In [1]:

    

%matplotlib inline
import os,sys,re
from matplotlib import cm
import matplotlib.pyplot as plt
import numpy as np
from sklearn.cross_validation import train_test_split
read_path = '/home/mckc/Face oNN/Complete Face Data/'
#read_path = '/home/mckc/Face oNN/Complete Face Data/'
save_path = '/home/mckc/new/'
os.chdir(read_path)
os.getcwd()
folders = os.listdir(os.getcwd())


    

In [2]:

    

def PCA_comp(data):
    from time import *
    from sklearn.decomposition import PCA
    n_components = 150
    
    h,w = 96,96

    print("Extracting the top %d eigenfaces from %d faces"
      % (n_components, image_data.shape[0]))
    t0 = time()
    pca = PCA(n_components=n_components, whiten=True).fit(images_geq)
    print("done in %0.3fs" % (time() - t0))

    #eigenfaces = pca.components_.reshape((n_components, h, w))

    print("Projecting the input data on the eigenfaces orthonormal basis")
    t0 = time()
    X = pca.transform(data)
    print X.shape
    print("done in %0.3fs" % (time() - t0))
    return(X)
#K-mediods file
import numpy as np
import random

def kMedoids(D, k, tmax=100):
    # determine dimensions of distance matrix D
    m, n = D.shape

    if k > n:
        raise Exception('too many medoids')
    # randomly initialize an array of k medoid indices
    M = np.arange(n)
    np.random.shuffle(M)
    M = np.sort(M[:k])

    # create a copy of the array of medoid indices
    Mnew = np.copy(M)

    # initialize a dictionary to represent clusters
    C = {}
    for t in xrange(tmax):
        # determine clusters, i. e. arrays of data indices
        J = np.argmin(D[:,M], axis=1)
        for kappa in range(k):
            C[kappa] = np.where(J==kappa)[0]
        # update cluster medoids
        for kappa in range(k):
            J = np.mean(D[np.ix_(C[kappa],C[kappa])],axis=1)
            j = np.argmin(J)
            Mnew[kappa] = C[kappa][j]
        np.sort(Mnew)
        # check for convergence
        if np.array_equal(M, Mnew):
            break
        M = np.copy(Mnew)
    else:
        # final update of cluster memberships
        J = np.argmin(D[:,M], axis=1)
        for kappa in range(k):
            C[kappa] = np.where(J==kappa)[0]

# return results
    return M, C


    

    




<ipython-input-2-13e1dc72abe0>:1: SyntaxWarning: import * only allowed at module level
  def PCA_comp(data):

In [157]:

    

folder = folders[86]
print(folder)

import cv2
from cv2 import resize
from skimage import exposure
from skimage import exposure
from skimage.morphology import disk
from skimage.filters import rank
selem = disk(30)
size = []
images_geq = []
images_eql = []
images = []

os.chdir(read_path+folder)
files = os.listdir(read_path+folder)
#files = os.listdir(read_path)

for i in files:
    image = resize(cv2.imread(i,0),(96,96))
    #image = cv2.imread(i,0)
    # Global equalize
    image_g_eq = exposure.equalize_hist(image)

    # Equalization
    image_eql = rank.equalize(image, selem=selem)
    images.append(image)
    images_geq.append(image_g_eq)
    images_eql.append(image_eql)
    size.append(image.shape[0])
size = np.array(size)
image_data = np.array(images).reshape(-1,9216)
images_geq = np.array(images_geq).reshape(-1,9216)
images_eql = np.array(images_eql).reshape(-1,9216)
X  = PCA_comp(images_geq)
X.shape

from scipy.cluster.vq import kmeans
from scipy.spatial.distance import cdist,pdist

##### cluster data into K=1..20 clusters #####
K_MAX = 80
KK = range(1,K_MAX,4)

KM = [kmeans(X[:,:],k,iter=50) for k in KK]
centroids = [cent for (cent,var) in KM]
D_k = [cdist(X[:,:], cent, 'euclidean') for cent in centroids]
cIdx = [np.argmin(D,axis=1) for D in D_k]
dist = [np.min(D,axis=1) for D in D_k]

tot_withinss = [sum(d**2) for d in dist]  # Total within-cluster sum of squares
totss = sum(pdist(X)**2)/X.shape[0]       # The total sum of squares
betweenss = totss - tot_withinss      

##### plots #####
kIdx = 1        # K=10
# elbow curve
fig = plt.figure()
ax = fig.add_subplot(111)
ax.plot(KK, betweenss/totss*100, 'b*-')
ax.plot(KK[kIdx], betweenss[kIdx]/totss*100, marker='o', markersize=12, 
    markeredgewidth=2, markeredgecolor='r', markerfacecolor='None')
ax.set_ylim((0,100))
plt.grid(True)
plt.xlabel('Number of clusters')
plt.ylabel('Percentage of variance explained (%)')
plt.title('Elbow for KMeans clustering')


    

    




himanshu
Extracting the top 150 eigenfaces from 76 faces
done in 0.021s
Projecting the input data on the eigenfaces orthonormal basis
(76, 76)
done in 0.003s






    
Out[157]:





<matplotlib.text.Text at 0x7f5d4a2cfc10>






    

In [156]:

    

from sklearn.metrics.pairwise import pairwise_distances
D = pairwise_distances(X, metric='euclidean')

# split into 2 clusters
M, C = kMedoids(D, 45)
cleaned = image_data[M,].reshape(-1,96,96)

os.chdir(save_path+folder)
import scipy.misc
for i in range(cleaned.shape[0]):
    scipy.misc.toimage(cleaned[i]).save(str(folder+'_'+str(i)) +'.jpg')
os.chdir(read_path)


    

In [ ]: