基于CNN的猫狗识别应用

数据准备

这边使用的数据来源：kaggle官网。

具体的数据有：训练数据8000张，猫和狗各4000张；验证数据2000张，猫和狗各1000张。

数据放置位置：D：\cat_dog\data\cat and dog

在cat and dog文件夹内分出train 和validation 两个子文件夹。

在train文件夹内需将4000张猫的图像放入cats内，4000张狗的图像放入dogs内。同样的validation内放1000张猫的图像放入cats内，1000张狗的图像放入dogs内。

做好这些数据准备后，开始写代码。

未数据增强的代码

未数据增强的


import os
import warnings

warnings.filterwarnings("ignore")
import tensorflow as tf
from tensorflow.keras.optimizers import Adam
from tensorflow.keras.preprocessing.image import ImageDataGenerator

# 数据所在文件夹
base_dir = 'D:\cat_dog\data\cat and dog'
train_dir = os.path.join(base_dir, 'train')
validation_dir = os.path.join(base_dir, 'validation')

# 训练集
train_cats_dir = os.path.join(train_dir, 'cats')
train_dogs_dir = os.path.join(train_dir, 'dogs')

# 验证集
validation_cats_dir = os.path.join(validation_dir, 'cats')
validation_dogs_dir = os.path.join(validation_dir, 'dogs')
model = tf.keras.models.Sequential([
    tf.keras.layers.Conv2D(32, (3, 3), activation='relu', input_shape=(64, 64, 3)),
    tf.keras.layers.MaxPooling2D(2, 2),

    tf.keras.layers.Conv2D(64, (3, 3), activation='relu'),
    tf.keras.layers.MaxPooling2D(2, 2),

    tf.keras.layers.Conv2D(128, (3, 3), activation='relu'),
    tf.keras.layers.MaxPooling2D(2, 2),

    # 为全连接层准备
    tf.keras.layers.Flatten(),

    tf.keras.layers.Dense(512, activation='relu'),
    # 二分类sigmoid就够了
    tf.keras.layers.Dense(1, activation='sigmoid')
])
model.summary()
model.compile(loss='binary_crossentropy',
              optimizer=Adam(lr=1e-4),
              metrics=['acc'])
# 归一化处理
train_datagen = ImageDataGenerator(rescale=1. / 255)
test_datagen = ImageDataGenerator(rescale=1. / 255)
# 训练数据的生成器
train_generator = train_datagen.flow_from_directory(
    train_dir,  # 文件夹路径
    target_size=(64, 64),  # 统一图片数据大小，要与CNN输入大小一致
    batch_size=10,   # 分批次训练
    # 二分类用binary就可以
    class_mode='binary')
#验证数据的生成器
validation_generator = test_datagen.flow_from_directory(
    validation_dir,
    target_size=(64, 64),
    batch_size=10,
    class_mode='binary')
# fit是数据放入内存中，generator是一个生成器，动态产生数据
history = model.fit_generator(
    train_generator,
    steps_per_epoch=800,  # 8000 images = batch_size * steps
    epochs=50,
    validation_data=validation_generator,
    validation_steps=200,  # 2000 images = batch_size * steps
    verbose=2)
import matplotlib.pyplot as plt

acc = history.history['acc']
val_acc = history.history['val_acc']
loss = history.history['loss']
val_loss = history.history['val_loss']

epochs = range(len(acc))

plt.plot(epochs, acc, 'bo', label='Training accuracy')
plt.plot(epochs, val_acc, 'b', label='Validation accuracy')
plt.title('Training and validation accuracy')

plt.figure()

plt.plot(epochs, loss, 'bo', label='Training Loss')
plt.plot(epochs, val_loss, 'b', label='Validation Loss')
plt.title('Training and validation loss')
plt.legend()

plt.show()

运行之后的结果图为：

运行结果出现了过拟合现象。下面采用数据增强和加入dropout层缓解过拟合现象。

数据增强后的代码

增强后的代码：


import os
import warnings

warnings.filterwarnings("ignore")
import tensorflow as tf
from tensorflow.keras.optimizers import Adam
from tensorflow.keras.preprocessing.image import ImageDataGenerator

# 数据所在文件夹
base_dir = 'D:\cat_dog\data\cat and dog'
train_dir = os.path.join(base_dir, 'train')
validation_dir = os.path.join(base_dir, 'validation')

# 训练集
train_cats_dir = os.path.join(train_dir, 'cats')
train_dogs_dir = os.path.join(train_dir, 'dogs')

# 验证集
validation_cats_dir = os.path.join(validation_dir, 'cats')
validation_dogs_dir = os.path.join(validation_dir, 'dogs')
model = tf.keras.models.Sequential([
    tf.keras.layers.Conv2D(32, (3, 3), activation='relu', input_shape=(64, 64, 3)),
    tf.keras.layers.MaxPooling2D(2, 2),

    tf.keras.layers.Conv2D(64, (3, 3), activation='relu'),
    tf.keras.layers.MaxPooling2D(2, 2),

    tf.keras.layers.Conv2D(128, (3, 3), activation='relu'),
    tf.keras.layers.MaxPooling2D(2, 2),

