參考文獻(xiàn):
《基于卷積神經(jīng)網(wǎng)絡(luò)的人臉識別研究》 李春利，柳振東，惠康華
文章中基于經(jīng)典的網(wǎng)絡(luò)LeNet-5的結(jié)構(gòu)，提出了一種適用于ORL數(shù)據(jù)集的CNN結(jié)構(gòu)，在該數(shù)據(jù)集上取得了較高的識別率。
本文是在參考此論文的基礎(chǔ)上，使用tensorflow實(shí)現(xiàn)了文中相關(guān)理論。
訓(xùn)練集下載解壓后可以看到，ORL訓(xùn)練集一共有40類，每一類有10張bmp類型的圖片。
目錄
s1中圖片
首先我們需要做的就是將這些數(shù)據(jù)讀入，制作我們自己的訓(xùn)練集和測試集。
input_path = "./orl"
train_path = "./train"
test_path = "./test"
if not os.path.exists(train_path):
os.mkdir(train_path)
if not os.path.exists(test_path):
os.mkdir(test_path)
for i in range(1, 41):
if not os.path.exists(train_path + '/' + str(i)):
os.mkdir(train_path + '/' + str(i))
if not os.path.exists(test_path + '/' + str(i)):
os.mkdir(test_path + '/' + str(i))
# 生成訓(xùn)練和測試的數(shù)據(jù)
def generate_data(train_path, test_path):
index = 1
output_index = 1
for (dirpath, dirnames, filenames) in os.walk(input_path):
# 打亂文件列表，相當(dāng)于是隨機(jī)選取8張訓(xùn)練集，2張測試
random.shuffle(filenames)
for filename in filenames:
if filename.endswith('.bmp'):
img_path = dirpath + '/' + filename
# 使用opencv 讀取圖片
img_data = cv2.imread(img_path)
# 按照論文中的將圖片大小調(diào)整為28 * 28
img_data = cv2.resize(img_data, (28, 28), interpolation=cv2.INTER_AREA)
if index < 3:
cv2.imwrite(test_path + '/' + str(output_index) + '/' + str(index) + '.jpg', img_data)
index += 1
elif 10 >= index >= 3:
cv2.imwrite(train_path + '/' + str(output_index) + '/' + str(index) + '.jpg', img_data)
index += 1
if index > 10:
output_index += 1
index = 1
運(yùn)行完后我們便得到了320張訓(xùn)練集，80張測試集，所得的樣本都是通過隨機(jī)選取。
訓(xùn)練集：
訓(xùn)練集
第一類別
測試集：
image.png
將train和test寫入到tfrecord的同時(shí)進(jìn)行標(biāo)注
# 生成整數(shù)型的屬性
def _int64_feature(value):
return tf.train.Feature(int64_list=tf.train.Int64List(value=[value]))
# 生成字符串類型
def _bytes_feature(value):
return tf.train.Feature(bytes_list=tf.train.BytesList(value=[value]))
train_path = "./train/"
test_path = "./test/"
classes = {i: i for i in range(1, 41)}
writer_train = tf.python_io.TFRecordWriter("orl_train.tfrecords")
writer_test = tf.python_io.TFRecordWriter("orl_test.tfrecords")
def generate():
# 遍歷字典
for index, name in enumerate(classes):
train = train_path + str(name) + '/'
test = test_path + str(name) + '/'
for img_name in os.listdir(train):
img_path = train + img_name # 每一個(gè)圖片的地址
img = cv2.imread(img_path)
img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
img_raw = img.tobytes()
example = tf.train.Example(features=tf.train.Features(feature={
'label': _int64_feature(index + 1),
'img_raw': _bytes_feature(img_raw)
}))
writer_train.write(example.SerializeToString())
for img_name in os.listdir(test):
img_path = test + img_name # 每一個(gè)圖片的地址
img = cv2.imread(img_path)
img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
img_raw = img.tobytes()
example = tf.train.Example(features=tf.train.Features(feature={
'label': _int64_feature(index + 1),
'img_raw': _bytes_feature(img_raw)
}))
writer_test.write(example.SerializeToString())
writer_test.close()
writer_train.close()
接下來開始訓(xùn)練：
def train(data, label):
x = tf.placeholder(tf.float32,
[BATCH_SIZE, SIZE, SIZE, orl_inference.NUM_CHANNELS],
name='x-input')
y_ = tf.placeholder(tf.float32, [None, orl_inference.OUTPUT_NODE], name='y-output')
# 使用L2正則化計(jì)算損失函數(shù)
regularizer = tf.contrib.layers.l2_regularizer(REGULARIZATION_RATE)
min_after_dequeue = 100
capacity = min_after_dequeue + 3 * BATCH_SIZE
image_batch, label_batch = tf.train.shuffle_batch(
[data, label], batch_size=BATCH_SIZE,
capacity=capacity, min_after_dequeue=min_after_dequeue
)
y = orl_inference.inference(x, False, regularizer)
global_step = tf.Variable(0, trainable=False)
variable_averages = tf.train.ExponentialMovingAverage(
MOVING_AVERAGE_DECAY, global_step
)
variable_averages_op = variable_averages.apply(tf.trainable_variables())
# 計(jì)算交叉熵作為刻畫預(yù)測值和真實(shí)值之間的損失函數(shù)
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(logits=y, labels=tf.argmax(y_, 1))
# 計(jì)算所有樣例中交叉熵的平均值
cross_entropy_mean = tf.reduce_mean(cross_entropy)
# 總損失等于交叉熵?fù)p失和正則化損失的和
loss = cross_entropy_mean + tf.add_n(tf.get_collection('losses'))
