Merge pull request #321 from alexlib/master

updated to 0.25.3 for openpiv_tk_gui

openpiv/smoothn.py

@@ -1,7 +1,7 @@
 from numpy import *
 from pylab import *
 import scipy.optimize.lbfgsb as lbfgsb
-import numpy.linalg
+import scipy
 from scipy.fftpack import dct, idct
 import numpy as np
 import numpy.ma as ma
@@ -803,7 +803,8 @@ def sparseTest(n=1000):
 # https://stackoverflow.com/questions/17115030/want-to-smooth-a-contour-from-a-masked-array
 
 def smooth(u, mask):
-    r = u.filled(0.)  # set all 'masked' points to 0. so they aren't used in the smoothing
+    m = ~mask
+    r = u*m  # set all 'masked' points to 0. so they aren't used in the smoothing
     a = 4*r[1:-1,1:-1] + r[2:,1:-1] + r[:-2,1:-1] + r[1:-1,2:] + r[1:-1,:-2]
     b = 4*m[1:-1,1:-1] + m[2:,1:-1] + m[:-2,1:-1] + m[1:-1,2:] + m[1:-1,:-2]  # a divisor that accounts for masked points
     b[b==0] = 1.  # for avoiding divide by 0 error (region is masked so value doesn't matter)

openpiv/validation.py

@@ -23,6 +23,7 @@
 import numpy as np
 from scipy.ndimage import generic_filter
 import matplotlib.pyplot as plt
+from openpiv.windef import PIVSettings
 
 
 
@@ -165,7 +166,13 @@ def sig2noise_val(
     return ind 
 
 
-def local_median_val(u, v, u_threshold, v_threshold, size=1):
+def local_median_val(
+        u: np.ndarray, 
+        v: np.ndarray, 
+        u_threshold: float, 
+        v_threshold: float, 
+        size: int=1
+        )->np.ndarray:
     """Eliminate spurious vectors with a local median threshold.
 
     This validation method tests for the spatial consistency of the data.
@@ -201,16 +208,20 @@ def local_median_val(u, v, u_threshold, v_threshold, size=1):
     """
 
     # kernel footprint
-    f = np.ones((2*size+1, 2*size+1))
-    f[size,size] = 0
-
-    masked_u = np.where(~u.mask, u.data, np.nan)
-    masked_v = np.where(~v.mask, v.data, np.nan)
+    # f = np.ones((2*size+1, 2*size+1))
+    # f[size,size] = 0
+
+    if np.ma.is_masked(u):
+        masked_u = np.where(~u.mask, u.data, np.nan)
+        masked_v = np.where(~v.mask, v.data, np.nan)
+    else:
+        masked_u = u
+        masked_v = v
 
     um = generic_filter(masked_u, np.nanmedian, mode='constant',
-                        cval=np.nan, footprint=f)
+                        cval=np.nan, size=(2*size+1, 2*size+1))
     vm = generic_filter(masked_v, np.nanmedian, mode='constant',
-                        cval=np.nan, footprint=f)
+                        cval=np.nan, size=(2*size+1, 2*size+1))
 
     ind = (np.abs((u - um)) > u_threshold) | (np.abs((v - vm)) > v_threshold)