    # 为全连接层准备
    tf.keras.layers.Flatten(),

    tf.keras.layers.Dense(512, activation='relu'),
    # 二分类sigmoid就够了
    tf.keras.layers.Dense(1, activation='sigmoid')
])
model.summary()
model.compile(loss='binary_crossentropy',
              optimizer=Adam(lr=1e-4),
              metrics=['acc'])
train_datagen = ImageDataGenerator(
    rescale=1. / 255,
    rotation_range=40,
    width_shift_range=0.2,
    height_shift_range=0.2,
    shear_range=0.2,
    zoom_range=0.2,
    horizontal_flip=True,
    fill_mode='nearest')

test_datagen = ImageDataGenerator(rescale=1. / 255)

train_generator = train_datagen.flow_from_directory(
    train_dir,
    target_size=(64, 64),
    batch_size=10,
    class_mode='binary')

validation_generator = test_datagen.flow_from_directory(
    validation_dir,
    target_size=(64, 64),
    batch_size=10,
    class_mode='binary')
history = model.fit_generator(
    train_generator,
    steps_per_epoch=800,  # 8000 images = batch_size * steps
    epochs=80,
    validation_data=validation_generator,
    validation_steps=200,  # 2000 images = batch_size * steps
    verbose=2)
import matplotlib.pyplot as plt

acc = history.history['acc']
val_acc = history.history['val_acc']
loss = history.history['loss']
val_loss = history.history['val_loss']

epochs = range(len(acc))

plt.plot(epochs, acc, 'bo', label='Training accuracy')
plt.plot(epochs, val_acc, 'b', label='Validation accuracy')
plt.title('Training and validation accuracy')

plt.figure()

plt.plot(epochs, loss, 'bo', label='Training Loss')
plt.plot(epochs, val_loss, 'b', label='Validation Loss')
plt.title('Training and validation loss')
plt.legend()

plt.show()

采用数据增强后的结果图：

再加入dropout层：

加入dropout层的代码

加入dropout层
import os
import warnings

warnings.filterwarnings("ignore")
import tensorflow as tf
from tensorflow.keras.optimizers import Adam
from tensorflow.keras.preprocessing.image import ImageDataGenerator

# 数据所在文件夹
base_dir = 'D:\cat_dog\data\cat and dog'
train_dir = os.path.join(base_dir, 'train')
validation_dir = os.path.join(base_dir, 'validation')

# 训练集
train_cats_dir = os.path.join(train_dir, 'cats')
train_dogs_dir = os.path.join(train_dir, 'dogs')

# 验证集
validation_cats_dir = os.path.join(validation_dir, 'cats')
validation_dogs_dir = os.path.join(validation_dir, 'dogs')
model = tf.keras.models.Sequential([
    tf.keras.layers.Conv2D(32, (3, 3), activation='relu', input_shape=(64, 64, 3)),
    tf.keras.layers.MaxPooling2D(2, 2),

    tf.keras.layers.Conv2D(64, (3, 3), activation='relu'),
    tf.keras.layers.MaxPooling2D(2, 2),

    tf.keras.layers.Conv2D(128, (3, 3), activation='relu'),
    tf.keras.layers.MaxPooling2D(2, 2),

    # 为全连接层准备
    tf.keras.layers.Flatten(),

    #退出层
    tf.keras.layers.Dropout(0.5),

    tf.keras.layers.Dense(512, activation='relu'),
    # 二分类sigmoid就够了
    tf.keras.layers.Dense(1, activation='sigmoid')
])
model.summary()
model.compile(loss='binary_crossentropy',
              optimizer=Adam(lr=1e-4),
              metrics=['acc'])
train_datagen = ImageDataGenerator(
    rescale=1. / 255,
    rotation_range=40,
    width_shift_range=0.2,
    height_shift_range=0.2,
    shear_range=0.2,
    zoom_range=0.2,
    horizontal_flip=True,
    fill_mode='nearest')

test_datagen = ImageDataGenerator(rescale=1. / 255)

train_generator = train_datagen.flow_from_directory(
    train_dir,
    target_size=(64, 64),
    batch_size=10,
    class_mode='binary')

validation_generator = test_datagen.flow_from_directory(
    validation_dir,
    target_size=(64, 64),
    batch_size=10,
    class_mode='binary')
history = model.fit_generator(
    train_generator,
    steps_per_epoch=800,  # 8000 images = batch_size * steps
    epochs=150,
    validation_data=validation_generator,
    validation_steps=200,  # 2000 images = batch_size * steps
    verbose=2)
import matplotlib.pyplot as plt

acc = history.history['acc']
val_acc = history.history['val_acc']
loss = history.history['loss']
val_loss = history.history['val_loss']

epochs = range(len(acc))

plt.plot(epochs, acc, 'bo', label='Training accuracy')
plt.plot(epochs, val_acc, 'b', label='Validation accuracy')
plt.title('Training and validation accuracy')

plt.figure()

plt.plot(epochs, loss, 'bo', label='Training Loss')
plt.plot(epochs, val_loss, 'b', label='Validation Loss')
plt.title('Training and validation loss')
plt.legend()

plt.show()

运行结果为：

具体的代码如上（原理不想解释了，啊啊啊

运行的环境需要：GPU、Windows10、Python:3.9.13、Anaconda:1.11.0、Conda:22.9.0、Tensorflow:2.10.0、Torch:2.0.0、Keras:2.4.3）