# 設(shè)置指數(shù)衰減的學(xué)習(xí)率
learning_rate = tf.train.exponential_decay(
LEARNING_RATE_BASE,
global_step,
320 / BATCH_SIZE,
LEARNING_RATE_DECAY,
staircase=True
)
# 優(yōu)化損失函數(shù)
train_step = tf.train.GradientDescentOptimizer(learning_rate) \
.minimize(loss, global_step=global_step)
with tf.control_dependencies([train_step, variable_averages_op]):
train_op = tf.no_op(name='train')
saver = tf.train.Saver()
# 驗(yàn)證
# accuracy = tf.reduce_mean()
with tf.Session() as sess:
tf.global_variables_initializer().run()
coord = tf.train.Coordinator()
threads = tf.train.start_queue_runners(sess=sess, coord=coord)
# 迭代的訓(xùn)練網(wǎng)絡(luò)
for i in range(TRAINING_STEPS):
xs, ys = sess.run([image_batch, label_batch])
xs = xs / 255.0
reshaped_xs = np.reshape(xs, (BATCH_SIZE,
SIZE,
SIZE,
orl_inference.NUM_CHANNELS))
# 將圖像和標(biāo)簽數(shù)據(jù)通過tf.train.shuffle_batch整理成訓(xùn)練時(shí)需要的batch
ys = get_label(ys)
_, loss_value, step = sess.run([train_op, loss, global_step],
feed_dict={x: reshaped_xs, y_: ys})
if i % 100 == 0:
# 每10輪輸出一次在訓(xùn)練集上的測試結(jié)果
acc = loss.eval({x: reshaped_xs, y_: ys})
print("After %d training step[s], loss on training"
" batch is %g. " % (step, loss_value))
saver.save(
sess, os.path.join(MODEL_SAVE_PATH, MODEL_NAME),
global_step=global_step
)
# logit = orl_inference.inference(image_batch)
coord.request_stop()
coord.join(threads)
訓(xùn)練過程
進(jìn)行驗(yàn)證：
def evaluate():
with tf.Graph().as_default() as g:
filename_queue = tf.train.string_input_producer(["orl_test.tfrecords"])
reader = tf.TFRecordReader()
_, serialized_example = reader.read(filename_queue)
features = tf.parse_single_example(serialized_example,
features={
'label': tf.FixedLenFeature([], tf.int64),
'img_raw': tf.FixedLenFeature([], tf.string),
})
img = tf.decode_raw(features['img_raw'], tf.uint8)
img = tf.reshape(img, [28, 28, 1])
label = tf.cast(features['label'], tf.int32)
min_after_dequeue = 100
capacity = min_after_dequeue + 3 * 200
image_batch, label_batch = tf.train.shuffle_batch(
[img, label], batch_size=80,
capacity=capacity, min_after_dequeue=min_after_dequeue
)
x = tf.placeholder(tf.float32,
[80,
orl_inference.IMAGE_SIZE,
orl_inference.IMAGE_SIZE,
orl_inference.NUM_CHANNELS],
name='x-input')
y_ = tf.placeholder(
tf.float32, [None, orl_inference.OUTPUT_NODE], name='y-input'
)
y = orl_inference.inference(x, None, None)
correct_prediction = tf.equal(tf.argmax(y, 1), tf.argmax(y_, 1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
variable_averages = tf.train.ExponentialMovingAverage(
orl_train.MOVING_AVERAGE_DECAY
)
variable_to_restore = variable_averages.variables_to_restore()
saver = tf.train.Saver(variable_to_restore)
# 每隔EVAL_INTERVAL_SECS秒調(diào)用一次
while True:
with tf.Session() as sess:
test = cv2.imread('./data/20/10.jpg')
test = cv2.cvtColor(test, cv2.COLOR_BGR2GRAY)
test = np.array(test)
test = test / 255.0
test_re = np.reshape(test, (1, 28, 28, 1))
coord = tf.train.Coordinator()
threads = tf.train.start_queue_runners(sess=sess, coord=coord)
xs, ys = sess.run([image_batch, label_batch])
ys = get_label(ys)
xs = xs / 255.0
validate_feed = {x: xs,
y_: ys}
cpkt = tf.train.get_checkpoint_state(
orl_train.MODEL_SAVE_PATH
)
if cpkt and cpkt.model_checkpoint_path:
# 加載模型
saver.restore(sess, cpkt.model_checkpoint_path)
# 通過文件名得到模型保存時(shí)迭代的輪數(shù)
global_step = cpkt.model_checkpoint_path \
.split('/')[-1].split('-')[-1]
# result = sess.run(y, feed_dict={x: test_re})
# re = np.where(result == np.max(result))
# ss = tf.argmax(result, 1)
# tt = np.argmax(result, 1)
# print('result is %d'%(tt[0] + 1))
# # print('hehe')
accuracy_score = sess.run(accuracy,feed_dict=validate_feed)
print("After %s training steps, validation "
"accuracy = %g" % (global_step, accuracy_score))
else:
print("No checkpoint file found")
return
time.sleep(EVAL_INTERVAL_SECS)
驗(yàn)證結(jié)果
此次實(shí)驗(yàn)參考了《Tensorflow 實(shí)戰(zhàn)Google深度學(xué)習(xí)框架》這本書的內(nèi)容，根據(jù)所學(xué)內(nèi)容，將文獻(xiàn)中的實(shí)驗(yàn)實(shí)踐了一遍，也算是加深了理解。
完整代碼：聽說star的人會變帥
