From 03f6da135baa6fe058c7fd3c8fba88b9b13bda0f Mon Sep 17 00:00:00 2001
From: Stefan van der Walt <stefan@sun.ac.za>
Date: Thu, 2 Feb 2012 22:41:54 -0800
Subject: [PATCH] DOC: Example of structural similarity.  The example is
 currently broken because structural_similarity does not auto-detect dynamic
 range.

---
 doc/examples/plot_ssim.py | 72 +++++++++++++++++++++++++++++++++++++++
 1 file changed, 72 insertions(+)
 create mode 100644 doc/examples/plot_ssim.py

diff --git a/doc/examples/plot_ssim.py b/doc/examples/plot_ssim.py
new file mode 100644
index 00000000..761dc29c
--- /dev/null
+++ b/doc/examples/plot_ssim.py
@@ -0,0 +1,72 @@
+'''
+===========================
+Structural similarity index
+===========================
+
+When comparing images, the mean squared error (MSE)--while simple to
+implement--is not highly indicative of perceived similarity.  Structural
+similarity aims to address this shortcoming by taking texture into account
+[1]_, [2]_.
+
+The example shows two modifications of the input image, each with the same MSE,
+but with very different mean structural similarity indices.
+
+.. [1] Zhou Wang; Bovik, A.C.; ,"Mean squared error: Love it or leave it? A new
+       look at Signal Fidelity Measures," Signal Processing Magazine, IEEE,
+       vol. 26, no. 1, pp. 98-117, Jan. 2009.
+
+.. [2] Z. Wang, A. C. Bovik, H. R. Sheikh and E. P. Simoncelli, "Image quality
+       assessment: From error visibility to structural similarity," IEEE
+       Transactions on Image Processing, vol. 13, no. 4, pp. 600-612,
+       Apr. 2004.
+
+'''
+
+from skimage import data, color, io, exposure, img_as_float
+from skimage.measure import structural_similarity as ssim
+
+import numpy as np
+
+img = img_as_float(data.camera())
+rows, cols = img.shape
+
+noise = np.ones_like(img) * 0.5 * (img.max() - img.min())
+noise[np.random.random(size=noise.shape) > 0.5] *= -1
+
+def mse(x, y):
+    return np.linalg.norm(x - y)
+
+img_noise = img + noise
+img_const = img + abs(noise)
+
+import matplotlib.pyplot as plt
+f, (ax0, ax1, ax2) = plt.subplots(1, 3)
+
+mse_none = mse(img, img)
+ssim_none = ssim(img, img)
+
+mse_noise = mse(img, img_noise)
+ssim_noise = ssim(img, img_noise)
+
+mse_const = mse(img, img_const)
+ssim_const = ssim(img, img_const)
+
+label = 'MSE: %2.f, SSIM: %.2f'
+
+ax0.imshow(img, cmap=plt.cm.gray)
+ax0.set_xlabel(label % (mse_none, ssim_none))
+ax0.set_title('Original image')
+
+# exposure.rescale_intensity(img_noise)
+img_noise -= img_noise.min()
+img_noise /= img_noise.max()
+
+ax1.imshow(img_noise, cmap=plt.cm.gray)
+ax1.set_xlabel(label % (mse_noise, ssim_noise))
+ax1.set_title('Image with noise')
+
+ax2.imshow(img_const, cmap=plt.cm.gray)
+ax2.set_xlabel(label % (mse_const, ssim_const))
+ax2.set_title('Image plus constant')
+
+plt.show()