CellProfiler-nightly 5.0.0.dev257__tar.gz → 5.0.0.dev259__tar.gz

{cellprofiler_nightly-5.0.0.dev257 → cellprofiler_nightly-5.0.0.dev259}/CellProfiler_nightly.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: CellProfiler-nightly
-Version: 5.0.0.dev257
+Version: 5.0.0.dev259
 Summary: CellProfiler is a free open-source software designed to enable biologists without training in computer vision or programming to quantitatively measure phenotypes from thousands of images automatically.
 Author: Anne Carpenter, Thouis (Ray) Jones, Lee Kamentsky, Vebjorn Ljosa, David Logan, Mark Bray, Madison Swain-Bowden, Allen Goodman, Claire McQuinn, Alice Lucas, Callum Tromans-Coia
 Author-email: Beth Cimini <bcimini@broadinstitute.org>, David Stirling <dstirling@glencoesoftware.com>, Nodar Gogoberidze <ngogober@broadinstitute.org>

{cellprofiler_nightly-5.0.0.dev257 → cellprofiler_nightly-5.0.0.dev259}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: CellProfiler-nightly
-Version: 5.0.0.dev257
+Version: 5.0.0.dev259
 Summary: CellProfiler is a free open-source software designed to enable biologists without training in computer vision or programming to quantitatively measure phenotypes from thousands of images automatically.
 Author: Anne Carpenter, Thouis (Ray) Jones, Lee Kamentsky, Vebjorn Ljosa, David Logan, Mark Bray, Madison Swain-Bowden, Allen Goodman, Claire McQuinn, Alice Lucas, Callum Tromans-Coia
 Author-email: Beth Cimini <bcimini@broadinstitute.org>, David Stirling <dstirling@glencoesoftware.com>, Nodar Gogoberidze <ngogober@broadinstitute.org>

{cellprofiler_nightly-5.0.0.dev257 → cellprofiler_nightly-5.0.0.dev259}/cellprofiler/_version.py

@@ -17,5 +17,5 @@ __version__: str
 __version_tuple__: VERSION_TUPLE
 version_tuple: VERSION_TUPLE
 
-__version__ = version = '5.0.0.dev257'
-__version_tuple__ = version_tuple = (5, 0, 0, 'dev257')
+__version__ = version = '5.0.0.dev259'
+__version_tuple__ = version_tuple = (5, 0, 0, 'dev259')

{cellprofiler_nightly-5.0.0.dev257 → cellprofiler_nightly-5.0.0.dev259}/cellprofiler/modules/identifytertiaryobjects.py

@@ -289,6 +289,19 @@ but the results will be zero or not-a-number (NaN).
             primary_mask = primary_labels == 0
         tertiary_labels[primary_mask == False] = 0
         #
+        # Check if a label was deleted as a result of the subtraction
+        #
+        secondary_unique_labels, secondary_unique_indices = numpy.unique(secondary_labels, return_index=True)
+        tertiary_unique_labels = numpy.unique(tertiary_labels)
+        missing_labels = numpy.setdiff1d(secondary_unique_labels, tertiary_unique_labels)
+        for missing_label in missing_labels:
+            # If a label was deleted, manually add a pixel to the tertiary_labels.
+            # This workaround ensures that ghost objects do not get created by identifytertiaryobjects.
+            
+            # first non-zero (top-left) coodrinate of the secondary object is used to add a pixel to the tertiary_labels
+            first_row, first_col = numpy.unravel_index(secondary_unique_indices[missing_label], secondary_labels.shape)
+            tertiary_labels[first_row, first_col] = missing_label
+        #
         # Get the outlines of the tertiary image
         #
         tertiary_outlines = outline(tertiary_labels) != 0
@@ -304,6 +317,7 @@ but the results will be zero or not-a-number (NaN).
         child_count_of_secondary, secondary_parents = secondary_objects.relate_children(
             tertiary_objects
         )
+
         if self.shrink_primary:
             child_count_of_primary, primary_parents = primary_objects.relate_children(
                 tertiary_objects

{cellprofiler_nightly-5.0.0.dev257 → cellprofiler_nightly-5.0.0.dev259}/cellprofiler/modules/measuretexture.py

@@ -547,9 +547,7 @@ measured and will result in a undefined value in the output file.
 
         gray_levels = int(self.gray_levels.value)
 
-        unique_labels = numpy.unique(labels)
-        if unique_labels[0] == 0:
-            unique_labels = unique_labels[1:]
+        unique_labels = objects.indices
 
         n_directions = 13 if objects.volumetric else 4
 
@@ -594,17 +592,16 @@ measured and will result in a undefined value in the output file.
                 pixel_data, in_range=(0, 255), out_range=(0, gray_levels - 1)
             ).astype(numpy.uint8)
         props = skimage.measure.regionprops(labels, pixel_data)
+        features = numpy.empty((n_directions, 13, max(unique_labels)))
 
-        features = numpy.empty((n_directions, 13, len(unique_labels)))
-
-        for index, prop in enumerate(props):
+        for prop in props:
             label_data = prop["intensity_image"]
             try:
-                features[:, :, index] = mahotas.features.haralick(
+                features[:, :, prop.label-1] = mahotas.features.haralick(
                     label_data, distance=scale, ignore_zeros=True
                 )
             except ValueError:
-                features[:, :, index] = numpy.nan
+                features[:, :, prop.label-1] = numpy.nan
 
         for direction, direction_features in enumerate(features):
             for feature_name, feature in zip(F_HARALICK, direction_features):