基于卷积神经网络的人脸情绪识别

import os import tensorflow as tf import matplotlib.pyplot as plt import matplotlib.image as mpimg from keras_preprocessing.image import ImageDataGenerator import numpy as np from keras.preprocessing import image import pandas as pd angry_dir = os.path.join('./train/angry/') disgusted_dir = os.path.join('./train/disgusted/') fearful_dir = os.path.join('./train/fearful/') happy_dir = os.path.join('./train/happy/') neutral_dir = os.path.join('./train/neutral/') sad_dir = os.path.join('./train/sad/') surprised_dir = os.path.join('./train/surprised/') print('total training angry images:', len(os.listdir(angry_dir))) print('total training disgusted images:', len(os.listdir(disgusted_dir))) print('total training fearful images:', len(os.listdir(fearful_dir))) print('total training happy images:', len(os.listdir(happy_dir))) print('total training neutral images:', len(os.listdir(neutral_dir))) print('total training sad images:', len(os.listdir(sad_dir))) print('total training surprised images:', len(os.listdir(surprised_dir))) angry_files = os.listdir(angry_dir) print(angry_files[:10]) disgusted_files = os.listdir(disgusted_dir) print(disgusted_files[:10]) fearful_files = os.listdir(fearful_dir) print(fearful_files[:10]) happy_files = os.listdir(happy_dir) print(happy_files[:10]) neutral_files = os.listdir(neutral_dir) print(neutral_files[:10]) sad_files = os.listdir(sad_dir) print(sad_files[:10]) surprised_files = os.listdir(surprised_dir) print(surprised_files[:10]) pic_index = 2 next_angry = [os.path.join(angry_dir,fname) for fname in angry_files[pic_index-2 : pic_index]] for i, img_path in enumerate(next_angry): img = mpimg.imread(img_path) plt.imshow(img) plt.axis('Off') plt.show() gpu_device_name = tf.test.gpu_device_name() print(gpu_device_name) TRAINING_DIR = './train/' training_datagen = ImageDataGenerator( rescale=1. / 255, # 值将在执行其他处理前乘到整个图像上 rotation_range=40, # 整数，数据提升时图片随机转动的角度 width_shift_range=0.2, # 浮点数，图片宽度的某个比例，数据提升时图片随机水平偏移的幅度 height_shift_range=0.2, # 浮点数，图片高度的某个比例，数据提升时图片随机竖直偏移的幅度 shear_range=0.2, # 浮点数，剪切强度（逆时针方向的剪切变换角度）。是用来进行剪切变换的程度 zoom_range=0.2, # 用来进行随机的放大 validation_split=0.25, horizontal_flip=True, # 布尔值，进行随机水平翻转。随机的对图片进行水平翻转，这个参数适用于水平翻转不影响图片语义的时候 fill_mode='nearest' # 'constant','nearest','reflect','wrap'之一，当进行变换时超出边界的点将根据本参数给定的方法进行处理 ) train_generator = training_datagen.flow_from_directory( TRAINING_DIR, subset='training', target_size=(48, 48), class_mode='categorical' ) validation_generator = training_datagen.flow_from_directory( TRAINING_DIR, subset='validation', target_size=(48, 48), class_mode='categorical' ) print(train_generator.class_indices) # os.environ["KMP_DUPLICATE_LIB_OK"] = "TRUE" model = tf.keras.models.Sequential([ tf.keras.layers.Conv2D(16, (5, 5), activation='relu', input_shape=(48, 48, 3), padding='same'), tf.keras.layers.MaxPooling2D(2, 2), tf.keras.layers.Conv2D(32, (5, 5), activation='relu', padding='same'), tf.keras.layers.MaxPooling2D(2, 2), tf.keras.layers.Conv2D(32, (5, 5), activation='relu', padding='same'), tf.keras.layers.MaxPooling2D(2, 2), tf.keras.layers.Flatten(), tf.keras.layers.Dense(128, activation='relu'), tf.keras.layers.Dropout(0.5), tf.keras.layers.Dense(7, activation='softmax') ]) model.summary() model.compile(loss='categorical_crossentropy', optimizer='Adam', metrics=['accuracy']) history = model.fit_generator(train_generator, epochs=100, validation_data=validation_generator, verbose=1) model.save('face(4).h5') acc = history.history['accuracy'] val_acc = history.history['val_accuracy'] loss = history.history['loss'] val_loss = history.history['val_loss'] epochs = range(len(acc)) plt.plot(epochs, acc, 'r', label='Training accuracy') plt.plot(epochs, val_acc, 'b', label='validation accuracy') plt.title('Training and validation accuracy') plt.legend(loc=0) plt.show() gpus = tf.config.experimental.list_physical_devices('GPU') tf.config.experimental.set_memory_growth(gpus[0], True) from keras.models import load_model model = load_model("face(2).h5") dataframe = pd.DataFrame(columns=['name','label']) os.chdir('./test/') # 原 D:\\python\\15、情绪识别 # test_num = 1 for file_name in os.listdir(): img = image.load_img(file_name,target_size=(48,48)) x = image.img_to_array(img) x = np.expand_dims(x, axis=0) images = np.vstack([x]) classes = model.predict(images, batch_size=10) num = np.argmax(classes) if num == 0: face = 'angry' print(file_name+ ' ' + face) # {'angry': 0, 'disgusted': 1, 'fearful': 2, 'happy': 3, 'neutral': 4, 'sad': 5, 'surprised': 6} if num == 1: face = 'disgusted' print(file_name+ ' ' + face) if num == 2: face = 'fearful' print(file_name+ ' ' + face) if num == 3: face = 'happy' print(file_name+ ' ' + face) if num == 4: face = 'neutral' print(file_name+ ' ' + face) if num == 5: face = 'sad' print(file_name+ ' ' + face) if num == 6: face = 'surprised' print(file_name+ ' ' + face) dataframe = dataframe.append({'name':file_name,'label':face},ignore_index=True) # test_num = test_num + 1 # if test_num > 10: # break dataframe.to_csv('submit.csv')