attack_iter.py

"""Implementation of sample attack."""

from __future__ import absolute_import
from __future__ import division
from __future__ import print_function

import os

import numpy as np
from scipy.misc import imread
from scipy.misc import imsave

import tensorflow as tf
from nets import inception_v3, inception_v4, inception_resnet_v2, resnet_v2

slim = tf.contrib.slim


tf.flags.DEFINE_string(
    'master', '', 'The address of the TensorFlow master to use.')

tf.flags.DEFINE_string(
    'checkpoint_path_inception_v3', '', 'Path to checkpoint for inception network.')

tf.flags.DEFINE_string(
    'checkpoint_path_adv_inception_v3', '', 'Path to checkpoint for inception network.')

tf.flags.DEFINE_string(
    'checkpoint_path_ens3_adv_inception_v3', '', 'Path to checkpoint for inception network.')

tf.flags.DEFINE_string(
    'checkpoint_path_ens4_adv_inception_v3', '', 'Path to checkpoint for inception network.')

tf.flags.DEFINE_string(
    'checkpoint_path_inception_v4', '', 'Path to checkpoint for inception network.')

tf.flags.DEFINE_string(
    'checkpoint_path_inception_resnet_v2', '', 'Path to checkpoint for inception network.')

tf.flags.DEFINE_string(
    'checkpoint_path_ens_adv_inception_resnet_v2', '', 'Path to checkpoint for inception network.')

tf.flags.DEFINE_string(
    'checkpoint_path_resnet', '', 'Path to checkpoint for inception network.')

tf.flags.DEFINE_string(
    'input_dir', '', 'Input directory with images.')

tf.flags.DEFINE_string(
    'output_dir', '', 'Output directory with images.')

tf.flags.DEFINE_float(
    'max_epsilon', 16.0, 'Maximum size of adversarial perturbation.')

tf.flags.DEFINE_integer(
    'num_iter', 10, 'Number of iterations.')

tf.flags.DEFINE_integer(
    'image_width', 299, 'Width of each input images.')

tf.flags.DEFINE_integer(
    'image_height', 299, 'Height of each input images.')

tf.flags.DEFINE_integer(
    'batch_size', 10, 'How many images process at one time.')

tf.flags.DEFINE_float(
    'momentum', 1.0, 'Momentum.')

FLAGS = tf.flags.FLAGS


def load_images(input_dir, batch_shape):
  """Read png images from input directory in batches.

  Args:
    input_dir: input directory
    batch_shape: shape of minibatch array, i.e. [batch_size, height, width, 3]

  Yields:
    filenames: list file names without path of each image
      Lenght of this list could be less than batch_size, in this case only
      first few images of the result are elements of the minibatch.
    images: array with all images from this batch
  """
  images = np.zeros(batch_shape)
  filenames = []
  idx = 0
  batch_size = batch_shape[0]
  for filepath in tf.gfile.Glob(os.path.join(input_dir, '*.png')):
    with tf.gfile.Open(filepath) as f:
      image = imread(f, mode='RGB').astype(np.float) / 255.0
    # Images for inception classifier are normalized to be in [-1, 1] interval.
    images[idx, :, :, :] = image * 2.0 - 1.0
    filenames.append(os.path.basename(filepath))
    idx += 1
    if idx == batch_size:
      yield filenames, images
      filenames = []
      images = np.zeros(batch_shape)
      idx = 0
  if idx > 0:
    yield filenames, images


def save_images(images, filenames, output_dir):
  """Saves images to the output directory.

  Args:
    images: array with minibatch of images
    filenames: list of filenames without path
      If number of file names in this list less than number of images in
      the minibatch then only first len(filenames) images will be saved.
    output_dir: directory where to save images
  """
  for i, filename in enumerate(filenames):
    # Images for inception classifier are normalized to be in [-1, 1] interval,
    # so rescale them back to [0, 1].
    with tf.gfile.Open(os.path.join(output_dir, filename), 'w') as f:
      imsave(f, (images[i, :, :, :] + 1.0) * 0.5, format='png')


def graph(x, y, i, x_max, x_min, grad):
  eps = 2.0 * FLAGS.max_epsilon / 255.0
  num_iter = FLAGS.num_iter
  alpha = eps / num_iter
  momentum = FLAGS.momentum
  num_classes = 1001

  with slim.arg_scope(inception_v3.inception_v3_arg_scope()):
    logits_v3, end_points_v3 = inception_v3.inception_v3(
        x, num_classes=num_classes, is_training=False)

  with slim.arg_scope(inception_v3.inception_v3_arg_scope()):
    logits_adv_v3, end_points_adv_v3 = inception_v3.inception_v3(
        x, num_classes=num_classes, is_training=False, scope='AdvInceptionV3')

  with slim.arg_scope(inception_v3.inception_v3_arg_scope()):
    logits_ens3_adv_v3, end_points_ens3_adv_v3 = inception_v3.inception_v3(
        x, num_classes=num_classes, is_training=False, scope='Ens3AdvInceptionV3')

  with slim.arg_scope(inception_v3.inception_v3_arg_scope()):
    logits_ens4_adv_v3, end_points_ens4_adv_v3 = inception_v3.inception_v3(
        x, num_classes=num_classes, is_training=False, scope='Ens4AdvInceptionV3')

  with slim.arg_scope(inception_v4.inception_v4_arg_scope()):
    logits_v4, end_points_v4 = inception_v4.inception_v4(
        x, num_classes=num_classes, is_training=False)

  with slim.arg_scope(inception_resnet_v2.inception_resnet_v2_arg_scope()):
    logits_res_v2, end_points_res_v2 = inception_resnet_v2.inception_resnet_v2(
        x, num_classes=num_classes, is_training=False)

