diff --git a/skimage/measure/_simple_metrics.py b/skimage/measure/simple_metrics.py
similarity index 69%
rename from skimage/measure/_simple_metrics.py
rename to skimage/measure/simple_metrics.py
index 2848b42c..883b4dbd 100644
--- a/skimage/measure/_simple_metrics.py
+++ b/skimage/measure/simple_metrics.py
@@ -7,7 +7,7 @@ __all__ = ['mse', 'nrmse', 'psnr']
 
 
 def mse(X, Y):
-    """ compute mean-squared error between two images.
+    """Compute the mean-squared error between two images.
 
     Parameters
     ----------
@@ -27,42 +27,38 @@ def mse(X, Y):
     return np.square(X - Y).mean()
 
 
-def nrmse(im_true, im, norm_type='Euclidean'):
-    """ compute the normalized root mean-squared error between two images.
+def nrmse(im_true, im_test, norm_type='Euclidean'):
+    """Compute the normalized root mean-squared error between two images.
 
     Parameters
     ----------
     im_true : ndarray
         Ground-truth image.
-    im : ndarray
+    im_test : ndarray
         Test image.
     norm_type : {'Euclidean', 'min-max', 'mean'}
         Controls the normalization method to use in the denominator of the
-        NRMSE.
+        NRMSE.  There is no standard method of normalization across the
+        literature [1]_.  The methods available here are as follows:
+
+        - 'Euclidean' : normalize by the Euclidean norm of ``im_true``.
+        - 'min-max'   : normalize by the intensity range of ``im_true``.
+        - 'mean'      : normalize by the mean of ``im_true``.
 
     Returns
     -------
     nrmse : float
         The NRMSE metric.
 
-    Notes
-    -----
-    There is no standard method of normalization across the literature [1]_.
-    The methods available here are as follows:
-
-    - 'Euclidean' : normalize by the Euclidean norm of ``im_true``.
-    - 'min-max'   : normalize by the intensity range of ``im_true``.
-    - 'mean'      : normalize by the mean of ``im_true``.
-
     References
     ----------
     .. [1] https://en.wikipedia.org/wiki/Root-mean-square_deviation
 
     """
-    if not im_true.dtype == im.dtype:
+    if not im_true.dtype == im_test.dtype:
         raise ValueError('Input images must have the same dtype.')
 
-    if not im_true.shape == im.shape:
+    if not im_true.shape == im_test.shape:
         raise ValueError('Input images must have the same dimensions.')
 
     norm_type = norm_type.lower()
@@ -72,17 +68,19 @@ def nrmse(im_true, im, norm_type='Euclidean'):
         denom = im_true.max() - im_true.min()
     elif norm_type == 'mean':
         denom = im_true.mean()
-    return np.sqrt(mse(im_true, im)) / denom
+    else:
+        raise ValueError("Unsupported norm_type")
+    return np.sqrt(mse(im_true, im_test)) / denom
 
 
-def psnr(im_true, im, dynamic_range=None):
+def psnr(im_true, im_test, dynamic_range=None):
     """ Compute the peak signal to noise ratio (PSNR) for an image.
 
     Parameters
     ----------
     im_true : ndarray
         Ground-truth image.
-    im : ndarray
+    im_test : ndarray
         Test image.
     dynamic_range : int
         The dynamic range of the input image (distance between minimum and
@@ -99,10 +97,10 @@ def psnr(im_true, im, dynamic_range=None):
     .. [1] https://en.wikipedia.org/wiki/Peak_signal-to-noise_ratio
 
     """
-    if not im_true.dtype == im.dtype:
+    if not im_true.dtype == im_test.dtype:
         raise ValueError('Input images must have the same dtype.')
 
-    if not im_true.shape == im.shape:
+    if not im_true.shape == im_test.shape:
         raise ValueError('Input images must have the same dimensions.')
 
     if dynamic_range is None:
@@ -110,7 +108,7 @@ def psnr(im_true, im, dynamic_range=None):
         dynamic_range = dmax - dmin
 
     im_true = im_true.astype(np.float64)
-    im = im.astype(np.float64)
+    im_test = im_test.astype(np.float64)
 
-    err = mse(im_true, im)
+    err = mse(im_true, im_test)
     return 10 * np.log10((dynamic_range ** 2) / err)
diff --git a/skimage/measure/tests/test_simple_metrics.py b/skimage/measure/tests/test_simple_metrics.py
index cab25ab7..43122e01 100644
--- a/skimage/measure/tests/test_simple_metrics.py
+++ b/skimage/measure/tests/test_simple_metrics.py
@@ -1,5 +1,6 @@
 import numpy as np
-from numpy.testing import assert_equal, assert_raises, assert_almost_equal
+from numpy.testing import (run_module_suite, assert_equal, assert_raises,
+                           assert_almost_equal)
 
 from skimage.measure import psnr, nrmse
 import skimage.data
@@ -41,3 +42,7 @@ def test_NRMSE():
 
     assert_raises(ValueError, nrmse, x.astype(np.uint8), y)
     assert_raises(ValueError, nrmse, x[:-1], y)
+
+
+if __name__ == "__main__":
+    run_module_suite()