使⽤Keras预训练模型ResNet50进⾏图像分类 版权声明：本⽂为博主原创⽂章，未经博主允许不得转载。 blog.csdn/u010632850/article/details/77926679
Keras提供了⼀些⽤ImageNet训练过的模型：Xception，VGG16，VGG19，ResNet50，InceptionV3。在使⽤这些模型的时候，有⼀个参数include_top表⽰是否包含模型顶部的全连接层，如果包含，则可以将图像分为ImageNet中的1000类，如果不包含，则可以利⽤这些参数来做⼀些定制的事情。在运⾏时⾃动下载有可能会失败，需要去⽹站中⼿动下载，放在“~/.keras/models/”中，使⽤WinPython 则在“settings/.keras/models/”中。
修正：表⽰当前是训练模式还是测试模式的参数K.learning_phase()⽂中表述和使⽤有误，在该函数说明中可以看到： The learning phase flag is a bool tensor (0 = test, 1 = train)，所以0是测试模式，1是训练模式，部分⽹络结构下两者有差别。
我使⽤的版本：
1. Ubuntu 16.04.3
2. Python 2.7
3. Keras 2.0.8
4. Tensoflow 1.3.0
5. Numpy 1.13.1
6. python-opencv 2.4.9.1+dfsg-1.5ubuntu1
7. h5py 2.7.0
函数：
def eachFile(filepath): #将⽬录内的⽂件名放⼊列表中
pathDir = os.listdir(filepath)
out = []
for allDir in pathDir:
child = allDir.decode('gbk') # .decode('gbk')是解决中⽂显⽰乱码问题
out.append(child)
return out
def get_data(data_name,train_left=0.0,train_right=0.7,train_all=0.7,resize=True,data_format=None,t=''): #从⽂件夹中获取图像数据
file_name = os.path.join(pic_dir_out,data_name+t+'_'+str(train_left)+'_'+str(train_right)+'_'+str(Width)+"X"+str(Height)+".h5")
print file_name
if ists(file_name): #判断之前是否有存到⽂件中
f = h5py.File(file_name,'r')
气囊止血带if t=='train':
X_train = f['X_train'][:]
y_train = f['y_train'][:]
f.close()
return (X_train, y_train)
elif t=='test':
X_test = f['X_test'][:]
y_test = f['y_test'][:]
f.close()
return (X_test, y_test)
else:
return
data_format = alize_data_format(data_format)
pic_dir_set = eachFile(pic_dir_data)
pic_dir_set = eachFile(pic_dir_data)
X_train = []
y_train = []
X_test = []
y_test = []
label = 0
for pic_dir in pic_dir_set:
print pic_dir_data+pic_dir
if not os.path.isdir(os.path.join(pic_dir_data,pic_dir)):
continue
pic_set = eachFile(os.path.join(pic_dir_data,pic_dir))
pic_index = 0
train_count = int(len(pic_set)*train_all)
train_l = int(len(pic_set)*train_left)
train_r = int(len(pic_set)*train_right)
for pic_name in pic_set:
if not os.path.isfile(os.path.join(pic_dir_data,pic_dir,pic_name)): continue
img = cv2.imread(os.path.join(pic_dir_data,pic_dir,pic_name)) if img is None:
continue
if (resize):
img = size(img,(Width,Height))
img = shape(-1,Width,Height,3)
if (pic_index < train_count):
if t=='train':
if (pic_index >= train_l and pic_index < train_r):
X_train.append(img)
y_train.append(label)
else:
if t=='test':
X_test.append(img)
y_test.append(label)
pic_index += 1
if len(pic_set) <> 0:
label += 1
f = h5py.File(file_name,'w')
if t=='train':
X_train = np.concatenate(X_train,axis=0)
y_train = np.array(y_train)
f.close()
return (X_train, y_train)
elif t=='test':
X_test = np.concatenate(X_test,axis=0)
y_test = np.array(y_test)
f.close()
return (X_test, y_test)
else:
return
调⽤：
global Width, Height, pic_dir_out, pic_dir_data
Width = 224
Height = 224
num_classes = 102 #Caltech101为102 cifar10为10
pic_dir_out = '/home/ccuux3/pic_cnn/pic_out/'
