In [1]:
from captcha.image import ImageCaptcha
import numpy as np
import matplotlib.pyplot as plt
from PIL import Image
import random
number = ['0', '1', '2', '3', '4', '5', '6', '7', '8', '9']
alphabet = ['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z']
ALPHABET = ['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z']
char_set = "".join(number + alphabet + ALPHABET)
char_set = "".join(number)
# 图像大小
height = 60
width = 160
captcha_size = 4
n_class = len(char_set)
In [2]:
# 随机生成长度为4的字符串
def random_captcha_text(char_set = char_set, captcha_size = captcha_size):
captcha_text = []
for _ in range(captcha_size):
c = random.choice(char_set)
captcha_text.append(c)
captcha_text = ''.join(captcha_text)
return captcha_text
In [3]:
# 生成图片和label
def gen_captcha_text_and_image():
image = ImageCaptcha()
captcha_text = random_captcha_text()
captcha = image.generate(captcha_text)
captcha_image = Image.open(captcha)
captcha_image = np.array(captcha_image)
return captcha_text, captcha_image
In [4]:
text, image = gen_captcha_text_and_image()
plt.title(text)
plt.imshow(image)
plt.show()
In [5]:
# 生成一个训练batch
def get_next_batch(batch_size=128):
# 创建2个空数组， 用来存放一个批次的数据
while True:
batch_x = np.zeros((batch_size, height, width, 3))
batch_y = [np.zeros((batch_size, n_class)) for i in range(captcha_size)]
# 有时生成图像大小不是(60, 160, 3)
def wrap_gen_captcha_text_and_image():
while True:
text, image = gen_captcha_text_and_image()
if image.shape == (60, 160, 3):
return text, image
for i in range(batch_size):
text, image = gen_captcha_text_and_image()
batch_x[i,:] = image
# one-hot编码label
for j, ch in enumerate(text):
batch_y[j][i, :] = 0
batch_y[j][i, char_set.find(ch)] = 1
yield batch_x, batch_y
In [6]:
from keras.models import *
from keras.layers import *
import numpy as np
import keras
def VGG():
input_tensor = Input((height, width, 3))
x = input_tensor
x = Conv2D(32, (3, 3), activation='relu', padding='SAME')(x)
x = Conv2D(32, (3, 3), activation='relu', padding='SAME')(x)
x = MaxPooling2D((2, 2), padding='SAME')(x)
x = Conv2D(64, (3, 3), activation='relu', padding='SAME')(x)
x = Conv2D(64, (3, 3), activation='relu', padding='SAME')(x)
x = MaxPooling2D((2, 2), padding='SAME')(x)
x = Conv2D(128, (3, 3), activation='relu', padding='SAME')(x)
x = Conv2D(128, (3, 3), activation='relu', padding='SAME')(x)
x = MaxPooling2D((2, 2), padding='SAME')(x)
x = Conv2D(256, (3, 3), activation='relu', padding='SAME')(x)
x = Conv2D(256, (3, 3), activation='relu', padding='SAME')(x)
x = MaxPooling2D((2, 2), padding='SAME')(x)
x = Flatten()(x)
x = Dropout(0.3)(x)
x = [Dense(n_class, activation='softmax', name='c%d'%(i+1))(x) for i in range(4)]
model = Model(inputs=input_tensor, outputs=x)
return model
model = VGG()
model.compile(loss='categorical_crossentropy', optimizer='adadelta', metrics=['accuracy'])
In [7]:
from IPython.display import SVG
from keras.utils.vis_utils import model_to_dot
SVG(model_to_dot(model).create(prog='dot', format='svg'))
Out[7]:
In [9]:
model.fit_generator(generator = get_next_batch(),
steps_per_epoch = 100,
epochs=10,
verbose=1,
validation_data = get_next_batch(),
validation_steps = 10)
Out[9]:
In [ ]: