上次战过RNN，这次来挑战一下CNN，对单个的手写汉字进行识别。
数据集
CASIA-HWDB
下载HWDB1.1数据集：
http://www.nlpr.ia.ac.cn/databases/download/feature_data/HWDB1.1trn_gnt.zip
http://www.nlpr.ia.ac.cn/databases/download/feature_data/HWDB1.1tst_gnt.zip
这个数据集由模式识别国家重点实验室共享
CNN架构参考论文： Deep Convolutional Network for Handwritten Chinese Character Recognition
数据处理
import os
import numpy as np
import struct
import PIL.Image
train_data_dir = "./data/HWDB1.1trn_gnt"
test_data_dir = "./data/HWDB1.1tst_gnt"
# 读取图像和对应的汉字
def read_from_gnt_dir(gnt_dir=train_data_dir):
def one_file(f):
header_size = 10
while True:
header = np.fromfile(f, dtype='uint8', count=header_size)
if not header.size: break
sample_size = header[0] + (header[1]<<8) + (header[2]<<16) + (header[3]<<24)
tagcode = header[5] + (header[4]<<8)
width = header[6] + (header[7]<<8)
height = header[8] + (header[9]<<8)
if header_size + width*height != sample_size:
break
image = np.fromfile(f, dtype='uint8', count=width*height).reshape((height, width))
yield image, tagcode
for file_name in os.listdir(gnt_dir):
if file_name.endswith('.gnt'):
file_path = os.path.join(gnt_dir, file_name)
with open(file_path, 'rb') as f:
for image, tagcode in one_file(f):
yield image, tagcode
# 统计样本数
train_counter = 0
test_counter = 0
for image, tagcode in read_from_gnt_dir(gnt_dir=train_data_dir):
tagcode_unicode = struct.pack('>H', tagcode).decode('gb2312')
# for image, tagcode in read_from_gnt_dir(gnt_dir=test_data_dir):
# tagcode_unicode = struct.pack('>H', tagcode).decode('gb2312')
# test_counter += 1
# 样本数
print(train_counter)
由于数据集文件格式是.gnt的，所以我们用PIL包对它进行解析，看看这些手写汉字长的啥样
# 提取点图像
if train_counter < 1000:
im = PIL.Image.fromarray(image)
im.convert('RGB').save('png/' + tagcode_unicode + str(train_counter) + '.png')
train_counter += 1
发现它们是这样的：
手写汉字
构造卷积神经网络
由于笔记本性能限制，跑完所有训练集所有汉字估计顶不住。所以我们取前140个进行识别。
代码：
import os
import numpy as np
import struct
import PIL.Image
train_data_dir = "./data/HWDB1.1trn_gnt"
test_data_dir = ".//data/HWDB1.1tst_gnt"
# 读取图像和对应的汉字
def read_from_gnt_dir(gnt_dir=train_data_dir):
def one_file(f):
header_size = 10
while True:
header = np.fromfile(f, dtype='uint8', count=header_size)
if not header.size: break
sample_size = header[0] + (header[1] << 8) + (header[2] << 16) + (header[3] << 24)
tagcode = header[5] + (header[4] << 8)
width = header[6] + (header[7] << 8)
height = header[8] + (header[9] << 8)
if header_size + width * height != sample_size:
break
image = np.fromfile(f, dtype='uint8', count=width * height).reshape((height, width))
yield image, tagcode
for file_name in os.listdir(gnt_dir):
if file_name.endswith('.gnt'):
file_path = os.path.join(gnt_dir, file_name)
with open(file_path, 'rb') as f:
for image, tagcode in one_file(f):
yield image, tagcode
import scipy.misc
from sklearn.utils import shuffle
import tensorflow as tf
# 前140个汉字进行测试
char_set = "的一是了我不人在他有这个上们来到时大地为子中你说生国年着就那和要她出也得里后自以会家可下而过天去能对小多然于心学么之都好看起发当没成只如事把还用第样道想作种开美总从无情己面最女但现前些所同日手又行意动方期它头经长儿回位分爱老因很给名法间斯知世什两次使身者被高已亲其进此话常与活正感"
def resize_and_normalize_image(img):
# 补方
pad_size = abs(img.shape[0] - img.shape[1]) // 2
if img.shape[0] < img.shape[1]:
pad_dims = ((pad_size, pad_size), (0, 0))
else:
pad_dims = ((0, 0), (pad_size, pad_size))
img = np.lib.pad(img, pad_dims, mode='constant', constant_values=255)
# 缩放
img = scipy.misc.imresize(img, (64 - 4 * 2, 64 - 4 * 2))
img = np.lib.pad(img, ((4, 4), (4, 4)), mode='constant', constant_values=255)
assert img.shape == (64, 64)
img = img.flatten()
# 像素值范围-1到1
img = (img - 128) / 128
return img
# one hot
def convert_to_one_hot(char):
vector = np.zeros(len(char_set))
vector[char_set.index(char)] = 1
return vector
# 数据量不大, 可一次全部加载到RAM
train_data_x = []
train_data_y = []
for image, tagcode in read_from_gnt_dir(gnt_dir=train_data_dir):
