图像去噪-MNIST

图像去噪-MNIST

from keras.layers import Input, Dense
from keras.models import Model
from keras.datasets import mnist
import numpy as np

# 加载mnist数据集并对其进行处理
(x_train, _), (x_test, _) = mnist.load_data()

x_train = x_train.astype('float32') / 255.
x_test = x_test.astype('float32') / 255.
np.prod(x_test.shape[1:])))
print(x_train.shape)
print(x_test.shape)
#对mnist数据集加噪
def noisy(image):
    print('min {} , max {} and mean {}'.format(np.min(image), np.max(image), np.mean(image)))
    p = 0.5
    noisy = np.random.binomial(1, p,image.shape)
    noisy = image*noisy
    print('min {} , max {} and mean {}'.format(np.min(noisy), np.max(noisy), np.mean(noisy)))
    return noisy

# this is the size of our encoded representations
encoding_dim = 32  # 32 floats -> compression of factor 24.5, assuming the input is 784 floats


input_img = Input(shape=(784,))
input_img_noise = Input(shape=(784,))

# noisy version of inputs
x_train_noise = noisy(x_train)
x_test_noise = noisy(x_test)

x_train = x_train.reshape((len(x_train), np.prod(x_train.shape[1:])))
x_test = x_test.reshape((len(x_test), np.prod(x_test.shape[1:])))
x_train_noise = x_train_noise.reshape((len(x_train_noise), np.prod(x_train_noise.shape[1:])))
x_test_noise = x_test_noise.reshape((len(x_test_noise), np.prod(x_test_noise.shape[1:])))

# "encoded" is the encoded representation of the input
encoded = Dense(encoding_dim, activation='relu')(input_img_noise)
# "decoded" is the lossy reconstruction of the input
decoded = Dense(784, activation='sigmoid')(encoded)

# this model maps an input to its reconstruction
autoencoder = Model(input_img_noise, decoded)

# this model maps an input to its encoded representation
encoder = Model(input_img_noise, encoded)

# create a placeholder for an encoded (32-dimensional) input
encoded_input = Input(shape=(encoding_dim,))
# retrieve the last layer of the autoencoder model
decoder_layer = autoencoder.layers[-1]
# create the decoder model
decoder = Model(encoded_input, decoder_layer(encoded_input))

autoencoder.compile(optimizer='adadelta', loss='binary_crossentropy')
#训练参数
autoencoder.fit(x_train_noise, x_train,
                epochs=100,
                batch_size=256,
                shuffle=True,
                validation_data=(x_test_noise, x_test))

# encode and decode some digits
# note that we take them from the *test* set
encoded_imgs = encoder.predict(x_test_noise)
decoded_imgs = decoder.predict(encoded_imgs)

# use Matplotlib (don't ask)
import matplotlib.pyplot as plt

n = 10  # how many digits we will display
plt.figure(figsize=(20, 4))
for i in range(n):
    # display original
    ax = plt.subplot(2, n, i + 1)
    plt.imshow(x_test_noise[i].reshape(28, 28))
    plt.gray()
    ax.get_xaxis().set_visible(False)
    ax.get_yaxis().set_visible(False)

    # display reconstruction
    ax = plt.subplot(2, n, i + 1 + n)
    plt.imshow(decoded_imgs[i].reshape(28, 28))
    plt.gray()
    ax.get_xaxis().set_visible(False)
    ax.get_yaxis().set_visible(False)
plt.savefig('test.png')

epoch=1 result