docs: add description of highlights and perf

README.md

@@ -2,6 +2,30 @@
 
 # fastmorph: multilabel 3D morphological image processing functions.
 
+This is a collection of morphological 3D image operations that are tuned for working with dense 3D labeled images. 
+
+We provide the following multithreaded operations:
+
+- Dilation
+- Erosion 
+- Opening 
+- Closing
+- Spherical Dilation (binary images)
+- Fill Voids
+
+Highlights compared to other libraries:
+
+- Handles multi-labeled images
+- Multithreaded
+- High performance single-threaded
+- Low memory usage
+- Dilate computes mode of surrounding labels
+
+Disadvantages versus other libraries:
+
+- Not ideal for grayscale images
+- Stencil (structuring element) is fixed size 3x3x3 and all on.
+
 
 ```python
 import fastmorph
@@ -45,3 +69,25 @@ morphed = fastmorph.spherical_erode(labels, radius=1, parallel=2, anisotropy=(1,
 filled_labels, ct = fastmorph.fill_holes(labels, return_fill_count=True, remove_enclosed=False)
 ```
 
+## Performance
+
+A test run on an M1 Macbook Pro on `connectomics.npy.ckl`, a 512<sup>3</sup> volume with over 2000 dense labels had the following results.
+
+```
+erode / 1 thread:  7.70 sec
+erode / 2 threads:  4.10 sec
+erode / 4 threads:  3.02 sec
+dilate / background_only=True / 1 thread:  1.15 sec
+dilate / background_only=True / 2 threads:  0.65 sec
+dilate / background_only=True / 4 threads:  0.48 sec
+dilate / background_only=False / 1 thread:  14.31 sec
+dilate / background_only=False / 2 threads:  7.47 sec
+dilate / background_only=False / 4 threads:  8.09 sec
+dilate / background_only=False / 8 threads:  5.46 sec
+skimage expand_labels / 1 thread:  75.20 sec
+```
+
+
+
+
+