We need the following libraries for Denoising of image:
from keras.datasets import mnist
import numpy as np
import matplotlib.pyplot as plt
from keras.layers import Dense, Conv2D, MaxPooling2D, UpSampling2D
from keras.models import Sequential
from keras import backend as K
First we have to load the data (here we load data from MNIST Database)
(x_train, _), (x_test, _) = mnist.load_data()
x_train = x_train.astype('float32') / 255.
x_test = x_test.astype('float32') / 255.
x_train = np.reshape(x_train, (len(x_train), 28, 28, 1))
x_test = np.reshape(x_test, (len(x_test), 28, 28, 1))
#For denoising of the image we will use Autoencoder (, Autoencoders is a Reinforcement learning.) & Bulid the CNN Model.
#code
input_img = (28, 28, 1)
model = Sequential()
model.add(Conv2D(16, (3, 3), activation='relu', padding='same', input_shape = input_img))
model.add(MaxPooling2D((2, 2), padding='same'))
model.add(Conv2D(8, (3, 3), activation='relu', padding='same'))
model.add(MaxPooling2D((2, 2), padding='same'))
model.add(Conv2D(8, (3, 3), activation='relu', padding='same'))
model.add(MaxPooling2D((2, 2), padding='same'))
# at this point the representation is (4, 4, 8) i.e. 128-dimensional
model.add(Conv2D(8, (3, 3), activation='relu', padding='same'))
model.add(UpSampling2D((2, 2)))
model.add(Conv2D(8, (3, 3), activation='relu', padding='same'))
model.add(UpSampling2D((2, 2)))
model.add(Conv2D(16, (3, 3), activation='relu'))
model.add(UpSampling2D((2, 2)))
model.add(Conv2D(1, (3, 3), activation='sigmoid', padding='same'))
model.compile(optimizer='adam',
loss='binary_crossentropy',
metrics=['accuracy'])
model.summary()
# we will show the machine how the data look like(fit the model ,“hist-is a variable”)
hist = model.fit(x_train, x_train,
epochs=50,
batch_size=128,
shuffle=True,
validation_data=(x_test, x_test),
)
# we will check the prediction that about the data.
decoded_imgs = model.predict(x_test)
n = 10
plt.figure(figsize=(20, 4))
for i in range(1, 10):
# display original
ax = plt.subplot(2, n, i)
plt.imshow(x_test[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 + n)
plt.imshow(decoded_imgs[i].reshape(28, 28))
plt.gray()
ax.get_xaxis().set_visible(False)
ax.get_yaxis().set_visible(False)
plt.show()
#Deniosing
#we will add now the noise in the image as we have take the image without noise.
noise_factor = 0.5
x_train_noisy = x_train + noise_factor * np.random.normal(loc=0.0, scale=1.0, size=x_train.shape)
x_test_noisy = x_test + noise_factor * np.random.normal(loc=0.0, scale=1.0, size=x_test.shape)
x_train_noisy = np.clip(x_train_noisy, 0., 1.)
x_test_noisy = np.clip(x_test_noisy, 0., 1.)
