better clustering

diff --git a/imago_pack/intrsc.py b/imago_pack/intrsc.py
index ce89cb2..04ef350 100644 (file)
--- a/imago_pack/intrsc.py
+++ b/imago_pack/intrsc.py
@@ -50,12 +50,16 @@ def board(image, lines, show_all, do_something):
     if show_all:
         import matplotlib.pyplot as pyplot
         pyplot.scatter(luma, saturation, 
-                       color=[(s[2][0]/255., s[2][1]/255., s[2][2]/255., 1.) 
-                              for s in board_raw])
+                       color=[(s[2][0]/255.,
+                               s[2][1]/255.,
+                               s[2][2]/255., 1.) 
+                                   for s in board_raw])
+        pyplot.xlim(0,1)
+        pyplot.ylim(0,1)
         pyplot.show()
 
     clusters = k_means.cluster(3, 2,zip(zip(luma, saturation), range(len(luma))),
-                               [[0., 0.], [0.5, 0.5], [1., 1.]])
+                               [[0., 0.], [0.5, 0.5], [1., 0.]])
    #clusters.sort(key=mean_luma)
 
     if show_all:
@@ -65,6 +69,8 @@ def board(image, lines, show_all, do_something):
                                                  color=(0,1,0,1))
         pyplot.scatter([d[0][0] for d in clusters[2]], [d[0][1] for d in clusters[2]],
                                                  color=(0,0,1,1))
+        pyplot.xlim(0,1)
+        pyplot.ylim(0,1)
         pyplot.show()
 
     clusters[0] = [(p[1], 'B') for p in clusters[0]]
@@ -109,12 +115,12 @@ def intersections_from_angl_dist(lines, size, get_all=True):
     return intersections
    
 def RGBtoSat(c):
+    """Using the HSI color model."""
     max_diff = max(c) - min(c)
     if max_diff == 0:
         return 0
     else:
-        #TODO simplify this
-        return max_diff / float(255. - abs(max(c) + min(c) - 255))
+        return 1. - ((3. * min(c)) / sum(c)) 
 
 def stone_color_raw(image, (x, y)):
     """Given image and coordinates, return stone color."""
@@ -127,7 +133,7 @@ def stone_color_raw(image, (x, y)):
                 pass
     luma = sum([0.30 * sum(s[0] for s in suma) / 25., 0.59 * sum(s[1] for s in suma) / 25., 
             0.11 * sum(s[2] for s in suma) / 25.]) / 255.
-    saturation = sum(RGBtoSat(s) for s in suma) / (25. * 255.)
+    saturation = sum(RGBtoSat(s) for s in suma) / 25.
     color = [sum(s[0] for s in suma) / 25., sum(s[1] for s in suma) / 25.,
              sum(s[2] for s in suma) / 25.]
     return luma, saturation, color

diff --git a/imago_pack/k_means.py b/imago_pack/k_means.py
index 0019597..c5ddba5 100644 (file)
--- a/imago_pack/k_means.py
+++ b/imago_pack/k_means.py
@@ -1,8 +1,9 @@
-"""K-means module"""
+"""K-means module."""
 
 import random
 
 def cluster(k, d, data, i_centers=None):
+    """Find *k* clusters on *d* dimensional *data*."""
     borders = [(min(p[0][i] for p in data), max(p[0][i] for p in data))
                for i in range(d) ]
     if i_centers:
@@ -39,5 +40,5 @@ def centroid(cluster):
     return [sum(c[0][i] for c in cluster) / l for i in range(d)]
 
 def delta(c1, c2):
-    return sum( (sum(abs(cc1 - cc2) for (cc1, cc2) in zip (ccc1, ccc2)) if ccc2
+    return sum((sum(abs(cc1 - cc2) for (cc1, cc2) in zip (ccc1, ccc2)) if ccc2
                else 0.) for (ccc1, ccc2) in zip(c1, c2))