  with slim.arg_scope(inception_resnet_v2.inception_resnet_v2_arg_scope()):
    logits_ensadv_res_v2, end_points_ensadv_res_v2 = inception_resnet_v2.inception_resnet_v2(
        x, num_classes=num_classes, is_training=False, scope='EnsAdvInceptionResnetV2')

  with slim.arg_scope(resnet_v2.resnet_arg_scope()):
    logits_resnet, end_points_resnet = resnet_v2.resnet_v2_101(
        x, num_classes=num_classes, is_training=False)
            
  pred = tf.argmax(end_points_v3['Predictions'] + end_points_adv_v3['Predictions'] + end_points_ens3_adv_v3['Predictions'] + \
                  end_points_ens4_adv_v3['Predictions'] + end_points_v4['Predictions'] + \
                  end_points_res_v2['Predictions'] + end_points_ensadv_res_v2['Predictions'] + end_points_resnet['predictions'], 1)

  first_round = tf.cast(tf.equal(i, 0), tf.int64)
  y = first_round * pred + (1 - first_round) * y
  one_hot = tf.one_hot(y, num_classes)

  logits = (logits_v3 + 0.25 * logits_adv_v3 + logits_ens3_adv_v3 + \
           logits_ens4_adv_v3 + logits_v4 + \
           logits_res_v2 + logits_ensadv_res_v2 + logits_resnet) / 7.25
  auxlogits = (end_points_v3['AuxLogits'] + 0.25 * end_points_adv_v3['AuxLogits'] + end_points_ens3_adv_v3['AuxLogits'] + \
              end_points_ens4_adv_v3['AuxLogits'] + end_points_v4['AuxLogits'] + \
              end_points_res_v2['AuxLogits'] + end_points_ensadv_res_v2['AuxLogits']) / 6.25
  cross_entropy = tf.losses.softmax_cross_entropy(one_hot,
                                                  logits,
                                                  label_smoothing=0.0,
                                                  weights=1.0)
  cross_entropy += tf.losses.softmax_cross_entropy(one_hot,
                                                   auxlogits,
                                                   label_smoothing=0.0,
                                                   weights=0.4)
  noise = tf.gradients(cross_entropy, x)[0]
  noise = noise / tf.reduce_mean(tf.abs(noise), [1,2,3], keep_dims=True)
  noise = momentum * grad + noise
  x = x + alpha * tf.sign(noise)
  x = tf.clip_by_value(x, x_min, x_max)
  i = tf.add(i, 1)
  return x, y, i, x_max, x_min, noise

def stop(x, y, i, x_max, x_min, grad):
  num_iter = FLAGS.num_iter
  return tf.less(i, num_iter)


def main(_):
  # Images for inception classifier are normalized to be in [-1, 1] interval,
  # eps is a difference between pixels so it should be in [0, 2] interval.
  # Renormalizing epsilon from [0, 255] to [0, 2].
  eps = 2.0 * FLAGS.max_epsilon / 255.0

  batch_shape = [FLAGS.batch_size, FLAGS.image_height, FLAGS.image_width, 3]

  tf.logging.set_verbosity(tf.logging.INFO)

  with tf.Graph().as_default():
    # Prepare graph
    x_input = tf.placeholder(tf.float32, shape=batch_shape)
    x_max = tf.clip_by_value(x_input + eps, -1.0, 1.0)
    x_min = tf.clip_by_value(x_input - eps, -1.0, 1.0)

    y = tf.constant(np.zeros([FLAGS.batch_size]), tf.int64)
    i = tf.constant(0)
    grad = tf.zeros(shape=batch_shape)
    x_adv, _, _, _, _, _ = tf.while_loop(stop, graph, [x_input, y, i, x_max, x_min, grad])

    # Run computation
    s1 = tf.train.Saver(slim.get_model_variables(scope='InceptionV3'))
    s2 = tf.train.Saver(slim.get_model_variables(scope='AdvInceptionV3'))
    s3 = tf.train.Saver(slim.get_model_variables(scope='Ens3AdvInceptionV3'))
    s4 = tf.train.Saver(slim.get_model_variables(scope='Ens4AdvInceptionV3'))
    s5 = tf.train.Saver(slim.get_model_variables(scope='InceptionV4'))
    s6 = tf.train.Saver(slim.get_model_variables(scope='InceptionResnetV2'))
    s7 = tf.train.Saver(slim.get_model_variables(scope='EnsAdvInceptionResnetV2'))
    s8 = tf.train.Saver(slim.get_model_variables(scope='resnet_v2'))

    with tf.Session() as sess:
      s1.restore(sess, FLAGS.checkpoint_path_inception_v3)
      s2.restore(sess, FLAGS.checkpoint_path_adv_inception_v3)
      s3.restore(sess, FLAGS.checkpoint_path_ens3_adv_inception_v3)
      s4.restore(sess, FLAGS.checkpoint_path_ens4_adv_inception_v3)
      s5.restore(sess, FLAGS.checkpoint_path_inception_v4)
      s6.restore(sess, FLAGS.checkpoint_path_inception_resnet_v2)
      s7.restore(sess, FLAGS.checkpoint_path_ens_adv_inception_resnet_v2)
      s8.restore(sess, FLAGS.checkpoint_path_resnet)

      for filenames, images in load_images(FLAGS.input_dir, batch_shape):
        adv_images = sess.run(x_adv, feed_dict={x_input: images})
        save_images(adv_images, filenames, FLAGS.output_dir)

if __name__ == '__main__':
  tf.app.run()