tagcode_unicode = struct.pack('>H', tagcode).decode('gb2312')
if tagcode_unicode in char_set:
train_data_x.append(resize_and_normalize_image(image))
train_data_y.append(convert_to_one_hot(tagcode_unicode))
# shuffle样本
train_data_x, train_data_y = shuffle(train_data_x, train_data_y, random_state=0)
batch_size = 128
num_batch = len(train_data_x) // batch_size
text_data_x = []
text_data_y = []
for image, tagcode in read_from_gnt_dir(gnt_dir=test_data_dir):
tagcode_unicode = struct.pack('>H', tagcode).decode('gb2312')
if tagcode_unicode in char_set:
text_data_x.append(resize_and_normalize_image(image))
text_data_y.append(convert_to_one_hot(tagcode_unicode))
# shuffle样本
text_data_x, text_data_y = shuffle(text_data_x, text_data_y, random_state=0)
X = tf.placeholder(tf.float32, [None, 64 * 64])
Y = tf.placeholder(tf.float32, [None, 140])
keep_prob = tf.placeholder(tf.float32)
def chinese_hand_write_cnn():
x = tf.reshape(X, shape=[-1, 64, 64, 1])
# 2 conv layers
w_c1 = tf.Variable(tf.random_normal([3, 3, 1, 32], stddev=0.01))
b_c1 = tf.Variable(tf.zeros([32]))
conv1 = tf.nn.relu(tf.nn.bias_add(tf.nn.conv2d(x, w_c1, strides=[1, 1, 1, 1], padding='SAME'), b_c1))
conv1 = tf.nn.max_pool(conv1, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')
w_c2 = tf.Variable(tf.random_normal([3, 3, 32, 64], stddev=0.01))
b_c2 = tf.Variable(tf.zeros([64]))
conv2 = tf.nn.relu(tf.nn.bias_add(tf.nn.conv2d(conv1, w_c2, strides=[1, 1, 1, 1], padding='SAME'), b_c2))
conv2 = tf.nn.max_pool(conv2, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')
"""
# 可以增加一层网络
w_c3 = tf.Variable(tf.random_normal([3, 3, 64, 128], stddev=0.01))
b_c3 = tf.Variable(tf.zeros([128]))
conv3 = tf.nn.relu(tf.nn.bias_add(tf.nn.conv2d(conv2, w_c3, strides=[1, 1, 1, 1], padding='SAME'), b_c3))
conv3 = tf.nn.max_pool(conv3, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')
conv3 = tf.nn.dropout(conv3, keep_prob)
"""
# 全连接层，8*32*64
w_d = tf.Variable(tf.random_normal([8 * 32 * 64, 1024], stddev=0.01))
b_d = tf.Variable(tf.zeros([1024]))
dense = tf.reshape(conv2, [-1, w_d.get_shape().as_list()[0]])
dense = tf.nn.relu(tf.add(tf.matmul(dense, w_d), b_d))
dense = tf.nn.dropout(dense, keep_prob)
w_out = tf.Variable(tf.random_normal([1024, 140], stddev=0.01))
b_out = tf.Variable(tf.zeros([140]))
out = tf.add(tf.matmul(dense, w_out), b_out)
return out
def train_hand_write_cnn():
output = chinese_hand_write_cnn()
loss = tf.reduce_sum(tf.nn.softmax_cross_entropy_with_logits(logits= output,labels= Y))
optimizer = tf.train.AdamOptimizer(learning_rate=0.001).minimize(loss)
accuracy = tf.reduce_mean(tf.cast(tf.equal(tf.argmax(output, 1), tf.argmax(Y, 1)), tf.float32))
# TensorBoard
tf.summary.scalar("loss", loss)
tf.summary.scalar("accuracy", accuracy)
merged_summary_op = tf.summary.merge_all()
saver = tf.train.Saver()
with tf.Session() as sess:
sess.run(tf.global_variables_initializer())
# 命令行执行 tensorboard --logdir=./log 打开浏览器访问http://0.0.0.0:6006
summary_writer = tf.summary.FileWriter('./log', graph=tf.get_default_graph())
for e in range(50):
for i in range(num_batch):
batch_x = train_data_x[i * batch_size: (i + 1) * batch_size]
batch_y = train_data_y[i * batch_size: (i + 1) * batch_size]
_, loss_, summary = sess.run([optimizer, loss, merged_summary_op],
feed_dict={X: batch_x, Y: batch_y, keep_prob: 0.5})
# 每次迭代都保存日志
summary_writer.add_summary(summary, e * num_batch + i)
print(e * num_batch + i, loss_)
if e * num_batch + i % 100 == 0:
# 计算准确率
acc = accuracy.eval({X: text_data_x[:500], Y: text_data_y[:500], keep_prob: 1.})
# acc = sess.run(accuracy, feed_dict={X: text_data_x[:500], Y: text_data_y[:500], keep_prob: 1.})
print(e * num_batch + i, acc)
train_hand_write_cnn()
跑起来之后，电脑风扇都快赶上直升飞机了，明天再去用实验室的台式跑起来看看，代码是没有问题的。
两千多轮的时候居然跑出了超过100%的准确率，应该是过拟合了
