一、定义模型 lenet.py
from keras.models import Sequential
from keras.layers.convolutional import Conv2D
from keras.layers.convolutional import MaxPooling2D
from keras.layers.core import Activation
from keras.layers.core import Flatten
from keras.layers.core import Dense
from keras import backend as K
class LeNet:
@staticmethod
def build(width, height, depth, classes):
model = Sequential()
inputShape = (height, width, depth)
# using "channels first",（样本数，通道数，行或高，列或宽）通道在前,修改参数位置
if K.image_data_format() == "channels_first":
inputShape = (depth, height, width)
"""
CONV => RELU => POOL layers
CONV => RELU => POOL layers
FC => RELU layers
softmax
"""
model.add(Conv2D(20, (5, 5), padding="same", input_shape=inputShape,name='Conv_1'))
model.add(Activation("relu"))
model.add(MaxPooling2D(pool_size=(2, 2), strides=(2, 2)))
model.add(Conv2D(50, (5, 5), padding="same",name='Conv_2'))
model.add(Activation("relu"))
model.add(MaxPooling2D(pool_size=(2, 2), strides=(2, 2)))
# first (and only) set of FC => RELU layers
model.add(Flatten())
model.add(Dense(500,name='Dense_1'))
model.add(Activation("relu"))
# softmax classifier
model.add(Dense(classes,name='Dense_2'))
model.add(Activation("softmax"))
# return the constructed network architecture
return model
二、定义训练模型 train_network.py
# set the matplotlib backend so figures can be saved in the background
import matplotlib
matplotlib.use("Agg")
# import the necessary packages
from keras.preprocessing.image import ImageDataGenerator
from keras.optimizers import Adam
from sklearn.model_selection import train_test_split
from keras.preprocessing.image import img_to_array
from keras.utils import to_categorical
from pyimagesearch.lenet import LeNet
from imutils import paths
import matplotlib.pyplot as plt
import numpy as np
import argparse
import random
import cv2
import os
ap = argparse.ArgumentParser()
ap.add_argument("-d", "--dataset", type=str, default="training_set" ,
help="path to input dataset")
ap.add_argument("-m", "--model", type=str, default="trained_model" ,
help="path to output model")
ap.add_argument("-p", "--plot", type=str, default="plot.png" ,
help="path to output loss/accuracy plot")
args = vars(ap.parse_args())
EPOCHS = 25
INIT_LR = 1e-3
BS = 32
data = []
labels = []
dir_labels = ()
num_class = 0
# finding the labels
print("[INFO] Finding Labels...")
for file in os.listdir(args["dataset"]) :
temp_tuple=(file,'null')
dir_labels=dir_labels+temp_tuple
dir_labels=dir_labels[:-1]
#print(dir_labels)
num_class=num_class+1
#print(num_class)
# grab the image paths and randomly shuffle them
print("[INFO] Loading Images...")
imagePaths = sorted(list(paths.list_images(args["dataset"])))
random.seed(42)
random.shuffle(imagePaths)
for imagePath in imagePaths:
image = cv2.imread(imagePath)
image = cv2.resize(image, (28, 28))
image = img_to_array(image)
data.append(image)
label = imagePath.split(os.path.sep)[-2]
for i in range(num_class) :
if label == dir_labels[i] :
label = i
labels.append(label)
data = np.array(data, dtype="float") / 255.0
labels = np.array(labels)
(trainX, testX, trainY, testY) = train_test_split(data,labels, test_size=0.25, random_state=42)
trainY = to_categorical(trainY, num_classes=num_class)
testY = to_categorical(testY, num_classes=num_class)
aug = ImageDataGenerator(rotation_range=30, width_shift_range=0.1,
height_shift_range=0.1, shear_range=0.2, zoom_range=0.2,
horizontal_flip=True, fill_mode="nearest")
print("[INFO] Compiling Model...")
model = LeNet.build(width=28, height=28, depth=3, classes=num_class)
opt = Adam(lr=INIT_LR, decay=INIT_LR / EPOCHS)
model.compile(loss="binary_crossentropy", optimizer=opt,
metrics=["accuracy"])
print("[INFO] Training Network...")
