请利用Python编写一个可以实现 图像识别功能的app

在Python中实现图像识别通常涉及到机器学习和深度学习技术，尤其是卷积神经网络（CNN）。以下是一个简单示例，展示如何使用预训练模型如TensorFlow的`tf.keras`库进行图像分类。假设我们想识别猫狗图片，可以采用经典的Keras `MobileNetV2` 模型。

# 首先，确保已安装必要的库
!pip install tensorflow numpy pillow

import tensorflow as tf
from tensorflow.keras.applications.mobilenet_v2 import MobileNetV2, preprocess_input
from tensorflow.keras.preprocessing.image import ImageDataGenerator
from tensorflow.keras.models import Model
from tensorflow.keras.layers import Dense, GlobalAveragePooling2D
import numpy as np
import os

# 1. 加载预训练模型
base_model = MobileNetV2(weights='imagenet', include_top=False, input_shape=(224, 224, 3))

# 2. 添加新的全连接层用于特定任务
x = base_model.output
x = GlobalAveragePooling2D()(x)
predictions = Dense(2, activation='softmax')(x) # 这里2代表猫和狗两个类别

# 创建新模型
model = Model(inputs=base_model.input, outputs=predictions)

# 冻结预训练模型的权重
for layer in base_model.layers:
    layer.trainable = False

# 编译模型
model.compile(optimizer=tf.keras.optimizers.Adam(), loss='categorical_crossentropy', metrics=['accuracy'])

# 准备数据集
train_dir = 'path_to_train_images'
validation_dir = 'path_to_validation_images'

train_datagen = ImageDataGenerator(preprocessing_function=preprocess_input)
val_datagen = ImageDataGenerator(preprocess_function=preprocess_input)

train_generator = train_datagen.flow_from_directory(
    train_dir,
    target_size=(224, 224),
    batch_size=32,
    class_mode='categorical')

validation_generator = val_datagen.flow_from_directory(
    validation_dir,
    target_size=(224, 224),
    batch_size=32,
    class_mode='categorical')

# 训练模型
history = model.fit(
    train_generator,
    epochs=10,  # 调整epochs根据实际情况
    validation_data=validation_generator)

# 3. 实际应用：预测新图片
def predict_image(image_path):
    img = tf.keras.preprocessing.image.load_img(image_path, target_size=(224, 224))
    x = tf.keras.preprocessing.image.img_to_array(img)
    x = np.expand_dims(x, axis=0)
    x = preprocess_input(x)
    predictions = model.predict(x)
    predicted_class = np.argmax(predictions)
    print(f"Predicted class: {predicted_class}")

# 使用predict_image函数识别新图片
predict_image('path_to_test_image.jpg')