imageIO

import dlib
import numpy
from skimage import io as ImageIO
from skimage.color import rgb2gray
from skimage.transform import resize
from scipy.ndimage import gaussian_filter


def extract_faces(filename, face_detector, face_size=100, padding=25, blur=True):
    img = gaussian_filter(ImageIO.imread(filename), sigma=1) if blur else ImageIO.imread(filename)
    #   make sure the image is grayscale
    detector = face_detector(img, 1)
    #   at a max we allocate at most 10 faces
    if len(img.shape) == 3:
        img = rgb2gray(img)
    faces = numpy.zeros((10, face_size, face_size), dtype='float32')
    # lets keep a counter so we know when to cut off our face array
    counter = 0
    for i, j in enumerate(detector):
        faces[i] = resize(img[j.top()-padding:
                                            j.bottom()+padding,
                                            j.left()-padding:
                                            j.right()+padding],
                     output_shape=(face_size, face_size),
                     preserve_range=True)
        counter += 1
    return numpy.asarray(faces[0:counter], dtype='float32') / 255