Hands-On Image Processing with Python

Eigenfaces

Given the property of PCA, the PCs computed are orthogonal to each other and each of them contains 4,096 pixels—and can be reshaped to a 64 x 64 image. These principal components are known as eigenfaces (since they are also the eigenvectors).

As can be seen, they represent certain properties of the faces. The following code block displays some of the computed eigenfaces:

fig = plt.figure(figsize=(5,2)) fig.subplots_adjust(left=0, right=1, bottom=0, top=1, hspace=0.05, wspace=0.05) # plot the first 10 eigenfacesfor i in range(10):     ax = fig.add_subplot(2, 5, i+1, xticks=[], yticks=[])     ax.imshow(np.reshape(pipeline.named_steps['pca'].components_[i,:], (64,64)), cmap=plt.cm.bone, interpolation='nearest')

The following screenshot  ...