added blob finding to fiber generation

Author: Joel Vincent Bernier

conda.recipe/meta.yaml

@@ -29,14 +29,14 @@ requirements:
     - matplotlib
     - numba
     - numpy
-    - progressbar >=2.3
+    - progressbar
     - python
     - python.app # [osx]
     - pyyaml
     - scikit-image
     - scikit-learn
     - scipy
-    - wxpython ==3
+    - wxpython
 
 test:
   imports:

hexrd/findorientations.py

@@ -26,9 +26,7 @@
 
 from hexrd.fitgrains import get_instrument_parameters
 
-logger = logging.getLogger(__name__)
-
-save_as_ascii = False  # FIX LATER...
+from skimage.feature import blob_dog, blob_log
 
 # just require scikit-learn?
 have_sklearn = False
@@ -43,9 +41,20 @@
     pass
 
 
+method = "blob_dog"  # !!! have to get this from the config
+save_as_ascii = False  # FIX LATER...
+
+logger = logging.getLogger(__name__)
+
+
+# =============================================================================
+# FUNCTIONS
+# =============================================================================
+
+
 def generate_orientation_fibers(
         eta_ome, chi, threshold, seed_hkl_ids, fiber_ndiv,
-        filt_stdev=0.8, ncpus=1):
+        method='blob_dog', filt_stdev=0.8, ncpus=1):
     """
     From ome-eta maps and hklid spec, generate list of
     quaternions from fibers
@@ -82,21 +91,42 @@ def generate_orientation_fibers(
     numSpots = []
     coms = []
     for i in seed_hkl_ids:
-        # First apply filter
-        this_map_f = -ndimage.filters.gaussian_laplace(
-            eta_ome.dataStore[i], filt_stdev)
-
-        labels_t, numSpots_t = ndimage.label(
-            this_map_f > threshold,
-            structureNDI_label
-            )
-        coms_t = np.atleast_2d(
-            ndimage.center_of_mass(
-                this_map_f,
-                labels=labels_t,
-                index=np.arange(1, np.amax(labels_t)+1)
+        if method == 'label':
+            # First apply filter
+            this_map_f = -ndimage.filters.gaussian_laplace(
+                eta_ome.dataStore[i], filt_stdev)
+            
+            labels_t, numSpots_t = ndimage.label(
+                this_map_f > threshold,
+                structureNDI_label
                 )
-            )
+            coms_t = np.atleast_2d(
+                ndimage.center_of_mass(
+                    this_map_f,
+                    labels=labels_t,
+                    index=np.arange(1, np.amax(labels_t) + 1)
+                    )
+                )
+        elif method in ['blob_log', 'blob_dog']:
+            # must scale map
+            this_map = eta_ome.dataStore[i]
+            this_map[np.isnan(this_map)] = 0.
+            this_map -= np.min(this_map)
+            scl_map = 2*this_map/np.max(this_map) - 1.
+
+            # FIXME: need to expose the parameters to config options.
+            if method == 'blob_log':
+                blobs = np.atleast_2d(
+                    blob_log(scl_map, min_sigma=0.5, max_sigma=5,
+                             num_sigma=10, threshold=0.01, overlap=0.1)
+                )
+            else:
+                blobs = np.atleast_2d(
+                    blob_dog(scl_map, min_sigma=0.5, max_sigma=5,
+                             sigma_ratio=1.6, threshold=0.01, overlap=0.1)
+                    )
+            numSpots_t = len(blobs)
+            coms_t = blobs[:, :2]
         numSpots.append(numSpots_t)
         coms.append(coms_t)
         pass