Tensorflow化骨绵掌第4式-LeNet、AlexNet、VGG16、VGG19打造自己的图像识别模型（1）
1、下载数据集
本期学习需要使用的数据集是17 Category Flower Dataset。17flowers是牛津大学Visual Geometry Group选取的在英国比较常见的17种花。其中每种花有80张图片，整个数据及有1360张图片，可以在官网下载。
下载路径：http://www.robots.ox.ac.uk/~vgg/data/flowers/17/
或者在tensorflow的官网下载：https://github.com/ck196/tensorflow-alexnet/blob/master/17flowers.tar.gz
2、了解lenet模型
LeNet-5：是Yann LeCun在1998年设计的用于手写数字识别的卷积神经网络，当年美国大多数银行就是用它来识别支票上面的手写数字的，它是早期卷积神经网络中最有代表性的实验系统之一。
LenNet-5共有7层（不包括输入层），每层都包含不同数量的训练参数。LeNet-5中主要有2个卷积层、2个下抽样层（池化层）、3个全连接层，3种连接方式如下图所示。
3、使用Lenet-5构建自己的图像识别模型
(1)首先得先把文件夹的图片数据读入模型，数据集图片大小是2828，现在需要把它变成100100.具体代码如下所示：
#coding=utf-8
from skimage import io,transform
import glob
import os
import tensorflow as tf
import numpy as np
import time
#数据集地址
path='D:/biancheng/'
#模型保存地址
model_path='D:/biancheng/LeNet_model/model.ckpt'
#将所有的图片resize成100*100
w=28
h=28
c=3
#读取图片
def read_img(path):
#
cate=[path+x for x in os.listdir(path) if os.path.isdir(path+x)]
imgs=[]
labels=[]
for idx,folder in enumerate(cate):
for im in glob.glob(folder+'/*.jpg'):
print('reading the images:%s'%(im))
img=io.imread(im)
img=transform.resize(img,(w,h,c))
imgs.append(img)
labels.append(idx)
return np.asarray(imgs,np.float32),np.asarray(labels,np.int32)
data,label=read_img(path)
#参数概要
def variable_summaries(var):
with tf.name_scope("summaries"):
mean = tf.reduce_mean(var)
tf.summary.scalar("mean",mean)#平均值
with tf.name_scope("stddev"):
stddev = tf.sqrt(tf.reduce_mean(tf.square(var-mean)))
tf.summary.scalar("stddev",stddev)#标准差
tf.summary.scalar("max",tf.reduce_max(var))#最大值
tf.summary.scalar("min",tf.reduce_min(var))#最小值
tf.summary.histogram("histogram",var)#直方图
#打乱顺序
num_example=data.shape[0]
arr=np.arange(num_example)
np.random.shuffle(arr)
data=data[arr]
label=label[arr]
#将所有数据分为训练集和验证集
ratio=0.7
s=np.int(num_example*ratio)
x_train=data[:s]
y_train=label[:s]
x_val=data[s:]
y_val=label[s:]
#-----------------构建网络----------------------
#占位符
x=tf.placeholder(tf.float32,shape=[None,w,h,c],name='x')
y_=tf.placeholder(tf.int32,shape=[None,],name='y_')
lr = tf.Variable(0.0001,dtype=tf.float32)
# 构建CNN网络
def inference(input_tensor, train, regularizer):
with tf.variable_scope('layer1-conv1'):
conv1_weights = tf.get_variable("weight",[5,5,3,32],
initializer=tf.truncated_normal_initializer(stddev=0.1))
conv1_biases = tf.get_variable("bias", [32],
initializer=tf.constant_initializer(0.0))
conv1 = tf.nn.conv2d(input_tensor, conv1_weights,
strides=[1, 1, 1, 1], padding='SAME')
relu1 = tf.nn.relu(tf.nn.bias_add(conv1, conv1_biases))
with tf.name_scope("layer1-pool1"):
pool1 = tf.nn.max_pool(relu1, ksize = [1,2,2,1],
strides=[1,2,2,1],padding="VALID")
with tf.variable_scope("layer2-conv2"):
conv2_weights = tf.get_variable("weight",[5,5,32,64],
initializer=tf.truncated_normal_initializer(stddev=0.1))
conv2_biases = tf.get_variable("bias", [64],
initializer=tf.constant_initializer(0.0))
conv2 = tf.nn.conv2d(pool1, conv2_weights, strides=[1, 1, 1, 1], padding='SAME')
relu2 = tf.nn.relu(tf.nn.bias_add(conv2, conv2_biases))
with tf.name_scope("layer2-pool2"):
pool2 = tf.nn.max_pool(relu2, ksize=[1, 2, 2, 1],
strides=[1, 2, 2, 1], padding='VALID')
nodes = 7*7*64
reshaped = tf.reshape(pool2,[-1,nodes])
with tf.variable_scope('layer3-fc1'):
fc1_weights = tf.get_variable("weight", [nodes, 512],
initializer=tf.truncated_normal_initializer(stddev=0.1))
if regularizer != None:
tf.add_to_collection('losses', regularizer(fc1_weights))
fc1_biases = tf.get_variable("bias", [512],
initializer=tf.constant_initializer(0.1))
fc1 = tf.nn.relu(tf.matmul(reshaped, fc1_weights) + fc1_biases)
if train:
fc1 = tf.nn.dropout(fc1, 0.5)
with tf.variable_scope('layer4-fc2'):
fc2_weights = tf.get_variable("weight", [512, 64],
initializer=tf.truncated_normal_initializer(stddev=0.1))
if regularizer != None:
tf.add_to_collection('losses', regularizer(fc2_weights))
fc2_biases = tf.get_variable("bias", [64],
initializer=tf.constant_initializer(0.1))
fc2 = tf.nn.relu(tf.matmul(fc1, fc2_weights) + fc2_biases)
if train:
fc2 = tf.nn.dropout(fc2, 0.5)
with tf.variable_scope('layer5-fc3'):
fc3_weights = tf.get_variable("weight", [64, 5],
initializer=tf.truncated_normal_initializer(stddev=0.1))
if regularizer != None:
tf.add_to_collection('losses', regularizer(fc3_weights))
fc3_biases = tf.get_variable("bias", [5], initializer=tf.constant_initializer(0.1))
logit = tf.matmul(fc2, fc3_weights) + fc3_biases
return logit
#---------------------------网络结束---------------------------
regularizer = tf.contrib.layers.l2_regularizer(0.0001)
logits = inference(x,False,regularizer)
#(小处理)将logits乘以1赋值给logits_eval，定义name，方便在后续调用模型时通过tensor名字调用输出tensor
b = tf.constant(value=1,dtype=tf.float32)
logits_eval = tf.multiply(logits,b,name='logits_eval')
with tf.name_scope("loss"):
loss=tf.nn.sparse_softmax_cross_entropy_with_logits(logits=logits, labels=y_)
tf.summary.scalar("loss", loss)
with tf.name_scope("train"):
train_op=tf.train.AdamOptimizer(learning_rate=0.0001).minimize(loss)
correct_prediction = tf.equal(tf.cast(tf.argmax(logits,1),tf.int32), y_)
acc= tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
#定义一个函数，按批次取数据
def minibatches(inputs=None, targets=None, batch_size=None, shuffle=False):
assert len(inputs) == len(targets)
if shuffle:
indices = np.arange(len(inputs))
np.random.shuffle(indices)
for start_idx in range(0, len(inputs) - batch_size + 1, batch_size):
if shuffle:
excerpt = indices[start_idx:start_idx + batch_size]
else:
excerpt = slice(start_idx, start_idx + batch_size)
yield inputs[excerpt], targets[excerpt]
print("--------------------------数据集训练开始---------------------------------")
#训练和测试数据，可将n_epoch设置更大一些
n_epoch=3000
batch_size=16
saver=tf.train.Saver()
with tf.Session() as sess:
sess.run(tf.global_variables_initializer())
for epoch in range(n_epoch):
start_time = time.time()
sess.run(tf.assign(lr, 0.001 * (0.95 ** epoch)))
#training
train_loss, train_acc, n_batch = 0, 0, 0
