TensorFlow框架下cnn卷积神经⽹络识别花卉
识别过程
1、第⼀步在⽹络上搜索与CNN卷积神经⽹络识别花卉的相关信息，这⾥你必定能到相关的花卉数据集数据集，⽹上下的花卉数据集⼤概有3670张图⽚，分为菊花，郁⾦⾹，玫瑰，蒲公英，和向⽇葵
2、⼤致过程都⼀样，跟⽹上各位⼤⽜的差不多（本⼈⼀名⼤⼆学⽣，刚接触不久，但也还算了解吧），导⼊数据，就是你下的哪些数据集
3、设置相关参数，n_epoch，banch_size什么的。训练模型后保存模型，然后拉出来训练了。
本⽂仅供参考，⽹上太多这个了，但是他们的测试结果我觉得并不真实，我⾃⼰亲⾝做过，参考过⼏篇不错的博主⽂章，⽂章后供给⼤家，以供参考。
**以下这段属于训练并保存模型的代码**
import os
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= 'G:/⼈⼯智能/第⼆次花卉识别/花卉数据集/'
#模型保存地址
model_path='G:/⼈⼯智能/第⼆次花卉识别/保存模型'
#将所有的图⽚resize成100*100
w=100
h=100
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)
size(img,(w,h))
imgs.append(img)
labels.append(idx)
return np.asarray(imgs,np.float32),np.asarray(labels,np.int32)
data,label=read_img(path)
#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=0.8
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----------------------
#-----------------构建⽹络----------------------
#占位符
x=tf.placeholder(tf.float32,shape=[None,w,h,c],name='x')
#可以避免⽣成⼤量常量来提供输⼊数据，提⾼了计算图的利⽤率（占位符的好处）
y_=tf.placeholder(tf.int32,shape=[None,],name='y_')
def inference(input_tensor, train, regularizer):
with tf.variable_scope('layer1-conv1'):
conv1_weights = tf.get_variable("weight",[5,5,3,32],uncated_normal_initializer(stddev=0.1))
conv1_biases = tf.get_variable("bias", [32], stant_initializer(0.0))
conv1 = v2d(input_tensor, conv1_weights, strides=[1, 1, 1, 1], padding='SAME')
relu1 = bias_add(conv1, conv1_biases))
with tf.name_scope("layer2-pool1"):
pool1 = tf.nn.max_pool(relu1, ksize = [1,2,2,1],strides=[1,2,2,1],padding="VALID")
with tf.variable_scope("layer3-conv2"):
conv2_weights = tf.get_variable("weight",[5,5,32,64],uncated_normal_initializer(stddev=0.1)) conv2_biases = tf.get_variable("bias", [64], stant_initializer(0.0))
conv2 = v2d(pool1, conv2_weights, strides=[1, 1, 1, 1], padding='SAME')
relu2 = bias_add(conv2, conv2_biases))
with tf.name_scope("layer4-pool2"):
pool2 = tf.nn.max_pool(relu2, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='VALID')
with tf.variable_scope("layer5-conv3"):
conv3_weights = tf.get_variable("weight",[3,3,64,128],uncated_normal_initializer(stddev=0.1)) conv3_biases = tf.get_variable("bias", [128], stant_initializer(0.0))
conv3 = v2d(pool2, conv3_weights, strides=[1, 1, 1, 1], padding='SAME')
relu3 = bias_add(conv3, conv3_biases))
with tf.name_scope("layer6-pool3"):
pool3 = tf.nn.max_pool(relu3, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='VALID')
with tf.variable_scope("layer7-conv4"):
conv4_weights = tf.get_variable("weight",[3,3,128,128],uncated_normal_initializer(stddev=0.1)) conv4_biases = tf.get_variable("bias", [128], stant_initializer(0.0))
conv4 = v2d(pool3, conv4_weights, strides=[1, 1, 1, 1], padding='SAME')
relu4 = bias_add(conv4, conv4_biases))
with tf.name_scope("layer8-pool4"):
pool4 = tf.nn.max_pool(relu4, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='VALID')
nodes = 6*6*128
reshaped = tf.reshape(pool4,[-1,nodes])
with tf.variable_scope('layer9-fc1'):
fc1_weights = tf.get_variable("weight", [nodes, 1024],
uncated_normal_initializer(stddev=0.1))
