python - ValueError : Error when checking input: expected conv2d_3_input to have shape (100, 100, 1) 但得到形状为 (100, 100, 3) 的数组

Question

当我尝试编写用于图像识别的神经网络时，出现错误:

ValueError: Error when checking input: expected conv2d_3_input to have shape (100, 100, 1) but got array with shape (100, 100, 3).

我的所有图像都是灰度，尺寸为100x100像素。代码如下:

# Importing the Keras libraries and packages
import tensorflow as tf
from keras.models import Sequential
from keras.layers import Conv2D
from keras.layers import MaxPooling2D
from keras.layers import Flatten
from keras.layers import Dense
# Initialising the CNN
classifier = Sequential()
# Step 1 - Convolution
classifier.add(Conv2D(32, (3, 3), input_shape = (100, 100, 1), activation = 'relu'))
# Step 2 - Pooling
classifier.add(MaxPooling2D(pool_size = (2, 2)))
# Adding a second convolutional layer
classifier.add(Conv2D(32, (3, 3), activation = 'relu'))
classifier.add(MaxPooling2D(pool_size = (2, 2)))
# Step 3 - Flattening
classifier.add(Flatten())
# Step 4 - Full connection
classifier.add(Dense(units = 128, activation = 'relu'))
classifier.add(Dense(units = 1, activation = 'sigmoid'))
# Compiling the CNN
classifier.compile(optimizer = 'adam', loss = 'binary_crossentropy', metrics = ['accuracy'])
# Part 2 - Fitting the CNN to the images
from keras.preprocessing.image import ImageDataGenerator
train_datagen = ImageDataGenerator(rescale = 1./255,
shear_range = 0.2,
zoom_range = 0.2,
horizontal_flip = True)
test_datagen = ImageDataGenerator(rescale = 1./255)
training_set = train_datagen.flow_from_directory('E:/exercise/dataset/train',
target_size = (100, 100),
batch_size = 32,
color_mode = "grayscale",
class_mode = 'binary')
test_set = test_datagen.flow_from_directory('E:/exercise/dataset/test',
target_size = (100, 100),
color_mode = "grayscale",
batch_size = 32,
class_mode = 'binary')
classifier.fit_generator(training_set,
steps_per_epoch = 40,
epochs = 10,                                                                                   
validation_data = test_set,
validation_steps = 8)

import numpy as np
from keras.preprocessing import image                           
test_image = image.load_img('E:/exercise/predict_2.jpg', target_size = (100, 100))
test_image = image.img_to_array(test_image)
test_image = np.expand_dims(test_image, axis = 0)
result = classifier.predict(test_image)
training_set.class_indices                                   
if result[0][0] >= 0.5: prediction = 'happy'
else: prediction = 'sad'
print(prediction)  

任何人都可以告诉我如何解决这个问题。谢谢大家!

Answer 1

尝试将参数 preprocessing_function=gray_to_rgb 添加到 train_datagen 和 test_datagen 的 ImageDataGenerator 中，如下所示:

def rgb_to_gray(rgb):     # Using the luminosity formula: grayscale =  0.21 R + 0.72 G + 0.07 B
    return np.dot(rgb[...,:3], [0.299, 0.587, 0.114])

train_datagen = ImageDataGenerator(rescale = 1./255,
shear_range = 0.2,
zoom_range = 0.2,
horizontal_flip = True,
preprocessing_function=rgb_to_gray)
test_datagen = ImageDataGenerator(rescale = 1./255, preprocessing_function=rgb_to_gray)