for x_train_a, y_train_a in minibatches(x_train, y_train, batch_size, shuffle=True):
_,err,ac=sess.run([train_op,loss,acc], feed_dict={x: x_train_a, y_: y_train_a})
train_loss += err;
train_acc += ac;
n_batch += 1
print(" train loss: %f" % (np.sum(train_loss)/ n_batch)," train acc: %f" % (np.sum(train_acc)/ n_batch))
#validation
val_loss, val_acc, n_batch = 0, 0, 0
for x_val_a, y_val_a in minibatches(x_val, y_val, batch_size, shuffle=False):
err, ac = sess.run([loss,acc], feed_dict={x: x_val_a, y_: y_val_a})
val_loss += err; val_acc += ac; n_batch += 1
count = 0
print(" validation loss: %f" % (np.sum(val_loss)/ n_batch)," validation acc: %f" % (np.sum(val_acc)/ n_batch))
saver.save(sess,model_path)
print("--------------------------数据集训练结束---------------------------------")
4、使用构建的lenet网络模型进行花卉识别
前面对lenet网络进行训练，以及保存了网络模型，现在我们要进行调用模型进行图像的测试识别，因此写了一个test.py文件进行测试。
test.py代码如下：
#coding=utf-8
from skimage import io, transform
import tensorflow as tf
import numpy as np
import os # os 处理文件和目录的模块
import glob # glob 文件通配符模块
#待测试数据集的路径
path = 'D:/biancheng/'
# 类别代表字典
flower_dict = {0:'花卉1',1:'花卉2',2:'花卉3',3:'花卉4',4:'花卉5'}
w = 28
h = 28
c = 3
# 读取图片+数据处理
def read_img(path):
# os.listdir(path) 返回path指定的文件夹包含的文件或文件夹的名字的列表
# os.path.isdir(path)判断path是否是目录
# b = [x+x for x in list1 if x+x<15 ] 列表生成式,循环list1，当if为真时，将x+x加入列表b
cate = [path + x for x in os.listdir(path) if os.path.isdir(path + x)]
imgs = []
for idx, folder in enumerate(cate):
# glob.glob(s+'*.py') 从目录通配符搜索中生成文件列表
for im in glob.glob(folder + '/*.jpg'):
# 输出读取的图片的名称
print('reading the images:%s' % (im))
# io.imread(im)读取单张RGB图片 skimage.io.imread(fname,as_grey=True)读取单张灰度图片
# 读取的图片
img = io.imread(im)
# skimage.transform.resize(image, output_shape)改变图片的尺寸
img = transform.resize(img, (w, h, c))
# 将读取的图片数据加载到imgs[]列表中
imgs.append(img)
# 将图片的label加载到labels[]中，与上方的imgs索引对应
# labels.append(idx)
# 将读取的图片和labels信息，转化为numpy结构的ndarr(N维数组对象（矩阵）)数据信息
return np.asarray(imgs, np.float32)
# 调用读取图片的函数，得到图片和labels的数据集
data = read_img(path)
with tf.Session() as sess:
saver = tf.train.import_meta_graph('D:/biancheng/LeNet_model/model.ckpt.meta')
saver.restore(sess,tf.train.latest_checkpoint('D:/biancheng/LeNet_model/'))
# sess：表示当前会话，之前保存的结果将被加载入这个会话
# 设置每次预测的个数
graph = tf.get_default_graph()
x = graph.get_tensor_by_name("x:0")
feed_dict = {x: data}
logits = graph.get_tensor_by_name("logits_eval:0") # eval功能等同于sess(run)
classification_result = sess.run(logits, feed_dict)
# 打印出预测矩阵
print(classification_result)
# 打印出预测矩阵每一行最大值的索引
print(tf.argmax(classification_result, 1).eval())
# 根据索引通过字典对应花卉的分类
output = []
output = tf.argmax(classification_result, 1).eval()
for i in range(len(output)):
print("第", i + 1, "朵花卉预测:" + flower_dict[output[i]])
本期作业：
1、了解dropout、正则化、批量归一化等
2、继续修改Lenet网络，提高准确率
3、改写VGG16网络，达到可以初步识别花卉数据集
4、继续tensorflow视频课程（P24-30）：https://www.bilibili.com/video/av20542427/?p=8&t=1221
本期学习就到这里，欢迎大家关注、批评指正。
7238
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