if regularizer != None: tf.add_to_collection('losses', regularizer(fc1_weights))
fc1_biases = tf.get_variable("bias", [1024], stant_initializer(0.1))
fc1 = lu(tf.matmul(reshaped, fc1_weights) + fc1_biases)
if train: fc1 = tf.nn.dropout(fc1, 0.5)
with tf.variable_scope('layer10-fc2'):
fc2_weights = tf.get_variable("weight", [1024, 512],
uncated_normal_initializer(stddev=0.1))
if regularizer != None: tf.add_to_collection('losses', regularizer(fc2_weights))
fc2_biases = tf.get_variable("bias", [512], stant_initializer(0.1))
fc2 = lu(tf.matmul(fc1, fc2_weights) + fc2_biases)
if train: fc2 = tf.nn.dropout(fc2, 0.5)
with tf.variable_scope('layer11-fc3'):
fc3_weights = tf.get_variable("weight", [512, 5],
uncated_normal_initializer(stddev=0.1))
uncated_normal_initializer(stddev=0.1))
if regularizer != None: tf.add_to_collection('losses', regularizer(fc3_weights))
fc3_biases = tf.get_variable("bias", [5], stant_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')
sparse_softmax_cross_entropy_with_logits(logits=logits, labels=y_)
train_ain.AdamOptimizer(learning_rate=0.001).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]
#训练和测试数据，可将n_epoch设置更⼤⼀些
n_epoch=10
batch_size=10
ain.Saver()
sess=tf.Session()
sess.run(tf.global_variables_initializer())
for epoch in range(n_epoch):
start_time = time.time()
#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))
print(" 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
print(" validation loss: %f" % (np.sum(val_loss)/ n_batch))
print(" validation acc: %f" % (np.sum(val_acc)/ n_batch))
saver.save(sess,model_path)
sess.close()
**调⽤训练好的模型**
import os
from skimage import io,transform
import tensorflow as tf
import numpy as np
path1 = "G:/⼈⼯智能/第⼆次花卉识别/测试样本1/daisy/2617111535_54c2ac8462.jpg"
path2 = "G:/⼈⼯智能/第⼆次花卉识别/测试样本1/dandelion/479495978_ee22cf05be.jpg" path3 = "G:/⼈⼯智能/第⼆次花卉识别/测试样本1/roses/2065522422_cfdd80044a_n.jpg" path4 = "G:/⼈⼯智能/第⼆次花卉识别/测试样本1/sunflowers/1379256773_bb2eb0d95b_n.jpg" path5 = "G:/⼈⼯智能/第
⼆次花卉识别/测试样本1/tulips/444963906_e41492b692.jpg"
flower_dict = {0:'dasiy',1:'dandelion',2:'roses',3:'sunflowers',4:'tulips'}
w=100
h=100
c=3
def read_one_image(path):
img = io.imread(path)
img = size(img,(w,h))
return np.asarray(img)
with tf.Session() as sess:
data = []
data1 = read_one_image(path1)
data2 = read_one_image(path2)
data3 = read_one_image(path3)
data4 = read_one_image(path4)
data5 = read_one_image(path5)
data.append(data1)
data.append(data2)
data.append(data3)
data.append(data4)
data.append(data5)
saver = tf.train.import_meta_graph('G:/⼈⼯智能/第⼆次花卉识别/保存模型.meta')
graph = tf.get_default_graph()
x = _tensor_by_name("x:0")
feed_dict = {x:data}
logits = _tensor_by_name("logits_eval:0")
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的分类
output = []
output = tf.argmax(classification_result,1).eval()
for i in range(len(output)):
print("第",i+1,"朵花预测:"+flower_dict[output[i]])
训练模型的测试结果如下
开始训练时的截图
测试结果图
validation框架
版权声明：本站内容均来自互联网，仅供演示用，请勿用于商业和其他非法用途。如果侵犯了您的权益请与我们联系QQ:729038198，我们将在24小时内删除。
发表评论