H = model.fit_generator(aug.flow(trainX, trainY, batch_size=BS),
validation_data=(testX, testY), steps_per_epoch=len(trainX) // BS,
epochs=EPOCHS, verbose=1 )
print("[INFO] Saving Model...")
model_base=args["model"]+'.h5'
model.save(model_base)
# plot the training loss and accuracy
plt.style.use("ggplot")
plt.figure()
plt.plot(np.arange(0,EPOCHS), H.history["loss"], label="train_loss")
plt.plot(np.arange(0,EPOCHS), H.history["val_loss"], label="val_loss")
plt.plot(np.arange(0,EPOCHS), H.history["acc"], label="train_acc")
plt.plot(np.arange(0,EPOCHS), H.history["val_acc"], label="val_acc")
plt.title("Training Loss and Accuracy")
plt.xlabel("Epoch #")
plt.ylabel("Loss/Accuracy")
plt.legend(loc="lower left")
plt.savefig(args["plot"])
print("[INFO] Completed...")
三、定义预测模型
# import the necessary packages
from keras.preprocessing.image import img_to_array
from keras.models import load_model
import numpy as np
import argparse
import imutils
import cv2
import os
import sys
# initialising
dir_labels=()
dir_predict=()
num_class=0
# construct the argument parse and parse the arguments
ap = argparse.ArgumentParser()
ap.add_argument("-d", "--dataset", required=False, default="training_set",
help="path to input dataset")
ap.add_argument("-m", "--model", required=False, default="trained_model",
help="path to trained model model")
ap.add_argument("-i", "--image", required=True,
help="path to input image")
args = vars(ap.parse_args())
#findings the labels
for file in os.listdir(args["dataset"]) :
temp_tuple=(file,'null')
dir_labels=dir_labels+temp_tuple
dir_labels=dir_labels[:-1]
num_class=num_class+1
print("[INFO] Labels are ",dir_labels)
# load the image
print("[INFO] Loading Image...")
try :
image = cv2.imread(args["image"])
orig = image.copy()
except AttributeError :
print("[INFO] Error in the test image... ")
print('[INFO] Exiting...')
sys.exit()
# pre-process the image for classification
image = cv2.resize(image, (28, 28))
image = image.astype("float") / 255.0
image = img_to_array(image)
image = np.expand_dims(image, axis=0)
# load the trained convolutional neural network
print("[INFO] Loading Network...")
model_base=args["model"]+'.h5'
model = load_model(model_base)
# classify the input image
dir_predict = model.predict(image)[0]
print(dir_labels)
print(dir_predict)
for i in range(num_class) :
var = 0
for j in range(num_class) :
if(dir_predict[i]>=dir_predict[j]) :
var=var+1
if(var==num_class) :
label=dir_labels[i]
proba=dir_predict[i]
elif(var==num_class-1) :
label2=dir_labels[i]
proba2=dir_predict[i]
label = "{}: {:.2f}%".format(label, proba * 100)
# draw the label on the image
output = imutils.resize(orig, width=400)
cv2.putText(output, label, (10, 25), cv2.FONT_HERSHEY_SIMPLEX,
0.7, (0, 255, 0), 2)
# show and save the output image
cv2.imshow("Output", output)
cv2.imwrite("Output.png",output)
cv2.waitKey(0) #Press any key to exit the output image
print('[INFO] Exiting...')
数据集下载地址：
执行命令如下：
python train_network.py --dataset training_set --model trained_model --plot plo
输出：
Using TensorFlow backend.
[INFO] Finding Labels...
[INFO] Loading Images...
[INFO] Compiling Model...
[INFO] Training Network...
