diff --git a/skimage/filter/_thresholding.pyx b/skimage/filter/_thresholding.pyx
index 498580da..c7a80478 100644
--- a/skimage/filter/_thresholding.pyx
+++ b/skimage/filter/_thresholding.pyx
@@ -9,27 +9,27 @@ cimport cython
 def _threshold_adaptive(np.ndarray[np.double_t, ndim=2] image, int block_size,
                         method, double offset, mode, param):
     cdef int r, c
-    cdef np.ndarray[np.double_t, ndim=2] thres_image
+    cdef np.ndarray[np.double_t, ndim=2] thresh_image
 
     if method == 'generic':
-        thres_image = scipy.ndimage.generic_filter(image, param, block_size,
+        thresh_image = scipy.ndimage.generic_filter(image, param, block_size,
             mode=mode)
     elif method == 'gaussian':
         if param is None:
             # automatically determine sigma which covers > 99% of distribution
             sigma = (block_size - 1) / 6.0
-        thres_image = scipy.ndimage.gaussian_filter(image, sigma, mode=mode)
+        thresh_image = scipy.ndimage.gaussian_filter(image, sigma, mode=mode)
     elif method == 'mean':
         mask = 1. / block_size * np.ones((block_size,))
         # separation of filters to speedup convolution
-        thres_image = scipy.ndimage.convolve1d(image, mask, axis=0, mode=mode)
-        thres_image = scipy.ndimage.convolve1d(thres_image, mask, axis=1,
+        thresh_image = scipy.ndimage.convolve1d(image, mask, axis=0, mode=mode)
+        thresh_image = scipy.ndimage.convolve1d(thresh_image, mask, axis=1,
             mode=mode)
     elif method == 'median':
-        thres_image = scipy.ndimage.median_filter(image, block_size, mode=mode)
+        thresh_image = scipy.ndimage.median_filter(image, block_size, mode=mode)
 
     for r in range(image.shape[0]):
         for c in range(image.shape[1]):
-            thres_image[r,c] = image[r,c] > (thres_image[r,c] - offset)
+            thresh_image[r,c] = image[r,c] > (thresh_image[r,c] - offset)
 
-    return thres_image.astype('bool')
+    return thresh_image.astype('bool')