n = 10
plt.figure(figsize=(20, 2))
for i in range(1, 10):
ax = plt.subplot(1, n, i)
plt.imshow(x_test_noisy[i].reshape(28, 28))
plt.gray()
ax.get_xaxis().set_visible(False)
ax.get_yaxis().set_visible(False)
plt.show()
#Again build the CNN model
input_img = (28, 28, 1)
model2 = Sequential()
model2.add(Conv2D(16, (3, 3), activation='relu', padding='same', input_shape = input_img))
model2.add(MaxPooling2D((2, 2), padding='same'))
model2.add(Conv2D(8, (3, 3), activation='relu', padding='same'))
model2.add(MaxPooling2D((2, 2), padding='same'))
model2.add(Conv2D(8, (3, 3), activation='relu', padding='same'))
model2.add(MaxPooling2D((2, 2), padding='same'))
# at this point the representation is (4, 4, 8) i.e. 128-dimensional
model2.add(Conv2D(8, (3, 3), activation='relu', padding='same'))
model2.add(UpSampling2D((2, 2)))
model2.add(Conv2D(8, (3, 3), activation='relu', padding='same'))
model2.add(UpSampling2D((2, 2)))
model2.add(Conv2D(16, (3, 3), activation='relu'))
model2.add(UpSampling2D((2, 2)))
model2.add(Conv2D(1, (3, 3), activation='sigmoid', padding='same'))
model2.compile(optimizer='adam',
loss='binary_crossentropy',
metrics=['accuracy'])
model2.summary()
#fit the model
hist2 = model2.fit(x_train_noisy, x_train,
epochs=50,
batch_size=128,
shuffle=True,
validation_data=(x_test_noisy, x_test),
)
#final denoising will be done
decoded_imgs = model2.predict(x_test_noisy)
n = 10
plt.figure(figsize=(20, 4))
for i in range(1, 10):
# display original
ax = plt.subplot(2, n, i)
plt.imshow(x_test_noisy[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 + n)
plt.imshow(decoded_imgs[i].reshape(28, 28))
plt.gray()
ax.get_xaxis().set_visible(False)
ax.get_yaxis().set_visible(False)
plt.show()
#We need the following libraries for Denoising of image:
from keras.datasets import mnist
import numpy as np
import matplotlib.pyplot as plt
from keras.layers import Dense, Conv2D, MaxPooling2D, UpSampling2D
from keras.models import Sequential
from keras import backend as K
#First we have to load the data (here we load data from MNIST Database)
(x_train, _), (x_test, _) = mnist.load_data()
x_train = x_train.astype('float32') / 255.
x_test = x_test.astype('float32') / 255.
x_train = np.reshape(x_train, (len(x_train), 28, 28, 1))
x_test = np.reshape(x_test, (len(x_test), 28, 28, 1))
#For denoising of the image we will use Autoencoder (, Autoencoders is a Reinforcement learning.) & Bulid the CNN Model.
#code
input_img = (28, 28, 1)
model = Sequential()
model.add(Conv2D(16, (3, 3), activation='relu', padding='same', input_shape = input_img))
model.add(MaxPooling2D((2, 2), padding='same'))
model.add(Conv2D(8, (3, 3), activation='relu', padding='same'))
model.add(MaxPooling2D((2, 2), padding='same'))
model.add(Conv2D(8, (3, 3), activation='relu', padding='same'))
model.add(MaxPooling2D((2, 2), padding='same'))
# at this point the representation is (4, 4, 8) i.e. 128-dimensional
model.add(Conv2D(8, (3, 3), activation='relu', padding='same'))
model.add(UpSampling2D((2, 2)))
model.add(Conv2D(8, (3, 3), activation='relu', padding='same'))
model.add(UpSampling2D((2, 2)))
model.add(Conv2D(16, (3, 3), activation='relu'))
model.add(UpSampling2D((2, 2)))
model.add(Conv2D(1, (3, 3), activation='sigmoid', padding='same'))
model.compile(optimizer='adam',
loss='binary_crossentropy',
metrics=['accuracy'])
model.summary()
# we will show the machine how the data look like(fit the model ,“hist-is a variable”)
hist = model.fit(x_train, x_train,
epochs=50,
batch_size=128,
shuffle=True,
validation_data=(x_test, x_test),
)
# we will check the prediction that about the data.
decoded_imgs = model.predict(x_test)
n = 10
plt.figure(figsize=(20, 4))
for i in range(1, 10):
# display original
ax = plt.subplot(2, n, i)
plt.imshow(x_test[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 + n)
plt.imshow(decoded_imgs[i].reshape(28, 28))
plt.gray()
ax.get_xaxis().set_visible(False)
ax.get_yaxis().set_visible(False)
plt.show()
#Deniosing
#we will add now the noise in the image as we have take the image without noise.
