Adding shape analysis

pyqae/shape_analysis.py

@@ -0,0 +1,84 @@
+"""
+Tools for analyzing the labels and data made in lungstage
+"""
+import warnings
+from warnings import warn
+
+import matplotlib.pyplot as plt
+import numpy as np
+import pandas as pd
+from pyqae.nd import autocrop
+from pyqae.viz import draw_3d_labels
+from skimage.measure import regionprops
+
+try:
+    # sklearn team renamed the core marching cubes and replaced it with something funkier
+    from skimage.measure import marching_cubes_classic as marching_cubes
+except ImportError:
+    from skimage.measure import marching_cubes
+
+from skimage.filters import gaussian
+from scipy.ndimage import zoom
+from mpl_toolkits.mplot3d.art3d import Poly3DCollection
+
+fake_HTML = lambda x: x  # dummy function
+
+
+def scalar_attributes_list(im_props):
+    """
+    Makes list of all scalar, non-dunder, non-hidden
+    attributes of skimage.measure.regionprops object
+    """
+
+    attributes_list = []
+
+    for i, test_attribute in enumerate(dir(im_props[0])):
+
+        # Attribute should not start with _ and cannot return an array
+        # does not yet return tuples
+        try:
+            if test_attribute[:1] != '_' and not \
+                    isinstance(getattr(im_props[0], test_attribute), np.ndarray):
+                attributes_list += [test_attribute]
+        except Exception as e:
+            warn("Not implemented: {} - {}".format(test_attribute, e), RuntimeWarning)
+
+    return attributes_list
+
+
+def regionprops_to_df(im_props):
+    """
+    Read content of all attributes for every item in a list
+    output by skimage.measure.regionprops
+    """
+
+    with warnings.catch_warnings():
+        warnings.simplefilter("ignore")
+        attributes_list = scalar_attributes_list(im_props)
+
+    # Initialise list of lists for parsed data
+    parsed_data = []
+
+    # Put data from im_props into list of lists
+    for i, _ in enumerate(im_props):
+        parsed_data += [[]]
+
+        for j in range(len(attributes_list)):
+            parsed_data[i] += [getattr(im_props[i], attributes_list[j])]
+
+    # Return as a Pandas DataFrame
+    return pd.DataFrame(parsed_data, columns=attributes_list)
+
+
+def region_analysis(cur_mask, cur_image):
+    assert cur_mask.shape == cur_image.shape, "Mask and image must have same shape\n {} != {}".format(cur_mask.shape,
+                                                                                                      cur_image.shape)
+    return regionprops_to_df(regionprops(cur_mask, intensity_image=cur_image))
+
+
+if __name__ == '__main__':
+    import doctest
+    # noinspection PyUnresolvedReferences
+    import pyqae.shape_analysis
+
+    doctest.testmod(pyqae.shape_analysis, verbose=True, optionflags=doctest.ELLIPSIS)