用python和opencv写一个基于卷积神经网络的疲劳驾驶识别检测+数据集

下面是一个基于卷积神经网络的疲劳驾驶识别检测的Python代码，包括数据集的生成和模型的训练和测试：

import cv2
import os
import numpy as np
from sklearn.model_selection import train_test_split
from keras.models import Sequential
from keras.layers import Dense, Dropout, Flatten
from keras.layers.convolutional import Conv2D, MaxPooling2D
from keras.utils import to_categorical

# 生成数据集
def load_dataset(dataset_path, img_size):
    # 读取数据集
    images = []
    labels = []
    for foldername in os.listdir(dataset_path):
        label = int(foldername)
        folderpath = os.path.join(dataset_path, foldername)
        for filename in os.listdir(folderpath):
            imagepath = os.path.join(folderpath, filename)
            image = cv2.imread(imagepath)
            image = cv2.resize(image, (img_size, img_size))
            images.append(image)
            labels.append(label)
    images = np.array(images, dtype=np.float32) / 255.0
    labels = np.array(labels, dtype=np.int32)
    return images, labels

# 划分数据集
def split_dataset(images, labels, test_size):
    x_train, x_test, y_train, y_test = train_test_split(images, labels, test_size=test_size, random_state=42)
    y_train = to_categorical(y_train)
    y_test = to_categorical(y_test)
    return x_train, x_test, y_train, y_test

# 定义模型
def create_model(img_size, num_classes):
    model = Sequential()
    model.add(Conv2D(32, (3, 3), input_shape=(img_size, img_size, 3), activation='relu'))
    model.add(MaxPooling2D(pool_size=(2, 2)))
    model.add(Dropout(0.25))
    model.add(Conv2D(64, (3, 3), activation='relu'))
    model.add(MaxPooling2D(pool_size=(2, 2)))
    model.add(Dropout(0.25))
    model.add(Flatten())
    model.add(Dense(128, activation='relu'))
    model.add(Dropout(0.5))
    model.add(Dense(num_classes, activation='softmax'))
    model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
    return model

# 训练模型
def train_model(x_train, y_train, x_test, y_test, batch_size, epochs):
    model = create_model(x_train.shape[1], y_train.shape[1])
    model.fit(x_train, y_train, validation_data=(x_test, y_test), epochs=epochs, batch_size=batch_size)
    return model

# 测试模型
def test_model(model, images, labels):
    loss, acc = model.evaluate(images, labels, verbose=0)
    return acc

# 主函数
if __name__ == '__main__':
    dataset_path = 'dataset'
    img_size = 32
    test_size = 0.2
    batch_size = 32
    epochs = 10
    num_classes = 2

    images, labels = load_dataset(dataset_path, img_size)
    x_train, x_test, y_train, y_test = split_dataset(images, labels, test_size)
    model = train_model(x_train, y_train, x_test, y_test, batch_size, epochs)
    acc = test_model(model, x_test, y_test)
    print('Test accuracy:', acc)

在这个代码中，我们首先使用`load_dataset`函数读取数据集，然后使用`split_dataset`函数划分数据集，接着使用`create_model`函数定义模型，使用`train_model`函数训练模型，最后使用`test_model`函数测试模型。其中，`num_classes`表示数据集中的类别数，本例中为2（疲劳和非疲劳），可以根据实际情况进行修改。

需要注意的是，在运行代码之前，需要先准备好数据集，将疲劳和非疲劳的图像分别放在两个不同的文件夹中，并将这两个文件夹放在一个名为`dataset`的文件夹下。

希望对你有所帮助！