5. Removing noise from the images

 AIM: Implement Multi-Layer Perceptron algorithm for Image denoising hyperparameter tuning.


PROGRAM:


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

# Load the MNIST dataset
(x_train, _), (x_test, _) = mnist.load_data()

# Normalize and reshape the data
x_train = x_train.astype('float32') / 255.0
x_test = x_test.astype('float32') / 255.0
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:])))

# Add noise to the training and test data
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)

# Clip the noisy images to ensure values are between 0 and 1
x_train_noisy = np.clip(x_train_noisy, 0., 1.)
x_test_noisy = np.clip(x_test_noisy, 0., 1.)

# Define the MLP denoising model
def create_denoising_mlp(input_shape, encoding_dim):
    input_img = Input(shape=input_shape)
    encoded = Dense(encoding_dim, activation='relu')(input_img)
    decoded = Dense(input_shape[0], activation='sigmoid')(encoded)

    autoencoder = Model(input_img, decoded)
    return autoencoder

# Define input shape and encoding dimension
input_shape = (784,)  # 28x28 flattened images
encoding_dim = 128

# Create the MLP denoising model
model = create_denoising_mlp(input_shape, encoding_dim)

# Compile the model
model.compile(optimizer='adam', loss='binary_crossentropy')

# Train the denoising model
batch_size = 32
epochs = 50
model.fit(x_train_noisy, x_train, epochs=epochs, batch_size=batch_size, shuffle=True, validation_data=(x_test_noisy, x_test))

# Denoise test images
decoded_imgs = model.predict(x_test_noisy)

# Display the noisy and denoised images
import matplotlib.pyplot as plt

n = 10  # Number of images to display
plt.figure(figsize=(20, 4))
for i in range(n):
    # Display original images
    ax = plt.subplot(2, n, i + 1)
    plt.imshow(x_test_noisy[i].reshape(28, 28), cmap='gray')
    plt.title('Noisy')
    ax.get_xaxis().set_visible(False)
    ax.get_yaxis().set_visible(False)

    # Display denoised images
    ax = plt.subplot(2, n, i + 1 + n)
    plt.imshow(decoded_imgs[i].reshape(28, 28), cmap='gray')
    plt.title('Denoised')
    ax.get_xaxis().set_visible(False)
    ax.get_yaxis().set_visible(False)
plt.show()


o/P:

Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz 11490434/11490434 [==============================] - 0s 0us/step Epoch 1/50 1875/1875 [==============================] - 18s 9ms/step - loss: 0.1531 - val_loss: 0.1203 Epoch 2/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1173 - val_loss: 0.1140 Epoch 3/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1132 - val_loss: 0.1126 Epoch 4/50 1875/1875 [==============================] - 14s 7ms/step - loss: 0.1112 - val_loss: 0.1103 Epoch 5/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1099 - val_loss: 0.1094 Epoch 6/50 1875/1875 [==============================] - 14s 7ms/step - loss: 0.1088 - val_loss: 0.1095 Epoch 7/50 1875/1875 [==============================] - 14s 7ms/step - loss: 0.1081 - val_loss: 0.1082 Epoch 8/50 1875/1875 [==============================] - 14s 7ms/step - loss: 0.1075 - val_loss: 0.1077 Epoch 9/50 1875/1875 [==============================] - 14s 7ms/step - loss: 0.1070 - val_loss: 0.1078 Epoch 10/50 1875/1875 [==============================] - 15s 8ms/step - loss: 0.1067 - val_loss: 0.1078 Epoch 11/50 1875/1875 [==============================] - 14s 7ms/step - loss: 0.1064 - val_loss: 0.1076 Epoch 12/50 1875/1875 [==============================] - 14s 7ms/step - loss: 0.1062 - val_loss: 0.1071 Epoch 13/50 1875/1875 [==============================] - 14s 7ms/step - loss: 0.1060 - val_loss: 0.1069 Epoch 14/50 1875/1875 [==============================] - 14s 7ms/step - loss: 0.1059 - val_loss: 0.1071 Epoch 15/50 1875/1875 [==============================] - 14s 7ms/step - loss: 0.1057 - val_loss: 0.1069 Epoch 16/50 1875/1875 [==============================] - 14s 7ms/step - loss: 0.1056 - val_loss: 0.1066 Epoch 17/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1054 - val_loss: 0.1064 Epoch 18/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1053 - val_loss: 0.1066 Epoch 19/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1052 - val_loss: 0.1066 Epoch 20/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1051 - val_loss: 0.1063 Epoch 21/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1049 - val_loss: 0.1065 Epoch 22/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1049 - val_loss: 0.1063 Epoch 23/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1048 - val_loss: 0.1062 Epoch 24/50 1875/1875 [==============================] - 15s 8ms/step - loss: 0.1047 - val_loss: 0.1063 Epoch 25/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1046 - val_loss: 0.1061 Epoch 26/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1045 - val_loss: 0.1061 Epoch 27/50 1875/1875 [==============================] - 15s 8ms/step - loss: 0.1044 - val_loss: 0.1060 Epoch 28/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1043 - val_loss: 0.1059 Epoch 29/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1042 - val_loss: 0.1059 Epoch 30/50 1875/1875 [==============================] - 14s 7ms/step - loss: 0.1042 - val_loss: 0.1058 Epoch 31/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1041 - val_loss: 0.1058 Epoch 32/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1040 - val_loss: 0.1059 Epoch 33/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1040 - val_loss: 0.1056 Epoch 34/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1039 - val_loss: 0.1059 Epoch 35/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1039 - val_loss: 0.1054 Epoch 36/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1038 - val_loss: 0.1057 Epoch 37/50 1875/1875 [==============================] - 15s 8ms/step - loss: 0.1037 - val_loss: 0.1056 Epoch 38/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1037 - val_loss: 0.1057 Epoch 39/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1036 - val_loss: 0.1055 Epoch 40/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1036 - val_loss: 0.1054 Epoch 41/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1035 - val_loss: 0.1055 Epoch 42/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1035 - val_loss: 0.1053 Epoch 43/50 1875/1875 [==============================] - 14s 7ms/step - loss: 0.1034 - val_loss: 0.1054 Epoch 44/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1034 - val_loss: 0.1055 Epoch 45/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1034 - val_loss: 0.1052 Epoch 46/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1033 - val_loss: 0.1052 Epoch 47/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1033 - val_loss: 0.1050 Epoch 48/50 1875/1875 [==============================] - 14s 7ms/step - loss: 0.1032 - val_loss: 0.1051 Epoch 49/50 1875/1875 [==============================] - 13s 7ms/step - loss: 0.1032 - val_loss: 0.1051 Epoch 50/50 1875/1875 [==============================] - 15s 8ms/step - loss: 0.1032 - val_loss: 0.1053 313/313 [==============================] - 1s 2ms/step

noise images
denoised images

Comments

Post a Comment

Popular posts from this blog

11. List of Capstone Projects for SOC - Deep Learning

 List of Capstone Projects for SOC - Deep Learning  1. Build your Own Neural Net from Scratch. 2. Image Classification with CIFAR-10 Dataset.(Multi Class Classification) CNN 3. Human Face Detection.( detecting the bounding boxes of the human faces from a given Image)  YOLO  4. Traffic Sign Classification.(Collect Various traffic Signal Images and train Neural Network Model) CNN 5. Text Summarizer. RNN 6. Chatbot Using Deep Learning ( Training/Generative) 7. Colouring old B&W photos (Deep Neural Networks OpenCV) 8. Animal detection by training from multiple classes (CNN) 9. Face detection
Read more

8. Advanced CNN (Build AlexNet using Advanced CNN)

  Building and training an AlexNet-like model using the CIFAR-10 dataset in TensorFlow/Keras: Program to build the model: import tensorflow as tf from tensorflow.keras.layers import Conv2D, MaxPooling2D, Flatten, Dense, Dropout from tensorflow.keras.datasets import cifar10 from tensorflow.keras.utils import to_categorical from sklearn.model_selection import train_test_split # Load and preprocess CIFAR-10 dataset (x_train, y_train), (x_test, y_test) = cifar10.load_data() x_train = x_train.astype( 'float32' ) / 255.0 x_test = x_test.astype( 'float32' ) / 255.0 y_train = to_categorical(y_train, num_classes= 10 ) y_test = to_categorical(y_test, num_classes= 10 ) # Split into training and validation sets x_train, x_val, y_train, y_val = train_test_split(x_train, y_train, test_size= 0.1 , random_state= 42 ) # Define the model architecture '''model = tf.keras.Sequential([     Conv2D(96, (11, 11), strides=(4, 4), activation='relu', input_shape=(32...
Read more