, -193, -520, -767, -907, -966, -988, -995] == (
+            arr[47:57, 279].tolist()
+        )
+        assert [-377, -121, 141, 383, 633, 910, 1198, 1455, 1638, 1732] == (
+            arr[328:338, 106].tolist()
+        )
+
 
 class TestUtils(object):
     """Test the pydicom.utils functions."""
@@ -272,3 +290,36 @@ def test_generate_frames_3s_0p(self):
         assert 'uint8' == arr.dtype
         assert (ds.Rows, ds.Columns, 3) == arr.shape
         assert [48,  128,  128] == arr[159, 290, :].tolist()
+
+
+class TestGetJ2KParameters:
+    """Tests for get_j2k_parameters."""
+    def test_parameters(self):
+        """Test getting the parameters for a JPEG2K codestream."""
+        base = b'\xff\x4f\xff\x51' + b'\x00' * 38
+        # Signed
+        for ii in range(135, 143):
+            params = get_j2k_parameters(base + bytes([ii]))
+            assert ii - 127 == params['precision']
+            assert params['is_signed']
+
+        # Unsigned
+        for ii in range(7, 17):
+            params = get_j2k_parameters(base + bytes([ii]))
+            assert ii + 1 == params['precision']
+            assert not params['is_signed']
+
+    def test_not_j2k(self):
+        """Test result when no JPEG2K SOF marker present"""
+        base = b'\xff\x4e\xff\x51' + b'\x00' * 38
+        assert {} == get_j2k_parameters(base + b'\x8F')
+
+    def test_no_siz(self):
+        """Test result when no SIZ box present"""
+        base = b'\xff\x4f\xff\x52' + b'\x00' * 38
+        assert {} == get_j2k_parameters(base + b'\x8F')
+
+    def test_short_bytestream(self):
+        """Test result when no SIZ box present"""
+        assert {} == get_j2k_parameters(b'')
+        assert {} == get_j2k_parameters(b'\xff\x4f\xff\x51' + b'\x00' * 20)