Epoch 1/25
2018-10-30 16:25:45.234853: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2
101/101 [==============================] - 18s 175ms/step - loss: 0.4371 - acc: 0.8043 - val_loss: 0.4090 - val_acc: 0.8098
Epoch 2/25
101/101 [==============================] - 13s 128ms/step - loss: 0.3919 - acc: 0.8224 - val_loss: 0.3658 - val_acc: 0.8415
Epoch 3/25
101/101 [==============================] - 12s 119ms/step - loss: 0.3714 - acc: 0.8333 - val_loss: 0.3616 - val_acc: 0.8515
Epoch 4/25
101/101 [==============================] - 12s 121ms/step - loss: 0.3661 - acc: 0.8352 - val_loss: 0.3442 - val_acc: 0.8547
Epoch 5/25
101/101 [==============================] - 14s 140ms/step - loss: 0.3393 - acc: 0.8486 - val_loss: 0.3250 - val_acc: 0.8636
Epoch 6/25
101/101 [==============================] - 15s 144ms/step - loss: 0.3274 - acc: 0.8575 - val_loss: 0.3387 - val_acc: 0.8528
Epoch 7/25
101/101 [==============================] - 14s 134ms/step - loss: 0.3270 - acc: 0.8564 - val_loss: 0.3232 - val_acc: 0.8614
Epoch 8/25
101/101 [==============================] - 15s 145ms/step - loss: 0.3192 - acc: 0.8626 - val_loss: 0.3176 - val_acc: 0.8627
Epoch 9/25
101/101 [==============================] - 14s 143ms/step - loss: 0.3033 - acc: 0.8695 - val_loss: 0.3333 - val_acc: 0.8562
Epoch 10/25
101/101 [==============================] - 13s 126ms/step - loss: 0.3141 - acc: 0.8641 - val_loss: 0.3155 - val_acc: 0.8647
Epoch 11/25
101/101 [==============================] - 17s 166ms/step - loss: 0.3032 - acc: 0.8672 - val_loss: 0.3088 - val_acc: 0.8667
Epoch 12/25
101/101 [==============================] - 17s 166ms/step - loss: 0.2906 - acc: 0.8755 - val_loss: 0.3450 - val_acc: 0.8562
Epoch 13/25
101/101 [==============================] - 16s 155ms/step - loss: 0.2884 - acc: 0.8759 - val_loss: 0.3135 - val_acc: 0.8688
Epoch 14/25
101/101 [==============================] - 16s 155ms/step - loss: 0.2943 - acc: 0.8733 - val_loss: 0.3143 - val_acc: 0.8652
Epoch 15/25
101/101 [==============================] - 15s 149ms/step - loss: 0.2731 - acc: 0.8868 - val_loss: 0.2883 - val_acc: 0.8791
Epoch 16/25
101/101 [==============================] - 14s 135ms/step - loss: 0.2776 - acc: 0.8806 - val_loss: 0.3088 - val_acc: 0.8723
Epoch 17/25
101/101 [==============================] - 13s 132ms/step - loss: 0.2612 - acc: 0.8864 - val_loss: 0.3020 - val_acc: 0.8808
Epoch 18/25
101/101 [==============================] - 13s 132ms/step - loss: 0.2639 - acc: 0.8887 - val_loss: 0.2781 - val_acc: 0.8830
Epoch 19/25
101/101 [==============================] - 14s 141ms/step - loss: 0.2600 - acc: 0.8878 - val_loss: 0.2882 - val_acc: 0.8797
Epoch 20/25
101/101 [==============================] - 15s 144ms/step - loss: 0.2476 - acc: 0.8938 - val_loss: 0.3219 - val_acc: 0.8690
Epoch 21/25
101/101 [==============================] - 14s 139ms/step - loss: 0.2518 - acc: 0.8938 - val_loss: 0.2979 - val_acc: 0.8806
Epoch 22/25
101/101 [==============================] - 13s 125ms/step - loss: 0.2565 - acc: 0.8894 - val_loss: 0.2977 - val_acc: 0.8782
Epoch 23/25
101/101 [==============================] - 15s 145ms/step - loss: 0.2423 - acc: 0.8976 - val_loss: 0.3188 - val_acc: 0.8684
Epoch 24/25
101/101 [==============================] - 15s 151ms/step - loss: 0.2317 - acc: 0.9030 - val_loss: 0.2979 - val_acc: 0.8825
Epoch 25/25
101/101 [==============================] - 14s 139ms/step - loss: 0.2312 - acc: 0.9033 - val_loss: 0.2858 - val_acc: 0.8869
[INFO] Saving Model...
[INFO] Completed...