pic_dir_data = '/home/ccuux3/pic_cnn/pic_dataset/Caltech101/'
sub_dir = '224_resnet50/'
if not os.path.isdir(os.path.join(pic_dir_out,sub_dir)):
os.mkdir(os.path.join(pic_dir_out,sub_dir))
pic_dir_mine = os.path.join(pic_dir_out,sub_dir)
(X_train, y_train) = get_data("Caltech101_color_data_",0.0,0.7,data_format='channels_last',t='train')
y_train = _categorical(y_train, num_classes)
载⼊预训练模型ResNet50，并将训练图像经过⽹络运算得到数据，不包含顶部的全连接层，得到的结果存成⽂件，以后可以直接调⽤（由于我内存不够，所以拆分了⼀下）：
input_tensor = Input(shape=(224, 224, 3))
base_model = ResNet50(input_tensor=input_tensor,include_top=False,weights='imagenet')
#base_model = ResNet50(input_tensor=input_tensor,include_top=False,weights=None)
get_resnet50_output = K.function([base_model.layers[0].input, K.learning_phase()],
[base_model.layers[-1].output])
file_name = os.path.join(pic_dir_mine,'resnet50_train_output'+'.h5')
if ists(file_name):
f = h5py.File(file_name,'r')
resnet50_train_output = f['resnet50_train_output'][:]
f.close()
else:
you商城resnet50_train_output = []
delta = 10
for i in range(0,len(X_train),delta):
print i
one_resnet50_train_output = get_resnet50_output([X_train[i:i+delta], 0])[0]
resnet50_train_output.append(one_resnet50_train_output)
resnet50_train_output = np.concatenate(resnet50_train_output,axis=0)
f = h5py.File(file_name,'w')
f.close()
将ResNet50⽹络产⽣的结果⽤于图像分类：
input_tensor = Input(shape=(1, 1, 2048))
x = Flatten()(input_tensor)
x = Dense(1024, activation='relu')(x)
predictions = Dense(num_classes, activation='softmax')(x)
model = Model(inputs=input_tensor, outputs=predictions)
modelpile(optimizer=Adam(), loss='categorical_crossentropy',metrics=['accuracy'])现代主义绘画
训练图像数据集：
print('\nTraining ------------') #从⽂件中提取参数，训练后存在新的⽂件中
cm = 0 #修改这个参数可以多次训练
cm_str = '' if cm==0 else str(cm)
cm2_str = '' if (cm+1)==0 else str(cm+1)
if cm >= 1:
model.load_weights(os.path.join(pic_dir_mine,'cnn_model_Caltech101_resnet50_'+cm_str+'.h5'))
model.fit(resnet50_train_output, y_train, epochs=10, batch_size=128,)
model.save_weights(os.path.join(pic_dir_mine,'cnn_model_Caltech101_resnet50_'+cm2_str+'.h5'))
测试图像数据集：
(X_test, y_test) = get_data("Caltech101_color_data_",0.0,0.7,data_format='channels_last',t='test')
y_test = _categorical(y_test, num_classes)
file_name = os.path.join(pic_dir_mine,'resnet50_test_output'+'.h5')
if ists(file_name):
f = h5py.File(file_name,'r')
正交试验resnet50_test_output = f['resnet50_test_output'][:]
f.close()
else:
resnet50_test_output = []
delta = 10
for i in range(0,len(X_test),delta):
print i
one_resnet50_test_output = get_resnet50_output([X_test[i:i+delta], 0])[0]
resnet50_test_output.append(one_resnet50_test_output)
resnet50_test_output = np.concatenate(resnet50_test_output,axis=0)
f = h5py.File(file_name,'w')
f.close()
print('\nTesting ------------') #对测试集进⾏评估
class_name_list = get_name_list(pic_dir_data) #获取top-N的每类的准确率
pred = model.predict(resnet50_test_output, batch_size=32)
