微博验证码识别系列三：根据微博验证码图片建立模型

我们在上期已经制作好了通用模型，这期就要制作微博验证码的识别模型了
首先我们去微博获取验证码图片并对这些图片打标签（注意：这里打标签的图片越多生成的模型越好，识别率越高）
废话不多说，我们看代码：

import numpy as np
from scipy import misc
from keras.applications.xception import Xception, preprocess_input
import glob
from keras.layers import Input, Dense, Dropout
from keras.models import Model
images_split = 5
img_size = (60, 160)
input_image = Input(shape=(img_size[0], img_size[1], 3))
base_model = Xception(input_tensor=input_image, weights=None, include_top=False, pooling='avg')
predicts = [Dense(36, activation='softmax')(Dropout(0.5)(base_model.output)) for i in range(images_split)]

model = Model(inputs=input_image, outputs=predicts)
model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])
model.load_weights('CaptchaForPython.h5')


def data_generator(data, batch_size):  # 样本生成器，节省内存
    while True:
        batch = np.random.choice(data, batch_size)
        x, y = [], []
        for img in batch:
            im = misc.imread(img)
            im = im[:, :, :3]
            im = misc.imresize(im, img_size)
            x.append(im)
            real_num = img.split('.')[0].split('/')[-1]
            y_list = []
            for i in real_num:
                i = i.upper()
                if ord(i) - ord('A') >= 0:
                    y_list.append(ord(i) - ord('A') + 10)
                else:
                    y_list.append(ord(i) - ord('0'))
            y.append(y_list)
        x = preprocess_input(np.array(x).astype(float))
        y = np.array(y)
        yield x, [y[:, i] for i in range(images_split)]


# 获取指定目录下的所有图片
samples = glob.glob('weiboimgs/*.jpg')
np.random.shuffle(samples)

nb_train = 450
train_samples = samples[:nb_train]
test_samples = samples[nb_train:]

# Continue training
model.fit_generator(data_generator(train_samples, 30), steps_per_epoch=150, epochs=7,
                    validation_data=data_generator(test_samples, 10), validation_steps=5)
model.save('weibo.h5')

这段代码执行完成后就能得到真正的微博验证码识别模型了，大家快动手试试吧