noise_factor = 0.5
x_train_noisy = x_train + noise_factor * np.random.normal(loc=0.0, scale=1.0, size=x_train.shape)
x_test_noisy = x_test + noise_factor * np.random.normal(loc=0.0, scale=1.0, size=x_test.shape)
x_train_noisy = np.clip(x_train_noisy, 0., 1.)
x_test_noisy = np.clip(x_test_noisy, 0., 1.)
n = 10
plt.figure(figsize=(20, 2))
for i in range(1, 10):
ax = plt.subplot(1, n, i)
plt.imshow(x_test_noisy[i].reshape(28, 28))
plt.gray()
ax.get_xaxis().set_visible(False)
ax.get_yaxis().set_visible(False)
plt.show()
#Again build the CNN model
input_img = (28, 28, 1)
model2 = Sequential()
model2.add(Conv2D(16, (3, 3), activation='relu', padding='same', input_shape = input_img))
model2.add(MaxPooling2D((2, 2), padding='same'))
model2.add(Conv2D(8, (3, 3), activation='relu', padding='same'))
model2.add(MaxPooling2D((2, 2), padding='same'))
model2.add(Conv2D(8, (3, 3), activation='relu', padding='same'))
model2.add(MaxPooling2D((2, 2), padding='same'))
# at this point the representation is (4, 4, 8) i.e. 128-dimensional
model2.add(Conv2D(8, (3, 3), activation='relu', padding='same'))
model2.add(UpSampling2D((2, 2)))
model2.add(Conv2D(8, (3, 3), activation='relu', padding='same'))
model2.add(UpSampling2D((2, 2)))
model2.add(Conv2D(16, (3, 3), activation='relu'))
model2.add(UpSampling2D((2, 2)))
model2.add(Conv2D(1, (3, 3), activation='sigmoid', padding='same'))
model2.compile(optimizer='adam',
loss='binary_crossentropy',
metrics=['accuracy'])
model2.summary()
#fit the model
hist2 = model2.fit(x_train_noisy, x_train,
epochs=50,
batch_size=128,
shuffle=True,
validation_data=(x_test_noisy, x_test),
)
#final denoising will be done
decoded_imgs = model2.predict(x_test_noisy)
n = 10
plt.figure(figsize=(20, 4))
for i in range(1, 10):
# display original
ax = plt.subplot(2, n, i)
plt.imshow(x_test_noisy[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 + n)
plt.imshow(decoded_imgs[i].reshape(28, 28))
plt.gray()
ax.get_xaxis().set_visible(False)
ax.get_yaxis().set_visible(False)
plt.show()
#We need the following libraries for Denoising of image:
from keras.datasets import mnist
import numpy as np
import matplotlib.pyplot as plt
from keras.layers import Dense, Conv2D, MaxPooling2D, UpSampling2D
from keras.models import Sequential
from keras import backend as K
#First we have to load the data (here we load data from MNIST Database)
(x_train, _), (x_test, _) = mnist.load_data()
x_train = x_train.astype('float32') / 255.
x_test = x_test.astype('float32') / 255.
x_train = np.reshape(x_train, (len(x_train), 28, 28, 1))
x_test = np.reshape(x_test, (len(x_test), 28, 28, 1))
#For denoising of the image we will use Autoencoder (, Autoencoders is a Reinforcement learning.) & Build the CNN Model.