输出测试集各类别top-5的准确率：
N = 5
pred_list = []
for row in pred:
pred_list.append(row.argsort()[-N:][::-1]) #获取最⼤的N个值的下标
pred_array = np.array(pred_list)
test_arg = np.argmax(y_test,axis=1)
class_count = [0 for _ in xrange(num_classes)]
class_acc = [0 for _ in xrange(num_classes)]
for i in xrange(len(test_arg)):
class_count[test_arg[i]] += 1
if test_arg[i] in pred_array[i]:
class_acc[test_arg[i]] += 1
print('top-'+str(N)+' all acc:',str(sum(class_acc))+'/'+str(len(test_arg)),sum(class_acc)/float(len(test_arg))) for i in xrange(num_classes):
print (i, class_name_list[i], 'acc: '+str(class_acc[i])+'/'+str(class_count[i]))
完整代码：
# -*- coding: utf-8 -*-
import cv2
import numpy as np
import h5py
import os
from keras.utils import np_utils, conv_utils
dels import Model
from keras.layers import Flatten, Dense, Input
from keras.optimizers import Adam
from snet50 import ResNet50
from keras import backend as K
def get_name_list(filepath): #获取各个类别的名字
pathDir = os.listdir(filepath)
out = []
for allDir in pathDir:
for allDir in pathDir:
if os.path.isdir(os.path.join(filepath,allDir)):
child = allDir.decode('gbk') # .decode('gbk')是解决中⽂显⽰乱码问题
out.append(child)
return out
def eachFile(filepath): #将⽬录内的⽂件名放⼊列表中
pathDir = os.listdir(filepath)
out = []
for allDir in pathDir:
child = allDir.decode('gbk') # .decode('gbk')是解决中⽂显⽰乱码问题
out.append(child)
return out
def get_data(data_name,train_left=0.0,train_right=0.7,train_all=0.7,resize=True,data_format=None,t=''): #从⽂件夹中获取图像数据 file_name = os.path.join(pic_dir_out,data_name+t+'_'+str(train_left)+'_'+str(train_right)+'_'+str(Width)+"X"+str(Height)+".h5")
print file_name
if ists(file_name): #判断之前是否有存到⽂件中
f = h5py.File(file_name,'r')
if t=='train':
X_train = f['X_train'][:]
y_train = f['y_train'][:]
f.close()
return (X_train, y_train)
elif t=='test':
X_test = f['X_test'][:]
y_test = f['y_test'][:]
f.close()
return (X_test, y_test)
数字化
else:
return
data_format = alize_data_format(data_format)
pic_dir_set = eachFile(pic_dir_data)
X_train = []
y_train = []
X_test = []
y_test = []
label = 0
for pic_dir in pic_dir_set:
print pic_dir_data+pic_dir
if not os.path.isdir(os.path.join(pic_dir_data,pic_dir)):
continue
pic_set = eachFile(os.path.join(pic_dir_data,pic_dir))
pic_index = 0
train_count = int(len(pic_set)*train_all)
train_l = int(len(pic_set)*train_left)
train_r = int(len(pic_set)*train_right)
for pic_name in pic_set:
if not os.path.isfile(os.path.join(pic_dir_data,pic_dir,pic_name)):
continue
img = cv2.imread(os.path.join(pic_dir_data,pic_dir,pic_name))
if img is None:
continue
if (resize):
img = size(img,(Width,Height))
img = shape(-1,Width,Height,3)
if (pic_index < train_count):
if t=='train':
if (pic_index >= train_l and pic_index < train_r):
X_train.append(img)
y_train.append(label)
else:
if t=='test':
X_test.append(img)
y_test.append(label)
pic_index += 1
指路器
豫公网安备41160202000603
站长QQ:729038198
关于我们
投诉建议