#code
input_img = (28, 28, 1)
model = Sequential()
model.add(Conv2D(16, (3, 3), activation='relu', padding='same', input_shape = input_img))
model.add(MaxPooling2D((2, 2), padding='same'))
model.add(Conv2D(8, (3, 3), activation='relu', padding='same'))
model.add(MaxPooling2D((2, 2), padding='same'))
model.add(Conv2D(8, (3, 3), activation='relu', padding='same'))
model.add(MaxPooling2D((2, 2), padding='same'))
# at this point the representation is (4, 4, 8) i.e. 128-dimensional
model.add(Conv2D(8, (3, 3), activation='relu', padding='same'))
model.add(UpSampling2D((2, 2)))
model.add(Conv2D(8, (3, 3), activation='relu', padding='same'))
model.add(UpSampling2D((2, 2)))
model.add(Conv2D(16, (3, 3), activation='relu'))
model.add(UpSampling2D((2, 2)))
model.add(Conv2D(1, (3, 3), activation='sigmoid', padding='same'))
model.compile(optimizer='adam',
loss='binary_crossentropy',
metrics=['accuracy'])
model.summary()
# we will show the machine how the data look like(fit the model)
hist = model.fit(x_train, x_train,
epochs=50,
batch_size=128,
shuffle=True,
validation_data=(x_test, x_test),
)
# we will check the prediction that about the data.
decoded_imgs = model.predict(x_test)
n = 10
plt.figure(figsize=(20, 4))
for i in range(1, 10):
# display original
ax = plt.subplot(2, n, i)
plt.imshow(x_test[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 + n)
plt.imshow(decoded_imgs[i].reshape(28, 28))
plt.gray()
ax.get_xaxis().set_visible(False)
ax.get_yaxis().set_visible(False)
plt.show()
#Deniosing
#we will add now the noise in the image as we have take the image without noise.
noise_factor = 0.5
x_train_noisy = x_train + noise_factor * np.random.normal(loc=0.0, scale=1.0, size=x_train.shape)
x_test_noisy = x_test + noise_factor * np.random.normal(loc=0.0, scale=1.0, size=x_test.shape)
x_train_noisy = np.clip(x_train_noisy, 0., 1.)
x_test_noisy = np.clip(x_test_noisy, 0., 1.)
n = 10
plt.figure(figsize=(20, 2))
for i in range(1, 10):
ax = plt.subplot(1, n, i)
plt.imshow(x_test_noisy[i].reshape(28, 28))
plt.gray()
ax.get_xaxis().set_visible(False)
ax.get_yaxis().set_visible(False)
plt.show()
#Again build the CNN model
input_img = (28, 28, 1)
model2 = Sequential()
model2.add(Conv2D(16, (3, 3), activation='relu', padding='same', input_shape = input_img))
model2.add(MaxPooling2D((2, 2), padding='same'))
model2.add(Conv2D(8, (3, 3), activation='relu', padding='same'))
model2.add(MaxPooling2D((2, 2), padding='same'))
model2.add(Conv2D(8, (3, 3), activation='relu', padding='same'))
model2.add(MaxPooling2D((2, 2), padding='same'))
# at this point the representation is (4, 4, 8) i.e. 128-dimensional
model2.add(Conv2D(8, (3, 3), activation='relu', padding='same'))
model2.add(UpSampling2D((2, 2)))
model2.add(Conv2D(8, (3, 3), activation='relu', padding='same'))
model2.add(UpSampling2D((2, 2)))
model2.add(Conv2D(16, (3, 3), activation='relu'))
model2.add(UpSampling2D((2, 2)))
model2.add(Conv2D(1, (3, 3), activation='sigmoid', padding='same'))
model2.compile(optimizer='adam',
loss='binary_crossentropy',
metrics=['accuracy'])
model2.summary()
#fit the model
hist2 = model2.fit(x_train_noisy, x_train,
epochs=50,
batch_size=128,
shuffle=True,
validation_data=(x_test_noisy, x_test),
)
#final denoising will be done
decoded_imgs = model2.predict(x_test_noisy)
n = 10
plt.figure(figsize=(20, 4))
for i in range(1, 10):
# display original
ax = plt.subplot(2, n, i)
plt.imshow(x_test_noisy[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 + n)
plt.imshow(decoded_imgs[i].reshape(28, 28))
plt.gray()
ax.get_xaxis().set_visible(False)
ax.get_yaxis().set_visible(False)
plt.show()
Removing unwanted noising in order to restore the original image.
minify code